1
i
:^.
•J
i
X-
I
MAGNA CARTA
PUBLISHED BY
JAMES MACLEHOSE AND SONS, GLASGOW,
^nblishrts to tht Hntbtrsttg.
MACMILLAN AND CO., LTD., LONDON.
New York,
Toronto, •
London, •
Cambridge,
Eainburgh,
Sydney, •
The Macmillan Co.
The Macmillan Co. of Canada.
Simpkin, Hamilion and Co.
Bowes and Bowes.
Douglas and Foulis.
Angus and Robertson.
MCMXIV.
"vo
MAGNA CARTA
A COMMENTARY ON THE GREAT
CHARTER OF KING JOHN
WITH AN
HISTORICAL INTRODUCTION
BY
WILLIAM SHARP McKECHNIE
M.A., LL.B., D.Phil.
LECTURER ON CONSTITUTIONAL LAW AND HISTORY IN THE UNIVERSITY OF GLASGOW
AUTHOR OF
'the STATE AND THE INDIVIDUAL,' 'tHE REFORM OF THE HOUSE OF LORDS*
'the NEW DEMOCRACY AND THE CONSTITUTION,' ETC.
SECOND EDITION
REVISED AND IN PART RE-WRITTEN
GLASGOW
JAMES MACLEHOSE AND SONS
PUBLISHERS TO THE UNIVERSITY
I9I4
147
PREFACE TO SECOND EDITION.
The numerous and weighty criticisms upon the first edition
of this Commentary (pubHshed in 1905 and now out of
print) were doubly welcome to the author as showing a
widespread interest in the subjects discussed, and as
enabling him to profit from the collaboration of eminent
specialists in the elucidation of Magna Carta and of the age
that gave it birth. The last eight years have been fertile in
discussions on the form and contents, the historical setting,
and the constitutional value of the Great Charter. Mono-
graphs and contributions to periodical literature, devoted
exclusively to Magna Carta, have been published in France,
Germany and the United States of America, as well as in
Great Britain ; while few books have appeared on English
medieval history or on the development of English law
without throwing light incidentally on one or more of the
Charter's various aspects.
An endeavour has been made, by severe condensation,
to find room in this new edition for whatever seemed
relevant and of permanent value in this mass of new
material, without sacrificing anything of importance con-
tained in the first edition. Effect has been given, so far as
space permitted, to the suggestions cordially offered by
critics and fellow-workers, both privately and in published
books and articles; while the author's own recent
researches have supplied additional illustrations, and have
led him to modify several of his earlier impressions.
Although no reason has been found for altering funda-
mental propositions, the whole work has been recast ;
hardly a page, either of Commentary or of Historical Intro-
vi PREFACE TO SECOND EDITION
duction, remains as originally written ; and care has been
taken to supply the reader with references to the most recent
authorities on the various topics discussed or referred to.
The new material will be found mainly (i) in the portions
of the Introduction treating respectively of scutages, the
Coronation Charter of Henry I., the juridical nature of
Magna Carta, its contemporary and permanent effects on
constitutional development, its reissues by Henry HI., and
the nature of the so-called " unknown charter " of John ;
and (2) in chapters 12, 13, 14, 18, 20, 25, 27, 34, 38, 39 and
61 of the Commentary. In the Appendix, Professor
Liebermann's amended text of Henry I.'s Charter of
Liberties has been adopted, and the Great Charter of 1225
substituted for that of 1217; while an attempt has been
made, by means of italics and foot-notes, to show at a glance
the chief points in which the three reissues by Henry III.
differ from one another and from the Charter as originally
granted by John.
Latin Charters, of which the full text is given in the
Appendix or elsewhere, have been printed literatim as in
the authorities cited in each case; but for detached Latin
words or phrases, whether occurring in the Historical Intro-
duction or the Commentary, a uniform spelling has been
adopted, in which the " ae " diphthong, where appro-
priate, has been substituted for the less familiar " e."
The author's grateful acknowledgments are due to the
Trustees of the Carnegie Foundation, for a grant towards
the expenses of this edition ; to Professor Vinogradoff, for
help courteously given in solving problems affecting the
interpretation of chapter 34; and to Mr. David B. Mungo,
LL.B., formerly the author's assistant in the University of
Glasgow, for his services in reading the proof-sheets and
for many useful suggestions.
The University,
Glasgow,
December y 1913.
FROM PREFACE TO FIRST EDITION.
No Commentary upon Magna Carta has hitherto been
written from the standpoint of modern research. No
serious attempt has yet been made to supersede, or even
adequately to supplement, the works of Coke and Richard
Thomson, pubHshed respectively in 1642 and 1829, and
now hopelessly out of date. That this conspicuous gap in
our historical and legal literature should have remained so
long unfilled is the more remarkable in view of the great
advance, amounting almost to a revolution, which has been
effected since Coke and Thomson wrote. Within the last
twenty years, in especial, a wealth of new material has been
explored with notable results. Discoveries have been made,
profoundly affecting our views of every branch of law, every
organ of government, and every aspect of social and
individual life in medieval England. Nothing, however,
has hitherto been done towards applying to the systematic
elucidation of Magna Carta the new stores of knowledge
thus accumulated.
With this object in view, I have endeavoured, throughout
several years of hard, but congenial work, to collect, sift,
and arrange the mass of evidence, drawn from many
scattered sources, capable of throwing light upon John's
Great Charter. The results have now been condensed into
the Commentary which fills two -thirds of the present
volume. This attempt to explain, point by point, the
sixty-three chapters of Magna Carta, embracing, as these
do, every topic — legal, political, economic and social — in
which John and his barons felt a vital interest, has involved
an analysis in some detail of the whole public and private
viii PREFACE TO FIRST EDITION
life of England during the thirteenth century. The Com-
mentary is preceded by a Historical Introduction, which
describes the events leading to the crisis of 1215, analyzes
the grievances which stirred the barons to revolt, discusses
the contents and characteristics of the Charter, traces its
connection with the subsequent course of English history,
and gives some account of previous editions and commen-
taries.
February^ 1905.
PAGE
CONTENTS.
HISTORICAL INTRODUCTION.
Part I. Events leading to Magna Carta :
'^I. William I. to Henry II. : Main Problem, the Monarchy - 3
II. William I. to Henry II. : Problem of Local Government 13
III. William I. to Henry II. : Problem of Church and State - 16
lV. Richard I. and John ig
V. The Years of Crisis (1213-15) ..... 27
VI. Runny mede, and after - 36
Part II. Feudal Grievances and Magna Carta :
^ / I. The Immediate Causes of the Crisis ^ - - - - 48
II. The Crown and Feudal Obhgations - . .
v/ I III. Royal Justice and Feudal Justice
Part III. Magna Carta : Its Form and Contents :
I. Its Prototypes : Earlier Charters 93
^ II. Magna Carta : Its Form and Juridical Nature - 104
^' III. Magna Carta : Its Contents and Characteristics - - 109 \
IV. Magna Carta : An Estimate of its Value *^ - 120
^^ ^ V. Magna Carta : Its Defects - - ,^^_^ - - - 129
^ » VI. Magna Carta : Value of Traditional Interpretations - 132
VII. Magna Carta : Its Traditional Relation to Trial by Jury 134
X CONTENTS
Part IV. Historical Sequel to Magna Carta :
PAGB
I. Reissues and Confirmations of the Great Charter - - 139'
(i) Reissue of 1216 139
(ii) Reissue of 1217 145
(iii) Reissue of 1225 152
(iv) Confirmations (1237 to 1297) - - - '- - 157
II. Magna Carta and the Reforms of Edward I. - - - 159
Part V. Magna Carta ; Original Versions, Printed
Editions, and Commentaries :
I. Manuscripts of Magna Carta and Relative Documents - 165
II. Previous Editions and Commentaries - - - - 176
MAGNA CARTA:
TEXT, TRANSLATION, COMMENTARY.
PREAMBLE.
I. The King's Title. II. The names of the consenting Nobles. III.
The Motives of the Grant - - - - - - -186
CHAPTER ONE.
1. The Rights of the National Church : (i) Quod Anglicana ecclesia
libera sit ; (2) Canonical Election. II. Civil and Political
Rights --- 190
CHAPTER TWO.
I. Assessment of Rehefs. II. Units of Assessment : (i) Feodum
militis integrum ; (2) Baronia Integra ; (3) Baronia comitis
Integra. III. LiabiUty of Church Property to " Relief " - • 196
CHAPTER THREE.
i£\ No Relief after Wardship • 203
9>
CHAPTER FOUR.
Wardship : (i) The Definition of Waste ; (2) Punishment of Waste-
ful Guardians ; (3) Provision against Recurrence - - - 205
CONTENTS xi
CHAPTER FIVE. pack
]^I. The Obligations of the Warden of a Lay-fief. II. Wardship over
^ Vacant Sees -- 209
CHAPTER SIX.
The Marriage of Wards -.* 212
CHAPTER SEVEN.
I, The Widow's Share of Real Estate : (i) Dower ; (2) Maritagium ;
(3) Hereditas. II. Her Share of Personal Estate. III. Pro-
vision for her Immediate Needs : (i) Quarantine ; (2) Estovers
of Common ---------- 215
CHAPTER EIGHT.
Marriage of Widows --------- 220
CHAPTER NINE.
Procedure for Enforcing Payment of Debts - • - - -221
CHAPTER TEN.
Usury. I. The History of the Jews in England. II. Legal Position
of the Jews 223
CHAPTER ELEVEN.
^ Widows and Children of Debtors to be Protected against Creditors 230
S^ CHAPTER TWELVE.
Protection of Crown Tenants from Arbitrary Exactions : (i)
Feudal Aids ; (2) Scutage. II. Protection of London from
Arbitrary Exactions : (i) Aid ; (2) Tallage ; (3) Comparison of
Aid and Tallage ; {4) London's attempts to escape Tallage ;
{5) Effects of Omission of Tallage from Magna Carta ; (6)
Nature of the Protection accorded to London ; (7) Later History ,-
of the Crown's Right to Tallage the Towns. III. Magna Carta 1 ^
and the Theory of ParUamentary Taxation - • • - 231
CHAPTER THIRTEEN.
Liberties and Free Customs of London «-••-• 240
CONTENTS
•X- CHAPTER FOURTEEN. p^ge
Method of obtaining the Common Council of the Kingdom. I.
Nature of the Summons. II. Composition of the Council.
III. Position of " Minor Barons." IV. Representation. V.
Powers of the Council. VI. Rights of Majorities and Minorities 248
9(r CHAPTER FIFTEEN.
Restrictions on Mesne Lords taking Aids. I. Points of difference
between tenants-in-chief and under-tenants. II. The influence /
of Magna Carta upon later practice - . - - , 256
CHAPTER SIXTEEN.
No one to perform greater service for a tenement than is due - • 260
•d^ CHAPTER SEVENTEEN.
Common Pleas. I. The Curia Regis as a Court of Law. II, Com-
mon Pleas and Royal Pleas. III. Influence of Magna Carta on
Genesis of the three Courts of Common Law. IV. Evolution
of the Court of Common Pleas. V. Common Pleas and the
Exchequer ... 261
:V CHAPTER EIGHTEEN.
Petty Assizes. I. The Curia Regis and the Travelling Justices,
II. Nature and Origin of the three Petty Assizes. III. Aims
of Magna Carta. IV. Effects of Magna Carta. V. An Erroneous
View. VI. Later History of the Justices of Assize - - ~ 269
;:(c CHAPTER NINETEEN.
Procedure at Petty Assizes • 282
^CHAPTER TWENTY.
Amercement. I. Three Stages of Criminal Law : (i) The blood-
feud ; (2) Fixed money-payments ; (3) Amercements. II.
Magna Carta and Amercements : (i) Of Freeholders ; (2) Of
Merchants ; {3) Of Villeins ; (4) Fines and Amercements ; (5)
Contenement 284
CHAPTER TWENTY-ONE.
Amercement of Earls and Barons 295
CONTENTS xiii
CHAPTER TWENTY-TWO. page
Amercement of the Clergy 298
CHAPTER TWENTY-THREE.
I. Origin of Obligation to make Bridges. II. The King's Interest
in the Repair of Bridges. III. Erroneous Interpretations - - 299
At CHAPTER TWENTY-FOUR.
I. Pleas of the Crown. II. Keeping and Trying Criminal Pleas.
III. The Intention of Magna Carta. IV. An Erroneous View.
V. Local Magistrates under John : (i) The Sheriff ; (2) The
Constable ; (3) The Coroner ; (4) The Bailiff - • - - 304
CHAPTER TWENTY-FIVE.
Farms of Counties and Hundreds - - - - - - -317
CHAPTER TWENTY-SIX.
Crown Debtors. I. Nature of the Grievance. II. The Right to
Bequeath - .-.- 321
CHAPTER TWENTY-SEVEN.
Intestate Succession 326
CHAPTER TWENTY-EIGHT.
I. Purveyance in General. II. Branches of it restricted by Magna
Carta. III. Its other Branches ..-..- 329
CHAPTER TWENTY-NINE.
Castle-Guard - 333
CHAPTER THIRTY.
Purveyance of Horses and Carts - 334
CHAPTER THIRTY-ONE.
Purveyance of Timber 336
xiv CONTENTS
CHAPTER THIRTY-TWO. page
I. The Crown's Claim to Felons' Property : (i) Lands ; (2) Chattels.
11. Indictment, Conviction, and Attainder - • - - 336
CHAPTER THIRTY-THREE.
Obstructions to be removed from Rivers . • # • - 343
9f CHAPTER THIRTY-FOUR.
The Writ " Praecipe." 1. Royal Writs and the Feudal Jurisdictions.
^ / II. The Intention of Magna Carta. III. Expedients for evading
^y Magna Carta. IV. Influence upon Later Legal Development - 346
CHAPTER THIRTY-FIVE.
Standards of Weights and Measures - - - • • • 356
4;' CHAPTER THIRTY-SIX.
Writ de Odio et Atia. I. Trial by Combat prior to John's Reign.
II. Writ of Life and Limb. 111. Its Subsidiary Uses. IV.
Later History of Appeal and Battle 359
%
CHAPTER THIRTY-SEVEN.
Prerogative Wardship -.-- 367
CHAPTER THIRTY-EIGHT.
No BaiUff to put anyone to his " law " without Witnesses. I.
Medieval Interpretations. 11. Modern Interpretations. III.
Nature of the Grievance ----••• 369
CHAPTER THIRTY-NINE.
. Its Main Object : (i) Judgment must precede execution ; (2) Per
judicium parium ; (3) Per legem terrae ; (4) Meaning of " vel."
11. The scope of the Protection afforded. Ill, What classes of
men enjoyed it. IV. Reactionary Aspects. V. Genesis of this
Chapter. VI. Later History of " Judgment of Peers." VII.
Erroneous Interpretations ----.-- 375
CHAPTER FORTY.
Justice not to be Sold, Refused, or Delayed ----- 395
CONTENTS XV
CHAPTER FORTY-ONE. p^ge
Freedom of Trade. I. Magna Carta favours Alien Merchants. II.
Customs and Tolls. III. The Motives prompting this Chapter.
IV. English Boroughs and Merchant Strangers - - - 398
CHAPTER FORTY-TWO.
Liberty to leave the Kingdom — Writs ne exeat regno - - . 407
CHAPTER FORTY-THREE.
Tenants of Escheated Baronies - . - -
411
CHAPTER FORTY-FOUR.
I. The Royal Forests. II. Their Origin. III. Forest Officials. IV.
Forest Courts. V. Chases, Parks, and Warrens. VI. Forest
Rights and Forest Grievances. VII. Later History of Forests
and Forest Laws .--.---.. ^i^
CHAPTER FORTY-FIVE.
Justices, Castellans, Sheriffs, and Bailiffs to be law-abiding men - 431
CHAPTER FORTY-SIX.
Wardship over Vacant Abbeys 433
CHAPTER FORTY-SEVEN.
Forests and River-banks encroached upon by John - . • 435
CHAPTER FORTY-EIGHT.
Procedure for aboUshing Evil Customs of Forests and elsewhere - 438
CHAPTER FORTY-NINE.
Hostages and Charters to be restored - - - - - -441
CHAPTER FIFTY.
List of those excluded from offices of trust in future - - - 444
CHAPTER FIFTY-ONE.
Banishment of Foreign Mercenaries ---••- 447
xvi CONTENTS
CHAPTER FIFTY-TWO. page
Procedure for redressing wrongful Disseisins - - - - - 448
CHAPTER FIFTY-THREE.
A Crusader's Respite allowed to John 450
CHAPTER FIFTY-FOUR.
fcjtTS Right of Appeal by Women 451
CHAPTER FIFTY-FIVE.
Remission of Unjust Fines and Amercements - - - 454
CHAPTER FIFTY-SIX.
Redress for Welshmen wrongfully disseised by John - - - 456
CHAPTER FIFTY-SEVEN.
Redress for Welshmen wrongfully disseised by Henry II. or Richard I. 457
CHAPTER FIFTY-EIGHT.
Welsh Hostages and Charters to be restored 458
CHAPTER FIFTY-NINE.
Justice to be done to King of Scots ; Relations of England and
Scotland 459
CHAPTER SIXTY.
Extension of Provisions of Charter to Tenants of Mesne Lords - 463
CHAPTER SIXTY-ONE.
Forma Securitatis or Legal Sanction of the Charter. I. Nature of
the Security. II. Details of Scheme : (i) Twenty-five
Executors ; (2) A quorum may act ; (3) Sub-committee of
four ; (4) Local Agents ; (5) Co-operation of Public. III.
Relations to Contemporary Theory. IV. Modem Criticism.
V. Failure of Scheme - - ----- - 465
CHAPTER SIXTY-TWO.
Prelates to issue Letters Testimonial 478
CONTENTS xvii
CHAPTER SIXTY-THREE. pace
Formal Clauses 479
APPENDIX.
Documents Relative to, or Illustrative of, John's
Magna Carta :
I. The Charter of Liberties of Henry I. (i 100) - - - 481
II. The Second or Oxford Charter of Stephen (1136) - - 483
III. Charter of Henry II. {circa 1154) ----- 485
IV. The so-called " Unknown Charter of Liberties " - - 485
V. The Articles of the Barons (1215) ----- 487
yi. Writs Supplementary of John's Great Charter - - 493
VII. The Great Charter of Henry III. (Third Re-issue, 1225) 497
VIII. Carta de Foresta (1217) 508
Select Bibliography and List of Authorities referred
TO - - - - - - - - - - 513
Index of Statutes 519
General Index 521
HISTORICAL INTRODUCTION
PART I.
EVENTS LEADING TO MAGNA CARTA.
The Great Charter is too often treated as the outcome of
accidental causes ; its sources are traced no deeper than the
personal tyrannies and blunders of King John. That mon-
arch's misdeeds are held to have goaded into action a
widespread opposition that never rested until it had achieved
success; and the outcome of this success was the Great
Charter of Liberties. The moving causes of events of
tremendous moment are thus sought in the characteristics
and vices of one man. If John had never lived and sinned,
so it would appear, the foundations of English freedom
would never have been laid.
Such shallow views of history fail to comprehend the
magnitude and inevitable nature of the sequence of causes
and effects upon which great issues depend. The com-
pelling logic of events forces a way for its fulfilment,
independent of the caprices, aims and ambitions of
individual men. The incidents of John's career are the
occasions, not the causes, of the movement that laid the
foundations of English liberties. The origin of Magna
Carta lies too deep to be determined by any purely contin-
gent phenomena. It is as unwise as it is unnecessary to
suppose that the course of constitutional development in
England was violently wrested into a new channel, merely
because of the incapacity or cruelties of the temporary occu-
pant of the throne. The source of the discontent fanned
to flame by John's oppressions must be sought in earlier
reigns. The genesis of the Charter cannot be understood
apart from its historical antecedents.
4 EVENTS LEADING TO MAGNA CARTA
It is thus necessary briefly to narrate how the scattered
Anglo-Saxon and Danish tribes and territories, originally
unconnected, were slowly welded together and grew into
England; how this fusion was made permanent by the
growth of a strong centralized government which crushed
out local independence, and threatened to become the most
absolute despotism in Europe; how, finally, the Crown,
because of the very plenitude of its power, called into play
opposing forces, which set limits to royal prerogatives and
laid the foundations of the reign of law. Such a survey of
the early history of England reveals two leading movements ;
the establishment of a strong Monarchy able to bring
order out of anarchy, and the establishment of safeguards
to prevent this source of order from degenerating into
an unrestrained tyranny, and so crushing out not merely
anarchy but legitimate freedom as well. The later move-
ment, in favour of liberty and the Great Charter, was the
natural complement, and, in part, the consequence of the
earlier movement in the direction of a strong government
able to enforce peace. In historical sequence, order pre-
cedes freedom.
These two problems, mutually complementary, arise in
the history of every nation, and in every age : the problem
of order, or how to found a central government strong
enough to suppress anarchy, and the problem of freedom,
or how to set limits to an autocracy threatening to over-
shadow individual liberty. Deep political insight may still
be acknowledged in '^sop's fable of Jupiter and the frogs.
King Log proves as ineffective against foreign invasion as
he is void of offence to domestic freedom; King Stork
secures the triumph of his subjects in time of war, but
devours them in time of peace. All nations in their early
efforts to obtain an efficient government have to choose
between these two types of ruler — between an executive,
harmless but weak ; and one powerful to direct the business
of government at home and abroad, but ready to use powers
entrusted to him for the good of all, for his own selfish aims
and the trampling out of his subjects' liberties.
On the whole, the miseries of the long centuries of Anglo-
EVENTS LEADING TO MAGNA CARTA 5
Saxon rule were the outcome of the Crown's weakness;
i while, at the Norman Conquest, England escaped from the
; mild sceptre of inefficiency, only to fall under the cruel
\^ sceptre of selfish strength. Yet the able kings of the new
dynasty, powerful as they were, had to struggle to maintain
their mastery ; for the unruly barons fought vigorously to
shake off the royal yoke.
During a century of Norman rule, constant warfare was
waged between two great principles — the monarchic, stand-
ing on the whole for order, seeking to crush anarchy, and
the oligarchic or baronial, standing on the whole for local
autonomy, protesting against the tyranny of autocratic
power. Sometimes one of these gained the ascendant;
sometimes the other. The history of medieval England is
the swing of the pendulum between.
The main plot, then, of early English history, centres
in the attempt to found a strong monarchy, and yet to
set limits to its strength. With this main plot subordinate
plots are interwoven. Chief among these must be reckoned
the necessity of defining the relations of the central to the
local government, and the need of an acknowledged frontier
between the domains of Church and State. On the other
hand, all that interesting group of problems connected with
the ideal form of government, much discussed in the days of
Aristotle as in our own, is notably absent, never having
been forced by the logic of events upon the mind of medieval
Europe. Monarchy was accepted as the only possible
scheme of government ; the merits of aristocracy and demo-
cracy, or of the much-vaunted constitution known as
"mixed" were not discussed, since these forms of consti-
tution did not lie within the sphere of practical politics.
The student of history will do well to begin by concentrat-"
ing his attention on the main problem, to which the others
are subsidiary.
I. William I. to Henry II.— Main Problem : the Monarchy.
The difficulties that surrounded the English nation in its
early struggles for existence were formidable. The great
problem was, first, how to get itself into being, and there-
6 EVENTS LEADING TO MAGNA CARTA
after how to guard against the forces of disintegration,
which strove without rest to tear it to pieces again. The
dawn of EngHsh history shows the beginning of that long
slow process of consolidation in which unconscious reason
played a deeper part than human will, whereby many dis-
cordant tribes and races, many independent provinces, were
crushed together into something bearing a rude likeness to
a united nation. Many forces converged to the achieve-
ment of this result. The coercion of strong tribes over
weaker neighbours, the pressure of outside foes, the growth
of a body of law, and of public opinion, the influence of
religion as the friend of peace, all helped to weld together a
chaos of incongruous and warring elements.
It is notable that each of the three influences, destined
ultimately to aid in this process of unification, threatened at
one time a contrary effect. Thus the rivalries of the smaller
kingdoms tended towards disruption before Wessex gained
undisputed supremacy; the Christianizing of England,
partly by Celtic missionaries from the north and partly by
emissaries from Rome, threatened to split the country into
two, until mutual rivalries were stilled after the Synod of
Whitby in 664; and one effect of the settlements of the
Danes was to create a barrier between the lands that lay
on either side of Watling Street, before the whole country
succumbed to the unifying pressure of Canute and his sons.
The stern discipline of foreign conquest was required to
make national unity possible; and, with the restoration of
the old Wessex dynasty in the person of Edward Confessor,
the forces of disintegration again made headway. England
threatened once more to fall to pieces, but the irofi rule of
the Normans came to complete what the Danes had begiin
half a century before. As the weakness of the Anglo-
Saxon kings and the disruption of the country had gone
hand in hand; so the complete unification of England was
the result of the Norman despotism.
Thereafter, it was the strength of its monarchy that
rendered England unique in medieval Europe. Three
x^ kings in particular contributed to this result — William
the Conqueror, Henry Beauclerk, and Henry Plantagenet.
WILLIAM I. TO HENRY II. 7
In a sense, the work of all three was the same; to build up
the central authority against the disintegrating effects of
feudal anarchy. But the policy of each was modified by
changing times and needs. The foundations of the edifice
were laid by the Conqueror, whose character and circum-
stances combined to afford him an opportunity unparalleled
in history. The difficulties of his task, and the methods
by which he secured a successful issue, are best understood
in relation to the nature of the obstacles to be overcome.
Feudalism was the great current of the age — a tide formed
by many converging streams, all flowing in the same
direction, unreasoning like the blind powers of Nature,
carrying away or submerging every obstacle in its path.
In other parts of Europe — in Germany, France, and Italy,
as in Scotland — the ablest monarchs found their thrones
endangered by this feudal current. In England alone the
monarchy stood firm. William I. refrained from any
attempt to stay the torrent; but, while accepting it, he
made it serve his own purposes. He watched and modified
the tendencies making for feudalism, which he found in
England, and he profoundly altered the feudal usages
and rights transplanted from Norman soil. The special
expedients used by him for this purpose are well known,
and are all closely connected with his crafty policy of
balancing Anglo-Saxon against Norman elements, and of
selecting what suited him in either. He encouraged the
adoption in England of feudalism, considered as a system
of land tenure and of social distinctions based on the
possession of land; but he successfully checked the evils
of its unrestrained growth as a system of local government
and jurisdiction.
William's policy was one of balancing. Not content to
depend entirely on the right of conquest, he insisted on
having his title confirmed by a body claiming to represent
the Witenagemot, and alleged that he had been named
successor by his kinsman, Edward Confessor, a nomination
strengthened by the renunciation of Harold in his favour.
Thus, to Norman followers claiming to have set him by
force of arms on his throne, William might point to the
8 ^ EVENTS LEADING TO MAGNA CARTA
election by the Witan, while for his English subjects,
claiming to have elected him, the presence of foreign
troops was an effective argument. Throughout his reign,
he played off the old English laws and institutions against
the new Norman ones, with himself as umpire over all.
He retained, too, the popular moots or meetings of the'
shire and hundred as a counterpoise to the feudal juris-
dictions; the fyrd or militia of all free men as a set-off -
to the feudal levy; and whatever incidents of the Anglo-
Saxon land tenures he thought fit.
Thus the Norman feudal superstructure was built on a
I ii"^' basis of Anglo-Saxon usage and tradition. William, how-
ever, did not shrink from innovations where these suited
his purpose. The great earldoms into which England
had been divided, even down to the Norman Conquest, were
abolished. New earldoms were indeed created, but on a
different basis. Even the great officers subsequently
known as Earls Palatine, always few in number, never
attained to the independence of the Anglo-Saxon Ealdor-
men. William was chary of creating even ordinary Earls,
and such as he did create soon became mere holders
of empty titles of honour, ousted from all real power by
the Norman vicecomites or sheriffs. No English earl was
a " Count " in the continental sense of a real ruler of a
" County." No earl was allowed to hold too large an
estate within his titular shire.
ingenious devices were used for checking the feudal
excesses so prevalent on the Continent. Rights of private
war, coinage, and castle-building, were jealously circum-
scribed; while private jurisdictions, although tolerated as
a necessary evil, were kept within bounds. The manor
was in England the normal unit of seignorial jurisdiction ;
the higher courts of Honours were exceptional. No appeal
lay from the manorial court of one magnate to that of his
over-lord, while, in later reigns at least, appeals were
encouraged to the Curia Regis, The results of this policy
have been aptly summarized as " a strong monarchy, a
relatively weak baronage, and a homogeneous people."
During the reign of William II. (1087-1100) the Con-
WILLIAM I. TO HENRY II. 9
stitution made no conspicuous advance. The foundations
had been laid; but Rufus was more intent on his hunting
and enjoyments, than on the deeper matters of statecraft.
Minor details of feudal organization were doubtless settled
by the King's Treasurer, Ralph Flambard ; but the extent
to which he innovated on the practice of the elder William
is matter of dispute. On the whole, the reign must be
reckoned a time of comparative rest between two periods
of advance.
Henry I. (1100-35) took up, with far-seeing statesman-
ship and much vigour, the work of consolidation. His
policy shows an advance upon that of his father. William
had been content to curb the main vices of feudalism.
Henry introduced within the Curia Regis itself a new class
of men, representing a new principle of government. The
great offices of state, previously filled by holders ofrs..
barbmes, were now given to creatures of Henry's own, men "^
of humble birth, whose merit had raised them to his favour,
and whose only title to power lay in his goodwill. Henry's
other great achievement was the organization of the Exche-^
quer,^ as a source of royal revenue, and as an instrument
for making his will felt in every corner of England. For
this great work he was fortunate to secure in Roger,
Bishop of Salisbury, the help of a minister who combined
genius with painstaking ability. At the Exchequer, as _
organized by the King and his minister, the sheriff of each
county twice a year, at Easter and at Michaelmas, rendered
account of every payment that had passed through his
hands. His balance was adjusted before all the great
officers of the King's household, who subjected his accounts
to close scrutiny. Official records were drawn up, one of
which — the famous Pipe Roll of 1130 — is extant at the
present day. As the sums received by the sheriff affected
every class of society in town and country, these half-yearly
audits enabled the King's advisers to scrutinize the lives
and conduct of high and low. , These half-yearly investiga-
tions were rendered more effective by the existence at the
Exchequer of a great record of every landed estate in Eng-
land. With this the sheriffs' returns could be compared
10 EVENTS LEADING TO MAGNA CARTA
and checked. Henry's Exchequer thus found one of its
most powerful weapons in the great Domesday Survey,
the most enduring proof of the statesmanship of the Con-
queror, by whose orders and under whose direction it had
been compiled.
The central scrutiny conducted within the Exchequer
was supplemented by occasional inspections conducted in
each county. The King's representatives, including
among them the officers who presided over the half-yearly
audit, visited, at intervals still irregular, the various shires.
These Eyres, as they were called, were at first under-
taken chiefly for financial purposes. The sheriffs' accounts
rendered at Westminster were checked locally on the scene
of their labours. These investigations necessarily involved
the trial of pleas. Complaints of oppression at the hands
of the local tyrant were made and determined on the spot ;
gradually, but not until a later reign, the judicial business
became equally important with the financial, and ultimately
even more important.
Henry, before his death in 1135, seemed to have carried
to completion the congenial task of building a strong mon-
archy on the foundations laid by William. Much of his
work was, however, for a time undone, while all of it
seemed in imminent danger of perishing for ever, because
he left no male heir of his body to succeed him. His
daughter's claims were set aside by Stephen^ son of the
Conqueror's daughter, and a cadet of the House of Blois,
to whom Henry had played the indulgent uncle, and who
repaid his benefactor's generosity by constituting himself
his heir. Stephen proved unequal to the task of preserving
the monarchy intact from the forces that beat around the
throne. His failure is attributed by some to personal
characteristics; by others, to the defective nature of his
title, combined with the presence of a rival in the field in
the person of his cousin, Henry's daughter, the ex-Empress
Matilda. The nineteen years of anarchy which nominally
formed his reign did nothing — and worse than nothing — to
continue the work of his great ancestors. The power of
the Crown was humbled : England was almost torn in
I
WILLIAM I. TO HENRY II. ii \
fragments by the claims of rival magnates to local inde-
pendence.
With the accession of Henry II. (1154) the tide quickly
turned, and turned for good7 Ofthe numerous steps taken
by him to complete the work of the earlier master-builders
of the English Monarchy, only a few need here be
mentioned. Ascending the throne in early manhood, he
brought with him a statesman's instinct peculiar to himself,
together with the unconquerable energy common to his
race. He rapidly overhauled every institution and every ^^
branch of administration^ The permanent Cj[ma_i?^g^>l*
was not only restored to working or3er, but improved~iri
eacE~or its many aspects-^asIHe Kmj^V^househoTg^^
finanoiFbugauT^^^ administrative centre nfjJT^Jdng,
dom, and as the vehicle of royal justice. The Exchequer>y^ -
which was originally merely tlie~Curia in its financial ^
aspect, received the re-organization so urgently needed
after the terrible strains to which it "Rad beerTsubjecfe^'.
The Pipe Rolls were revived and financial reforms effected. <
The old popular courts of hundred and county^, and the^'^ ^'
feudal jurisdictions were brought under more effective
control of the central gfovernment by the restoratiorToTThe
„.^ — — £2 — .. . iww J
system^^oTEvres with their travelling justices, whose visits
were now placed on a more systematic basis. Equally
important were the Kiaglsj^re in Jhe selff tion of fitmen v^U
for the duties of Sheriff, the frequent punishment and
removal from ofBce of offenders, and the restored control
over all in authority. Henry was strong enough to employ
more substantial men than the novi homines of his grand-
father without suffering them to get out of hand. Another
expedient for controlling local courts was the calling up^|K -.
of cases to his own central feudal Curia, or before those "
benches of professional judges, the future King's Bench
and Common Pleas, that formed as yet merely committees
of the Curia as a whole.
Closely connected with these innovations was the new
system of procedure instituted by Henry. The chief
feature was that each litigation must commence with an
appropriate, rgy^l writ issued from the Chancery. Soon
12 EVENTS LEADING TO MAGNA CARTA
for each class of action was devised a special writ, and the
system came to be known as " the writ system." A striking
feature of Henry's policy was the bold manner in which he
threw open the doors of his royal Courts of Law to all-
yf comers~^excepting villeins), and^provided" there^ — always
iiV- return for hard cash» be it said — a better article in^ame
of^justice than could be procured elsewhere in England,
or, for that matter, elsewhere in Europe. "Thus, not oTily
was the Exchequer filled with fines and fees, but, insidi-
ously and without the danger involved in a frontal attack,
Henry sapped the strength of the great feudal magnates,
and dmil£d the stream of litigants from manorial courts
to his own. The same policy had_aJlllther result in taciIT-
tating the growth of a body of common law, uniform
throughout the length and breadth of England, opposed
to the varying usages of localities and individual baronial
courts.
C^; The reorganization of the army was another reform that
helped to strengthen the throne of Henry and his sons.
This was effected in various ways : partly by the revival
and more strict enforcement of obligations connected with
the Anglo-Saxon fyrd, under the Assize of Arms (1181),
which compelled every freeman to maintain at his own
expense weapons and warlike equipment suited to his station
in life; partly by the ingenious method of increasing
the amount of feudal service due from Crown tenants,
based upon an investigation instituted by the Crown and
upon the written replies returned by the barons, known
to historians as "the Cartae of 1166"; and partly by the
development of the principle of scutage, a means whereby
unwilling military service, limited as it was by annoying
restrictions as to time and place, might be exchanged at
the option of the Crown for money, with which a more
7- flexible army of mercenaries might be hired.
By these expedients and many others, Henry raised the
English monarchy, always in the ascendant since the Con-
quest, to the very zenith of its power, and left to his sons
the entire machinery of government in perfect working
order, combining high administrative efficiency with great
WILLIAM I. TO HENRY II. 13
strength. Full of bitter strifes and troubles as his reign
of thirty-five years had been, nothing had interfered with
the vigour and success of the policy whereby he tightened
his hold on England. Neither the long struggle with Becket,
ending as it did in Henry's personal humiliation, nor the
unnatural warfare with his sons, which hastened his death
in 1 189, was allowed to interfere with his projects of reform
in England.
The last twenty years of life had been darkened for him,
and proved troubled and anarchic in the extreme to his
continental dominions; but in England profound peace
reigned. The last serious revolt of the powers of feudal
anarchy had been suppressed in 11 74 with characteristic
thoroughness and moderation. After that date, the Eng-
lish monarchy retained its supremacy almost without an
effort.
M William I. to Henry II.— Problem of Local Government.
It is necessary to retrace our steps in order to consider
the subsidiary problem of local government. The failure
of the Princes of the House of Wessex to devise adequate
machinery for keeping the Danish and Anglian provinces
in subjection to their will was one main source of the weak-
ness of their monarchy. When Duke William solved this
problem, he took an enormous stride towards establishing
his throne on a securer basis.
Every age has to face, in its own way, a group of diffi-
culties essentially the same, although assuming different
names as Home Rule, Local Government, or Federation.
Problems as to the proper nature of the local authority,
the extent of its powers, and its relation to the central
government, require constantly to be re-stated and solved
anew. The difficulties involved, always great, were
unspeakably greater in an age when no proper administra-
tive machinery existed, and when rapid communication
and serviceable roads were unknown. Lively sympathy
is excited by consideration of the difficulties that beset the
path of King Edgar or King Ethelred, endeavouring to *
rule from Winchester the distant and alien races of North-
14 EVENTS LEADING TO MAGNA CARTA
umbria, Mercia, and East Anglia. If a weakling governed
a distant province, anarchy would result and the King's
authority might suffer with that of his inefficient repre-
sentative; while a powerful viceroy might consolidate his
own authority and then defy his King. The two horns of
this dilemma are amply illustrated by the course of early
English history. The West-Saxon Princes vacillated
between two lines of policy ; spasmodic attempts at
centralization alternated with periods of local autonomy.
The scheme of Edgar and Dunstan has sometimes been
described as a federal or home-rule policy — as a frank
surrender of the attempt to control exclusively from one
centre the mixed populations of Northern and Midland
England. Their solution was to relax rather than tighten
the bond ; to entrust with wide powers the local viceroy in
each district, and to aim at a loose federal empire — a union
of hearts, rather than a centralized despotism founded on
coercion. The dangers of such a system are obvious,
where each ealdorman commanded the troops of his
province.
Canute's consolidating policy has been the subject of
much discussion, and has sometimes been misunderstood.
The better opinion is that, with his Danish troops behind
him, he felt strong enough to reverse Dunstan's tactics by
decisive action in the direction of centralization. His
provincial viceroys (jarls or earls, as they were now called)
were appointed on a new basis : England was mapped out
into new administrative districts under viceroys having no
hereditary connection with the provinces they governed.
In this way Canute sought to arrest the process by which
England was breaking up into a number of petty kingdoms.
If these viceroys were a source of strength to the powerful
Canute, they proved a source of weakness to the saintly
Confessor, who was forced to submit to the control of his
provincial rulers, such as Godwin and Leofric, as each in
turn gained the upper-hand in the field or among the Witan.
The process of disintegration continued until the coming of
the Conqueror changed the relations between the monarchy
and the other factors in the national life.
WILLIAM I. TO HENRY 11.
Among the expedients adopted by the Norman Duke
for curbing his feudatories in England, one of the most
important was the reorganization of the system of provincial
rulers. The real representative of the King in each group
of counties was now the sheriff, not the earl. His Latin,
name of vicecomes is misleading, since that officer in no
sense represented the earl or comes, but acted as the direct
agent of the Crown. The name " viceroy " more accurately
describes his actual position and functions. •
The problem of local government, however, was not
eradicated: it only took a different form. The sheriffs
themselves, relieved from the earl's rivalry, tended to
become too powerful. If they never dreamed of openly
defying the royal authority, they thwarted its exercise,
appropriated to their private uses items of revenue, pushed
their own interests, and punished their own enemies, while
acting in the King's name. The office threatened to
become territorial and hereditary,^ and its holders aimed
at independence. Safeguards were found against the
sheriffs' growing powers, partly in the organization of the
Exchequer and partly in the itinerant justices, who took
precedence of the sheriff and heard complaints against his
misdeeds in his own county. By such measures, Henry I.
seemed almost to have solved these problems before his
death; but his success was apparent rather than real.
The incompleteness of Henry's solution became evident
under Stephen, when the leading noble of each locality
tried, generally with success, to capture both offices for
himself : great earls like Ralph of Chester and Geoffrey
of Essex compelled the King not only to confirm them as
sheriffs in their own titular counties, but also to confer on
them exclusive right to act as justices.
With the accession of Henry II. some advance was made
towards a permanent solution. That great ruler was strong
enough to prevent the growth of the hereditary principle
as applied to offices either of the Household or of local
magistrates. The sheriffs were frequently changed, not
only by the drastic and unique measure known as the
^ In one county, Westmoreland, the office did become hereditary.
l6 EVENTS LEADING TO MAGNA CARTA
Inquest of Sheriffs, but systematically, and as a matter of
routine. Their power tended in the thirteenth century to
decrease, chiefly because they found important rivals not
only in the itinerant judges, but also in two new officers
first heard of in the reign of Richard I., the forerunners
of the modern Coroner and Justice of the Peace respectively.
All fear that the sheriffs as administrative heads of districts
might defy the Crown was thus ended. Yet each of them
remained a petty tyrant over the inhabitants of his own
bailiwick. While the Crown was able and willing to
avenge neglect of its own interests, it was not always
sufficiently alert to punish wrongs inflicted upon its humble
subjects. The problem of local government, then, was
fast taking a new form, namely, how best to protect the
weak from unjust fines and oppressions inflicted on them-
by local magistrates. The sheriff's local power was no
longer a source of danger to the monarch, but had become
an effective part of the machinery which enabled the Crown
to levy with impunity its always increasing taxation.
III. William I. to Henry II.— Problem of Church and
State.
The Church had been, from an early date, in tacit
alliance with the Crown. The friendly aid of a line of
statesman-prelates from Dunstan downwards had given to
the Anglo-Saxon monarchy much of the little strength it
possessed. Before the Conquest the connection between
Church and State had been exceedingly close, so much so
that no one thought of drawing a sharp dividing line
between. What afterwards became two separate entities
were at first merely two aspects of one society, which
comprehended all classes of the people. Change came with
the Norman Conquest ; for the English Church was brought
into closer contact with Rome, and- with the ecclesiastical
ideals prevailing on the Continent. Yet no fundamental
alteration resulted; the friendly relations that bound the
prelates to the English throne remained intact, while Eng-
lish Churchmen continued to look to Canterbury, rather
than to Rome, for guidance.
WILLIAM I. TO HENRY XL 17
Gratitude to the Pope for moral support in eifecting the
Conquest never modified William's determination to allow
no unwarranted papal interference in his new domains.
His letter, both outspoken and courteous, in reply to papal
demands is still extant : — " 1 refuse to do fealty nor will I,
because neither have I promised it, nor do I find that my
predecessors did it to your predecessors." Peter's pence
he was willing to pay at the rate recognized by his Saxon
Xjredecessors; but all encroachments would be politely repelled.
In settling the country newly reduced to his domination,
the Duke of Normandy found his most valuable adviser
in a former prior of the Norman Abbey of Bee, whom he
raised to be Primate of all England. No record has come
down to us of any serious dispute between William and
Lanfranc. Friendly relations between King and Arch-
bishop continued, notwithstanding Anselm's condemnation
of the evil deeds of Rufus. Anselm supported that
King's authority over the Norman magnates, even while
he resented his evil practices towards the Church. He
contented himself with a dignified protest (made emphatic
by a withdrawal of his presence from England) against
unfair exactions from English prelates, and against the
long intervals during which vacancies remained unfilled.
Returning at Rufus's death from a sort of honourable
banishment at Rome, Anselm found himself compelled, by
his conscience and the recent decrees of a Lateran Council, "
to enter on the great struggle of the investitures.
In many respects, the spiritual and temporal powers were
still indissolubly locked together. Each bishop was a
vassal of the king, holder of a Crown barony, as well as a
prelate of Holy Church. By whom, then, should a bishop
be appointed, by the spiritual or by the temporal power?
Could he without sin perform homage for the estates of
his See? Who ought to invest him with ring and crozier?
Anselm adopted one view; Henry the other. A happy
compromise, suggested by the King's statesmanship, or
possibly by Bishop Ivo of Chartres,^ healed the breach for
^ Adams, Pol. Hist, of Engl., II. 141. See, however, Davis, England under
Norman s^ 1 32.
B
i8 EVENTS LEADING TO MAGNA CARTA
the time being. The symbols of spiritual authority were
to be conferred by the Church, but each prelate must
perform fealty to the King before receiving them, and da
homage thereafter, but before he was actually anointed
as bishop. This compromise of 1106 did not embrace, it
would appear, any final understanding as to the method
of appointing bishops : " Canonical election " formed no
part of Henry's express concessions. 1
Henry, however, does not seem to have rejected openly
the claims of the capitular clergy, but only to have taken
steps to render them nugatory in practice. Some of the
leading prelates, administrative officials on whom the
Monarch could depend, took part in the election of bishops
and were usually able to secure the appointment of a candi-
date acceptable to the King.
The Church gained in power during Stephen's reign,
arid deserved the power it gained, since it remained the
only stable centre of good government, while other institu-
tions crumbled around it. It was not unnatural that
Churchmen should advance new claims, and we find them
adopting the watchword, afterwards so famous, " that the
Church should be free," a vague phrase, destined to be
embodied in Magna Carta. The extent of immunity thus
claimed was never defined : an elastic phrase might be
expanded with the ever-growing pretensions of the Church.
Churchmen made it clear, however, that they meant it to
include at least two principles — "benefit of clergy," and
"canonical election."
Henry II. attempted to define the position in the Con-'
stitutions of Clarendon (1164), clause 12 of which provided
that in filling vacant Sees the King should summon potiores
personas ecclesiae and that the " election " should>take place
in the King's chapel with consent of the King and consilio
personarum regni, vague words which seem to reserve to
Henry the decision as to who constituted " the more
influential persons of the church," whom he ought to
summon, thus enabling him to control elections (as his
^ Adams, FoL Hist, of Engl., II. 148. Contrast the older view in Stubbs,
Const. Hist., I. 342-3.
WILLIAM I. TO HENRY II. 19
grandfather had done) by means of ecclesiastics whose
loyalty to the Crown was undoubted. Henry, in conse- f\^
quence of his humiliation following on Becket's murder,
had to release the bishops from their oath to observe the .^
Constitutions. In 11 73 he gave a definite promise to allow
greater liberty in elections, and it was part of a new agree-
ment with Rome in 1176, that in normal circumstances
vacant sees should not be kept in the King's hands for
more than a year.^ Yet, in practice, he continued to
exercise a control not inferior to that enjoyed by his grand-
father. On the whole, the rights of the Church at the close
of the reign of Henry Plantagenet were not far different
from what had been set down in the Constitutions of
Clarendon. A new definition of the frontier between the
spiritual and temporal powers was the outcome of John's
need of allies on the eve of Magna Carta.
IV. Richard I. and John.
Henry II., before his death, had fulfilled the task of
restoring order : to effect this, he had brought to perfection
machinery of rare excellence, equally adapted for purposes
of taxation, of dispensing justice, and of general adminis-
tration. Great as was the power for good of this new
instrument in the hands of a wise and justice-loving king, -.
it was equally powerful for evil in the hands of an arrogant,? (Jy
or even of a careless monarch. All the old enemies of>
the Crown had been crushed. Local government, now?
systematized, formed a source of strength, not of weakness ;j
while the Church, whose highest offices were filled withi
officials trained in Henry's own Exchequer (differing widel}^
from the type of saintly monks like Anselm), still remaineq
the fast friend of the Crown. The monarchy was strong _
enough to defy any one section of the nation.
The very thoroughness with which the monarchy had
surmounted its early difficulties, induced in Henry's sue- /^,/\.
cessors an exaggerated feeling of security. The very "^
abjectness of the various factors of the nation, now prostrate
beneath the heel of the Crown, prepared them to sink their
^ Makower, Const. Hist, of Church, 24-26.
20 EVENTS LEADING TO MAGNA CARTA
mutual suspicions and to form a tacit alliance in order to
join issue with their common oppressor. Powers used
k ; ^; ,, moderatelyand^onj^ for national ends by Henry,'
^; were^bused~Tor^seIfisji ends by both his sons. Richard's
Ch heayy taxation and contemptuous indifference_to_Enjglish
/ interests reconciled men's minds to thoughts of change, and
prepared the basis of a combined oppositionto a power jHaT
threatened to grind all other powers t^ powder .
"rn~no direction were these^buses felt so seyerely as in
O taxation. Financial machinery had been elaborated to
perfection, and large additional sums could be squeezed
from eyery class by an extra turn of the screw. Richard
did not eyen require to incur the odium, since ministers,
his instruments, shielded him from the unpopularity of
his measures, while he pursued his own good pleasure
abroad in war and tournament without yisiting the subjects
« he oppressed. Twice only, for a few months in either case,
did Richard yisit England during a reign of ten years.
In his absence new methods of taxation were deyised,
affecting new classes of property ; in particuTarTpersoiial
effects — merchandise andother chattels — only once belcJre
(in 1 187, for the Saladin tithey placed under contribution —
now became a regular source of royal reyenue. The
isolated precedent of Henry's reign was followed when an
extraordinarily heayy leyy was required for Richard's
ransom. The yery heartiness with which England made
sacrifices to succour the Monarch in his hour of need was
turned against the tax-payers. Richard^showed no grati-
tude ; and, being deyoid of kindly interest in his subjects,
he argued that what had been paid once might equally^ well
be paid again. With exaggerated notions of the reyenue
to be extracted from England, he sent from abroad demand
after demand to his oyerworked justiciars for eyer-increas-
ing sums of money. The chief lessons of the reign are
connected with this excessiye taxation ; the consequent
discontent prepared the way for a new grouping of political
forces under John.
Some minor lessons may be noted :
(i) In Richard's absence the odium for his exactions fell
RICHARD I. AND JOHN 21
upon his ministers at home, who bore the burden meet
for his own callous shoulders, while he enjoyed an
undeserved popularity by reason of his bravery and
achievements, exaggerated as these were by the halo of
romance which surrounds a distant hero. Thus may be
traced some dim foreshadowing of the doctrine of minis-
terial responsibility, although analogies with modern
politics must not be pushed too far.
(2) Throughout the reign, parts of Henry's system,
technical details of taxation and reforms in the adminis-
tration of justice, were elaborated by Archbishop Hubert V-
Walter, connected with trial by jury on the one hand and
with election on the other.
(3) Richard is sometimes said to have inaugurated the \
golden age of municipalities. Many Charters, still extant, /
bear witness to the lavish hand with which he granted, on
paper at least, privileges to the nascent towns. John
Richard Green finds the true interest of the reign not in
the King's Crusade and French wars, so much as in his
supposed fostering care over the growth of municipal
enterprise.
The death of Richard on 6th April, 1199, brought with
it at least one important change; England was no longer
to be governed by an absentee. JjQhn; endeavoured to
shake himself free from the restraintsj2f_power|ii^^
and^onduct the workjol^overnment in his own way. The ,
result was an abrupt end to tTie progress~madH^in the I
previous reign towards ministerial responsibility. The
odium formerly exhausting itself on the justiciars of
Richard was now expended on John. While, previously,
men had sought redress in a change of minister, such
expectations could no longer deceive. A new element of
bitterness was added to injuries long resented, and the c
nobles who felt the pinch of heavy taxation were compelled
to seek redress in a new direction. All the forces of dis-
content played openly around the throne.
As is usual at the opening of a reign, the discontented
hoped that a change of sovereign would bring relief.
Heavy taxation had been the result of exceptional circum-
22 EVENTS LEADING TO MAGNA CARTA
stances : the new king would revert to the less burdensome
scale of his father's exactions. Such hopes were quickly
Q j disappointed. John's needs proved as great as Richard's :-
the excessive demands, both for money and for service, -
coupled with the unpoBularJuses-lo which these were_put,
form the keynote of the reign : they form also_Jhe back-
ground of Magna Carta. .
The reign falls naturally into three periods; the years
in which John waged a losing war with the King of France
(i 199-1206), the quarrel with the Pope (1206-13), the great
struggle with the barons (12 13-16).
The first seven years were for England comparatively
uneventful, except in the gradual deepening of disgust with
the King and all his ways. The continental dominions
were ripe for losing, and John precipitated the catastrophe
by injustice and dilatoriness. The ease with which Nor-
mandy was lost showed something more than the incapacity
of the King as a ruler and leader — John Softsword as
contemporary writers call him. It showed that the feudal
army of Normandy had come to regard the English Sove-
ireign as an alien. The unwillingness of the English
nobles to succour John has also its significance. The
descendants of the men who helped William I. to conquer
England had now a less vital interest in the land from
which they came. The estates of many of the original
Norman baronage, not unequally divided on both sides of
the Channel, had been split up by inheritance or escheat.
Some of John's barons were purely English landowners
with no interest at stake in France.
By his arbitrary and selfish home policy, the King had
alienated their sympathies. Some of his father's innova-
tions had been unpopular fromtlie^first, and be75ame~Tfie
objects of bitter^pposition m John's tactless hands^ TKe
whole administralion of justice, along with the entire feudal
system of land-tenure, with its military obligations, aids
^and incidents, were degraded into instruments of extortion,
of which details will be given under appropriate chapters
of the subjoined commentary. English discontent con-
tributed to the loss of Normandy, and that in turn left
I
RICHARD I. AND JOHN 23
English barons more free to attend to insular matters, and
so prepared the way for Magna Carta.
The death of Archbishop Hubert Walter on 13th July,
1205, deprived John of the services of the most experienced
statesman in England. It did more, for it marked the
termination of the long friendship between the English
Crown and the English Church : its immediate effect was
to create a vacancy, the filling of which led to a quarrel
with Rome.
John failed, as usual, to recognize the merits of abler
men, and saw in the death of his great Minister merely
the removal of an unwelcome restraint, and the opening
to the Crown of a desirable piece of patronage. He pre-
pared to strain to the utmost his rights in the election of
a successor to the See of Canterbury, in favour of one of
his own creatures, John de Grey, already by royal influence
Bishop of Norwich. Unexpected opposition to his will
was offered by the canons of the Cathedral Church, who
determined to appoint their own nominee, without waiting
either for the King's approval or the co-operation of the
suffragan bishops of the Province, who, in the three last
vacancies, had participated in the election, and had invari-
ably used their influence on behalf of the King's nominee.
Reginald, the sub-prior, was secretly elected by the monks,
and hurried abroad to obtain confirmation at Rome before
the appointment was made public. Reginald's vanity pre-
vented his keeping his pledge of secrecy, and a rumour
reached the ear of John, who brought pressure to bear on
a section of the monks, now frightened at their own
temerity, and secured de Grey's appointment in a second
election. The Bishop of Norwich was enthroned at
Canterbury, and invested by the King with the temporali-
ties of the See. All parties now sent representatives to
Rome. This somewhat petty squabble benefited none of
the original disputants; for Innocent III. was quick to
seize his opportunity. Both elections were set aside by
decree of the Papal Curia, in favour of the Pope's own
nominee, a certain Cardinal, English-born, but hitherto
little known in England, Stephen Langton by namej^O
24 EVENTS LEADING TO MAGNA CARTA
destined to play an important part in the history of the
land of his birth.
John refused to view this triumph of papal iarroganc^
in the light of a compromise — the view diplomatically
suggested by Innocent. The King, with the hot blood
common to his race, and the bad judgment peculiar to
himself, rushed headlong into a quarrel with Rome which
he was incapable of carrying to a successful issue. Full
details of the struggle, the interdicts and excommunications
hurled by the Pope, and John's measures of retaliation
against the unfortunate English clergy, need not be here
discussed; but it should be noted that Innocent, in 121 1,
released the English people from allegiance to their King.^
John was one day to reap the fruits of this quarrel in
bitter humiliation and in the defeat of his most cherished
aims ; but, for the moment, the breach with Rome seemed
to lead to a triumph for the King. The papal encroach-
ments furnished him with a pretext for confiscating the
property of the clergy. Thus his Exchequer was amply
replenished, while he was able for a time to conciliate his
most inveterate opponents, the northern barons, by remit-
ting during several years the hated burden of a scutage.
John had no intention, however, to forego his right to
resume the practice of annual scutages : on the contrary,
he executed a measure intended to make them more
remunerative. This was the Inquest of Service, ordered
on ist June, 1212.2
During these years, however, John temporarily relaxed
the pressure on his feudal tenants. His doing so failed to
gain back their goodwill, while he broadened the basis of
future resistance by shifting his oppressions to the clergy
and through them to the poor. Meanwhile, his power
was great. Speaking of 12 10, a contemporary chronicler
declares : " All men bore witness that never since the time
of Arthur was there a King who was so greatly feared in
England, in Wales, in Scotland, or in Ireland." ^
Some incidents of the autumn of 12 12 require brief
* Petit-Dutaillis, Louis VII I. ^ 30. ^See Round, Commune of London, 2.tt^
^ Histoire des diics^ p. 109.
RICHARD I. AND JOHN 25
notice, as well from their inherent interest as because they
find an echo in Magna Carta. Serious trouble had arisen
with Wales. Llywelyn (who had married John's natural
daughter Joan, and had consolidated his power under pro-
tection of the English King) now seized the occasion to
cross the border, while John was preparing for a new
continental expedition. The King changed his plans, and
prepared to lead his troops to Wales instead of France.
A muster was summoned for September at Nottingham,
and John went thither to meet his troops. Before tasting
meat, in Roger of Wendover's graphic narrative, he
hanged twenty-eight Welsh hostages, boys of noble family,
whom he held as sureties that Llywelyn would keep the
peace. ^
Almost immediately thereafter, two messengers arrived
simultaneously from Scotland and from Wales with un-
expected tidings. John's daughter, Joan, and the King of
Scots, each independently warned him that his English
barons were prepared to revolt, under shelter of the Pope's
absolution from their allegiance, and either to slay him or
betray him to the Welsh. In a panic he disbanded the
feudal levies ; and, accompanied only by his mercenaries,
moved slowly back to London. ^
Two of the barons, Robert Fitz- Walter, afterwards the
Marshal of the army which opposed John at Runnymede,
and Eustace de Vesci, showed their knowledge of John's
suspicions by withdrawing secretly from his Court and
taking to flight. The King caused them to be outlawed
in their absence, and thereafter seized their estates and
demolished their castles.-'^
These events of September, 121 2, rudely shook John
out of the false sense of security in which he had wrapped
^R. Wendover, III. 239.
^W. Coventry, II. 207; R. Wendover, III. 239.
^ From their possible connection with chapter 39 of Magna Carta, it may be
worth while to quote the words of Ralph de Coggeshall, Chronicon Anglicanuru,
p. 165 : '* Rex Eustachium de Vesci et Robertum fiHum Walteri, in comitatibus
tertio requisitos, cum eorum fautoiibus utlaghiari fecit, castra eorum subvertit,.
praedia occupavit."
26 EVENTS LEADING TO MAGNA CARTA
himself. In the spring of the same year, he had still
seemed to enjoy the full tide of prosperity ; and he must
have been a bold prophet who dared, like Peter of Wake-
field, to foretell the speedy downfall of the King.^
John's apparent security was deceptive; he had under-
estimated the powers arrayed against him. In January,
1 2 13, by Innocent's command, formal sentence of excom-
munication was passed on John, and Philip of France was
appointed as its executor. The chance had come for which
the barons, particularly the eager spirits of the North, had
long been waiting. The King, on his part, realised that
the time had arrived to make his peace with Rome.
On 13th May, 12 13, John met Pandulf, the papal legate,
and accepted unconditionally the same demands which
he had refused contemptuously some months before. Full
reparation was to be made to the Church. Stephen
Langton was to be received as archbishop in all honour
with his banished bishops, friends and kinsmen. All
church property was to be restored, with compensation
for damage done. One of the minor conditions of John's
absolution was the restoration to Eustace de Vesci and
Robert Fitz- Walter of the estates which, they persuaded
Innocent, had been forfeited because of their loyalty to
Rome.2
Two days later, apparently on his own initiative, he
resigned the Crowns of England and Ireland, and received
them again as the Pope's feudatory, promising to perform
personal homage should occasion allow. John hoped thus
to be free to avenge himself on his baronial enemies. The
surrender was embodied in a formal document which bears
to be made by John, " with the common council of our
barons." Were these merely words of form? They may
have been so when first used ; yet two years later the envoys
of the barons claimed at Rome that the credit (so they now
represented it) for the whole transaction lay with them.
In any case, no protest seems to have been raised at the
time of the surrender. This step, so repugnant to later
^See Miss Norgate, yi?/^« Lackland^ 170, and authorities there cited.
"^Jbid., 292-3.
RICHARD I. AND JOHN 27
writers, seems not to have been regarded by contemporaries
as a disgrace. Matthew Paris, indeed, writing in the next
generation, describes it as " a thing to be detested for all
time " ; but events had ripened in Matthew's day, and he
was a keen politician rather than an impartial onlooker.^
Stephen Langton, now assured of a welcome to the high
office into which he had been thrust against John's will,
landed at Dover and was received by the King at Win-
chester on 20th July, 1 2 13. John swore on the Gospels to
cherish and defend Holy Church, to restore the good laws
of Edward, and to render to all men their rights, repeating
practically the words of the coronation oath. He agreed
further to make reparation of all property taken from the
Church or churchmen.
V. The Years of Crisis, 1213-15.
Once more the short-sighted character of John's abilities
was illustrated : a brief~triumph led to a~deeper fall. For
a season, however, after he had made his peace with Rome,
he seemed to enjoy substantial fruits of his diplomacy.
Philip's threatened invasion had to be abandoned; the
people renewed their allegiance on the removal of the papal
sentence; the barons had to make their peace as best
they could, awaiting a better opportunity to rebel. H
John had confined himself to home affairs, he might hav^^
postponed the final explosion : he could not, however,
reconcile himself to the loss of the continental heritage of
his ancestors. His attempts to recover Normandy and ""
Anjou led to new exactions and new murmurings, while
their complete failure left him, discredited and penniless,
at the mercy of the malcontents at home.
His projected campaign in Poitou required all the levies
he could raise. More than once John demanded, and his
barons refused, their feudal service. Many excuses were
put forward. At first they declined to follow a King who
had not yet been fully absolved. After 20th July, 12 13,
^For the complacency with which contemporary opinion viewed John's surrender,
see Petit Dutaillis, Louis VIII. p. 39. Cf. ibid. p. 181. See also Cardinal
Manning, Contevip. Rev., December, 1875 ; Adams, Origin Engl. Const., 152 n.
28 EVENTS LEADING TO MAGNA CARTA
their new plea was that the tenure on which they held their
lands did not compel them to serve abroad : they added
that they were already exhausted by expeditions within
England. 1 John took this as defiance, and determined,
with troops at his back (per vim et arma)y to compel
obedience. Before his preparations were completed, an
important assembly met at St. Albans on 4th August, to
make sworn inquest as to the extent of damage inflicted
on church property during John's quarrel with Rome.^
From this Council directions were issued in the King's
name commanding sheriffs, foresters, and others to observe
the laws of Henry I. and to abstain from unjust exactions,
as they valued their lives and limbs. ^
On 25th August, after John had set out with his mercen-
aries to punish his northern magnates, Stephen Langton
held a meeting with the great men of the south. Many
bishops, abbots, priors and deans, together with some lay
magnates of the southern counties, met him at St. Paul's,
London, ostensibly to determine what use the Archbishop
should make of his power to grant partial relaxation of the
interdict, still casting its blight over England. In the
King's absence, Stephen reminded the magnates that
John's absolution had been conditional on a promise of -
good governments^ He showed them Henry I.'s coronation
charter : " by which, if you desire, you can recall your
long lost liberties to their pristine state."* All present
swore to " fight for those liberties, if it were needful, even
unto death." The Archbishop promised his help, "and a
confederacy being thus made between them, the conference
was dissolved." 5
iR. Coggeshall, p. 167.
2 For the latest views on this council and the writs of summons, see Prof. A. B,
White, Am. Hist. Rev., XVII. 12-16.
»R. Wendover, III. 261-2.
^R. Wendover, III. 263-6. Blackstone (6^r^a/ Charter, Introduction, p. vi.),
makes the apposite comment that it seems unlikely that the discovery of a charter
probably already well known " should be a matter of such novelty and triumph."
^ R. Wendover, III. 263-6. Ramsay, Angevin Empire, 444, doubts the
authenticity of this meeting, the incidents of which have a suspicious resemblance to
what took place some fourteen months later at Bury St. Edmunds : see infra, p. 32.
THE YEARS OF CRISIS, 1213-15 29
Stephen Langton desired a peaceable solution. We find
him, accordingly, at Northampton, on the 28th of August,
striving to avert civil war. His line of argument is worthy
of note : the King must not levy war on his subjects before
he had obtained a legal judgment against them (absque
judicio curiae suae). These words should be compared
with the " unknown charter " ^ and with chapter 39 of
Magna Carta.
John continued his march to Nottingham, bidding the
archbishop not to meddle in aifairs of state ; but threats of
excommunication caused him to consent to substitute legal
process for violence, and to appoint a day for the trial of
defaulters before the Curia Regis — a trial which never took
place.2 John apparently continued his journey as far north
as Durham, but returned to meet the new papal legate
Nicholas, to whom he performed the promised homage and
repeated the act of surrender in St. Paul's on 3rd October. ^
Having completed his alliance with Rome, he was confident
of worsting his enemies in France and England.
Yet most, if not all, of the magnates were against him,
and this fact may possibly explain John's issue of writs,
on gth November, 12 13, inviting four discreet men of each
county to discuss with him affairs of the Kingdom.* This
has sometimes been interpreted as a deliberate design to
broaden the basis of the commune concilium by adding to
it representatives of classes other than Crown-tenants. &
Miss Norgate, indeed, lays stress on the fact that these writs
were issued after the death of the great Justiciar, Geoffrey
Fitz-Peter, and before any successor had been appointed.
John, she argues, acted on his own initiative, and is thus
entitled to the credit of being the first statesman to intro-
duce representatives of the counties into the national
assem.bly. Knights who were tenants of mesne lords (Miss
Norgate says " yeomen ") were invited to act as a counter-
^ See Appendix. " R. Wendover, III. 262-3.
^The charter recording this act may be read in Neru Ryvier^ I. 115. It was
sealed not in perishable wax, but in gold.
* Sel Chart. 287. ^John Lackland, 195.
30 EVENTS LEADING TO MAGNA CARTA
poise to the barons. This innovation is held to have
anticipated the line of progress afterwards followed by
de Montfort and Edward I. : compared with it, the often-
praised provisions of chapter 14 of Magna Carta are
regarded as antiquated and even reactionary.
Recent research and criticism, however, have tended to
throw doubts on the authenticity and purport of these writs,
and to postpone the introduction of the representative
principle into the central council to a considerably later
date. It would be unwise to build far-reaching inferences
on the supposed participation of county representatives in
the debates of November, 1213.^
In the early spring of 12 14, John considered his home
troubles ended, and that he was now free to use against
France the coalition formed by his diplomacy. He went
abroad early in February, leaving Peter de Roches, the
unpopular Bishop of Winchester, as Justiciar, to guard his
interests, in concert with the papal legate.^ Deserted by
the northern barons, John relied partly on his mercenaries,-
but chiefly on the Emperor Otto and his other powerful
allies. Fortune favoured him at first, only to ruin him
more completely in the end. On 2nd July, 1214, John had
hastily to abandon the siege of Roches au Moine, leaving
his baggage to the enemy. The final crash came on
Sunday, 27th July, when the King of France triumphed
over John's allies at the decisive battle of Bouvines. On
1 8th September, John was compelled to sign a five years*
truce with Philip, abandoning all pretensions to his
continental dominions.
He had left even more dangerous enemies at home, to
watch with trembling eagerness the vicissitudes of his
fortunes abroad. His earlier successes struck dismay into
the malcontents in England, apprehensive of the probable
sequel to his triumphant return home. They waited with
anxiety, but not in idleness, the culmination of his cam-
paign, wisely refraining from open rebellion until news
reached them of his failure or success. Meanwhile, they
quietly organized their programme of reform and their
^ See (f.^. Adams, Origin, 340-1. ^ See RoL Pat. I. no, no, <5.
THE YEARS OF CRISIS, 1213-15 31
measures of resistance. John's strenuous endeavours to
exact money and service, while faiHng to fill his Exchequer,
had ripened dormant hostility into an active confederacy
organized for resistance. The English barons felt that the
moment for action had arrived when news came of the
disaster at Bouvines.
Even while abroad, John had not relaxed his efforts
to wring exactions from England. Without consent or
warning, he had imposed a scutage at the unprecedented
rate of three marks on the knight's fee. Writs for its
collection had been issued on 26th May, 12 14, an exception
being indeed allowed for tenants personally present in the
King's army in Poitou. The northern barons, who had
already refused to serve in person, now refused likewise to
pay the scutage. This repudiation was couched in words,
peculiarly bold and sweeping ; they denied liability to
follow the King not merely to Poitou, but to any part of'
the Continent.^
. When John returned, vanquished and humiliated, on
T5th October, 12 14, he found himself confronted with a
crisis unique in English history. During his absence, the
opponents of his misrule had drawn together, formulated
their grievances, and matured their plans. The embar-
rassments on the Continent which weakened the King,
heartened the opposition. The northern barons took the
lead. Their cup of wrath, which had long been filling,
overflowed when the scutage of three marks was imposed.
Within three weeks of his landing, John held parley with
the malcontents at Bury St. Edmunds (on 4th November). 2
No compromise was possible : John pressed for payment,
and the barons refused.
It seems probable that, after John's retiral, a conference
of a more private nature was held at which, under cloak
of attending the Abbey for worship, a conspiracy against
John was sworn. Roger of Wendover gives a graphic
account : the magnates came together " as if for prayers ;
but there was something else in the matter, for after they
had held much secret discourse, there was brought forth
^ See W. Coventry, II. 217. ^See Norgate,yM« Lackland, p. 221.
32 EVENTS LEADING TO MAGNA CARTA
in their midst the charter of King Henry I., which the
same barons had received in London . . . from Arch-
bishop Stephen of Canterbury."! A solemn oath was
taken to withdraw their fealty (a threat carried into effect
on 5th May of the following year), and to wage war on the
King, unless he granted their liberties. A date — soon after
Christmas — was fixed for making their formal demands.
Meanwhile they separated to prepare for war. The King
luXalso realized that a resort to arms was imminent. While
collecting mercenaries, he tried to sow dissension among
his opponents : he hoped to buy off the hostility of the
Church by a charter, issued on 21st November, professing
to be granted "of the common consent of our barons."
Its object was to gain the Church's support by granting
freedom of election to vacant sees. The appointment of
prelates should henceforth really lie with the canons of the
various cathedral or conventual churches and monasteries,
saving, however, to the Crown the right of wardship during
vacancies. John promised never to deny or delay his
consent to an election, and conferred powers on the electors,
if he should do so, to proceed without him. The King
was bitterly disappointed in his hope that by this bribe
he would bring over the Church from the barons' side to
his own.
John held what must have been an anxious Christmas
at Worcester, but tarried only for a day, hastening to the
Temple, London, where the proximity of the Tower gave
him a feeling of security. There, on 6th January, 1215, a
, \ deputation from the insurgents met him without disguising
that their demands were backed by force. These demands,
they told him, included the confirmation of the laws of
r Edward, with the liberties set forth in Henry's Charter.
On the advice of the Archbishop and the Marshal, who
acted as mediators, John asked a truce till Easter, which
was granted on his promise that he would then give
reasonable satisfaction. The Archbishop, the Marshal,
and the Bishop of Ely were named as the King's sureties.
John was in desperate straits for money : " the pleas of
^ R. Wendover, III. 293. Cf. supra 28.
THE YEARS OF CRISIS, 1213-15 33
the exchequer and the counties ceased throughout England,
for nobody was found who would pay tax to the King, or
obey him in anything."^ On 15th January, he reissued
the Charter to the Chifrch, and demanded a renewal of
homage. The sheriffs in each county were instructed to
administer the oath in a stringent form ; all Englishmen
must now swear to " stand by him against all men." Mean-
while, emissaries were dispatched by both sides to Rome.
Eustace de Vesci, as spokesman of the malcontents, asked
Innocent, as overlord of England, to compel John to restore
the ancient liberties, and claimed consideration on the
ground that John's surrender to the Pope had been made
under pressure put on the King by them — all to no effect.
John thought to propitiate the Pope by swearing to go upon
Crusade, a politic oath which would serve to protect him
from personal violence, and which afforded him, as is well
illustrated by several chapters of Magna Carta, a fertile
excuse for delay in remedying abuses. In April, the
northern barons met in arms at Stamford, and after Easter\\)
(when the truce had expired) marched southward to Brack-
ley, in Northampton. There they were met, on 27th April,
by the Archbishop and the Marshal, as emissaries from
the King, to enquire as to their demands. They received
in reply, and took back with them to John, a certain
schedule, which, so Roger of Wendover informs us, con-
sisted for the most part of ancient laws and customs of the
realm, with an added threat that, if the King did not
immediately adhibit his seal, the rebels would constrain
him by seizing his castles, lands, and goods. 2
John's answer when he read these demands, was em-
phatic. " Why do not the barons, with these unjust
exactions, ask my kingdom ? " Then furious, he declared
with an oath that he would never grant them liberties which
would make him a slave. 3
A metrical chronicle * records the threat to depose the
King, unless he fully amended the law and furnished
* R. Wendover, III. 301.
2 R. Wendover, III. 298. For the schedule see infra^ pp. 37-9.
^ R. Wendover, III. 298. * Chronica de Mailrosy sub anno 1215.
%
34 EVENTS LEADING TO MAGNA CARTA
undoubted guarantees for a lasting peace. On 5th May,
the barons went through the ceremony of diffidatio, or
formal renunciation of allegiance, ^ a recognised feudal
right, and not involving treason if justified by events and
properly intimated to the overlord.2 They chose as their
commander, Robert Fitz- Walter, who, as though conduct-
ing a Crusade, styled himself piously and grandiloquently,
" Marshal of the army of God and Holy Church."
The insurgents, still shivering on the brink of civil war,
delayed to march southwards. Much would depend on the
attitude of London, with its wealth and central position;
and John bade high for the support of its citizens. On 9th
May a new charter^ was granted to the Londoners, who
now received a long-coveted privilege, the right to elect
their mayor annually and to remove him at the year's end.
This marked the culmination of a long series of progressive
grants in their favour. Previously the mayor had held
office for life, and Henry Fitz-Aylwin, the earliest holder
of the office (appointed perhaps in 1191), had died in 12 12.
Apparently no price was paid for this charter; but John
doubtless expected in return the grateful support of the
Londoners, exactly as he had expected the support of
churchmen when he twice granted a charter in their favour.
In both instances he was disappointed. Next day he made,
probably as a measure of delay, an offer of arbitration to
the barons. In the full tide of military preparations, he
issued a writ in these words : " Know that we have con-
ceded to our barons who are against us that we shall not
take or disseise them or their men, nor go against them
per vim vel per anna, unless by the law of our land, or by
the judgment of their peers in curia nostra, until considera-
tion shall have been had by four whom we shall choose on
our part and four whom they shall choose on their part,
and the lord Pope who shall be oversman over them " —
^ Blackstone, Great Charter, p. xiii, citing Annals of Dunstable (p. 43), says
they were absolved at Wallingford by a Canon of Durham.
^ Cf, Adams, Origin, 181 n. ; 306, 312 ; cf. also infra under c. 61.
^The Charter appears Rot. Chart., p. 207. Cf. under chapter 13 infra, where
the rights of the Londoners are discussed.
THE YEARS OF CRISIS, 1213-15 35
words worthy of careful comparison with cliapter 39 of
Magna Carta. The offer could not be taken seriously,
since it left the decision of every vital issue virtually to the
Pope, whom the barons distrusted.^
Another royal writ, of two days later, shows a rapid
change of policy, doubtless due to the contemptuous
rejection of arbitration. On 12th May, John ordered the
sheriffs to do precisely what he had offered not to do.
They were told to take violent measures against the rebels
without waiting for a " judgment of peers." Lands, goods,
and chattels of the King's enemies were to be seized and
applied to his benefit.^ The barons, rejecting all offers,
marched by Northampton, Bedford, and Ware, towards the
capital. London opened its gates on 17th May. 3 The
example was quickly followed by other towns and by many
hesitating magnates. The confederates felt strong enough
to issue letters to all who still adhered to John, bidding
them forsake him on pain of forfeiture.
John found himself, for the moment, without power of
effective resistance; and, probably with a view of gaining
time rather than of committing himself irretrievably to any
abatement of his prerogatives, agreed to a conference.
As a preliminary, he issued, on 8th June, a safe-conduct
for the barons' representatives to meet him at Staines within
the three days following. This was too short notice : on
loth June, John, now at Windsor, granted an extension
of the safe-conduct till Monday, 15th June. William
Marshal and other envoys were dispatched from Windsor
to the barons in London with a message of surrender :
John " would freely accede to the laws and liberties which
they asked," if they would appoint a place and day of
meeting. The intermediaries, in the words of Roger of
Wendover,* " without guile carried back to the barons the
1 The writ is given in Rot. Fat., I. 141, and also in N'ew Rynner, I. 128.
* For writ, see Rot. Claus.j 204.
^ Some authorities give 24th May, but Nerv Hymer, p. 121, under 17th May, prints
a writ of John, informing Rowland Blaot of the surrender of London. This was
followed on 20th May [^N. R., p. 121) by another writ, ordering bailiffs and other
to molest the Londoners in every possible way.
nn. 301.
W
36 EVENTS LEADING TO MAGNA CARTA
message which had been guilefully imposed on them."
The barons, immenso fluctuantes gaudio, fixed as the time
of meeting, the last day of the extended truce, Monday,
15th June, at a certain meadow between Staines and
Windsor, known as Runnymede.
VI. Runnymede, and after.
On 15th June, 1215, a five days' conference between
King and Barons began. On the side of the insurgents
appeared a great host; on the monarch's, a small band of
magnates, loyal to the person of the King, but only half-
hearted, at the best, in his support. Their names may be
read in the preamble to the Charter : the chief among them,
Stephen Langton, still nominally neutral, was known to be
in full sympathy with the rebels.
Dr. Stubbs,! maintaining that the whole baronage of
England was implicated in these stirring events, analyses
its more conspicuous members into four groups: (i) the
Northumhrani or Norenses of the chroniclers, the first to
raise the standard of revolt; (2) other barons from various
parts of England, who had shown themselves ready to
co-operate with the Northerners — " the great baronial
families that had been wise enough to cast away the feudal
aspirations of their forefathers, and the rising houses which
had sprung from the ministerial nobility " ; (3) the moderate
party, who followed the lead of London, including even the
King's half-brother (the Earl of Salisbury), the loyal
Marshal, Hubert de Burgh, and other Ministers of the
Crown, whose names may be read in the preamble to the
Charter; and (4) the tools of John's misgovernment, mostly -
men of foreign birth, tied to John by interest as well as'
loyalty, since their differences with the baronial leaders lay
too deep for reconciliation, a few of whom are branded by
name in Magna Carta as for ever incapable of holding
office. These men of desperate fortunes alone remained
whole-hearted on John's side when the crisis came. 2
1 ConsL Hist., I. 581-3.
'^ The names may be read in Stubbs, Ibid, ; and readers in search of biographical
knowledge are referred to Bemont, Chartes, 39-40. and for fuller, less reliable
information, to Thomson, Magna Charta, 270-322.
RUNNYMEDE AND AFTER 37
When the conference began, the fourth group was in
command of castle garrisons or of troops actually in the
field; the third group, a small one, was with John; the
first and second groups were, in their imposing strength,
arrayed against him.
Unfortunately, the vagueness of contemporary accounts
prevents us from reproducing with certainty the progress
of negotiations on that eventful 15th of June and the few
following days. Some inferences, however, may be drawn
from the words of the completed Charter and of several
closely related documents. One of these, the Articles of
the Barons,! is sometimes supposed to be identical in its
terms with the schedule which had been already presented
to the King's emissaries at Brackley, on 27th April. It is
more probable that during seven eventful weeks the original
demands had been somewhat modified. The schedule of
April was probably only a rough outline of the Articles
as we now know them, and these formed in turn the draft
on which the Charter was based. Articles and Charter are
alike authenticated by the impress of the King's seal.
There is thus a strong presumption that an interval elapsed
between the King's acceptance of the first and the com-
pletion of the second ; since it would have been absurd to
seal a superseded draft at the same time as the principal
instrument. The probability of such an interval must not
be lost sight of in any attempt to reconstruct the stages of
negotiations at Runnymede.
A few undoubted facts form a starting-point on which
inferences may be based. John's headquarters were at
Windsor from Monday, 15th June, to the afternoon of
Tuesday the 23rd. On each of these nine days (with the
possible exception of the i6th and 17th) he visited Runny-
mede to confer with the barons. 2 Two crucial stages were
reached on Monday the 15th (the date borne by Magna
Carta itself) and on Friday the 19th (the day on which John
* See Appendix.
^So far there can be no doubt. Either on Close or Patent Rolls (q.v.) copies of
writs are preserved dated from Windsor on each of these days, and also one or
more dated from Runnymede on 15th, i8th, 19th, 20th, 21st, 22nd, and 23rd June.
38 EVENTS LEADING TO MAGNA CARTA
in more than one writ stated that peace had been concluded).
What happened exactly on each of these two days is matter
of conjecture. It is here maintained, with some confidence,
[that on Monday the substance of the barons' demands was
[provisionally accepted and that the Articles were then
iealed; while on Friday this arrangement was confirmed
md Magna Carta itself, in several duplicates, was sealed.
To justify these inferences, a more detailed examination
'of the evidence available is required. The earliest meeting
between John and the baronial leaders, all authorities are
agreed, took place on Monday, 15th June, probably in the
early morning. The barons undoubtedly brought to the
conference a list of grievances they were determined to
redress. On the previous 27th of April the rebels had sent
a written schedule to the King ; ^ they are not likely to
have been less fully prepared on 15th June.
John, on his part, would naturally try a policy of evasions
and delays ; and, when these were clearly useless, would
then endeavour to secure modifications of the terms offered.
These tactics met with no success. His opponents asked
a plain acceptance of their plainly expressed demands.
Before nightfall, John, overawed by their firmness and by
the numbers of the armed force behind them, was con-
strained to surrender, and signified his acceptance of the
barons' demands, as contained in a list of 49 Articles
(apparently drawn out on the spot), by imprinting his great
seal on the w^ax of its label, where it may still be seen.^
Ralph of Coggeshall's brief account gives the contem-
porary opinion : " By intervention of the archbishop of
Canterbury, with several of his fellow-bishops and some
barons, a sort of peace was made." ^ The document bears
traces of the discussions that preceded it. The first article
postpones a definition of the customary "relief," leaving
this to be expressed "in carta." * Articles 45 and 46 (less
vital to the barons as affecting their allies, not themselves)
are joined by a rude bracket ; and their suggested modifica-
1 R. Wendover, III. 298.
^ In the British Museum. See in/ra under Part V.
^R. Coggeshall, 172. "* See ////ra, c. 2
RUNNYMEDE AND AFTER 39
tion in favour of John is referred to Stephen Langton's
decision.^ The last article, or forma securitatis, the dregs
of John's cup of humiliation, is separated by a blank space
from the rest. 2
The document is in a running hand and appears to have
been rapidly though carefully written : a diligent copyist
would be able to complete his task within a few hours.
There are thus ample reasons for holding that it was not
the identical schedule of the preceding April, but that it
was written out between two conferences on Monday, 15th
June, by one of the clerks of the royal Chancery. This is
in keeping with the contemporary heading : " Ista sunt
capitula quae harones petunt et dominus rex concedit."
Comparison with the final Charter suggests that further
conferences led to alterations in regard to various details : ^
thus, chapter 14 contains provisions not contained in the
Articuli, though forming a necessary supplement to the
substance of article 32. New influences would seem to have
been at work, favourable to the claims of the English
Church ; effecting some slight modifications in favour of
the Crown ; * and apparently not too careful of the interests
of the towns or of native traders. ^
It is not difficult to infer the nature of the forces at work.*^\
John was fighting for his own hand ; the barons merely
demanded a fair statement of their just rights, and had no-V
desire to take undue advantage of the King; the towns
found the barons more ready to meet the King by sacri-
ficing their allies' rights than their own ; Stephen Langton,
while acting as mediator, looked well after the interests of
the Church.
Tuesday, Wednesday and Thursday were probably con-
sumed in adjusting these matters of detail ; in reducing the
^ See infra, cc. 58 and 59. Cf. Blackstone, Great Charter, xvii. : ** subjoined in
a more hasty hand, ... as if added at the instance of the King's commissioners
upon more mature deliberation."
^See infra, c. 61.
^Blackstone, Great Charter, xviii., has given a careful analysis of the points of
<lifference.
^E.g. chapters 48 and 52 infra. ^ E.g. chapters 12, 13, 35, and 4I itifra.
40 EVENTS LEADING TO MAGNA CARTA
heads of agreement to the more binding form of a feudal
Charter; and in engrossing several copies for greater
security. Everything was ready for settlement on Friday^
the 19th. On that day, the final concord probably included
several steps; the nomination by the opposition, with the
King's acquiescence, of twenty-five barons to act as
"Executors" under chapter 61,^ the solemn sealing and
delivery of several originals of the Charter in its final form,
the taking of an oath by all parties to abide by its provi-
sions, and the issue of the first batch of writs of instructions
to the sheriffs.
The barons on that day renewed their oaths of fealty and
homage: this was the stipulated price of "the liberties."
They promised a guarantee in any form John wished,
except the delivery of hostages or the surrender of strong-
holds— a promise they failed to keep.^
The statement that Friday, 19th June, was the day on
which peace was finally concluded rests on unmistakable
evidence. On 21st June, John wrote from Windsor to
William of Cantilupe, one of his captains, instructing him
not to enforce payment of any unpaid balances of " ten-
series " 3 demanded since the preceding Friday, " on which
day peace was made between the King and his barons."*
It has been usually assumed that peace was concluded,
and the Charter sealed on the 15th. The fact that all four
copies of Magna Carta still extant bear this date seems to
have been regarded as conclusive. Elaborate charters,
however, which occupied time in preparation, usually bore
the date, not of their actual execution, but of the day on
which occurred the transactions they record. Thus it is
^ The powers and constitutional position of these " executors " are fully discussed
infra under c. 6l.
' See Protest in Appendix.
'Round explains this {Geoffrey cie Mandeville, 414) as "blackmail," i.e^
*' money extorted under pretence of protection or defence."
^See Rot. Clans. ^ p, 225. This writ does not stand alone. In another writ,,
dated 19th June, John informs his half-brother that he has just concluded peace.
See also Annals of Dunstable, III. 43, reporting peace made ^^ die Gervassi et
Protasii" i.e. on 19th June.
RUNNYMEDE AND AFTER 41
far from safe to infer from Magna Carta's mention of its
own date that the seal was actually adhibited on 15th June.
Such presumption as exists is all the other way. The
Great Charter is a lengthy document, and it is barely
possible that any one of the four originals known to us
could have been engrossed (to say nothing of the adjust-
ment of substance and form) within one day. Not only is
it much longer than the Articles on which it is founded;
but even the most casual comparison will convince any
unbiassed mind of the slower rate of engrossment of the
Charter. All four copies show marks of deliberation,
while those at Lincoln and Salisbury are models of leisurely
and exquisite penmanship. The highly finished initial
letters of the first line and other ornamental features may
be instructively compared with the plain, business-like,
rapid hand of the Articles. How many additional copies,
now lost, were once in existence bearing the same date, it
is impossible to say ; but each of those still extant may well
have occupied more than one day in the writing. ^
In addition to the various originals of the Charter issued
under the great seal, chapter 62 provides that authenticated
copies should be made and certified as correct by " Letters
Testimonial," under the seals of the two archbishops with
the legate and the bishops.^ These were intended for the
* Miss Norgate, John Lackland^ p. 234, in fixing on Monday as the day of final
concord, relies for evidence on a more than doubtful interpretation of an error in
the copy of a writ, which in the Patent Rolls bears to be dated i8th June (errone-
ously as will be shown), addressed to Stephen Harengod, announcing that terms
of peace had been agreed upon "last Friday." Miss Norgate contends that on
the Friday preceding the i8th negotiations had not even begun, and is confident
that the " die Veneris" which occurs three times in the writ is an unaccountable
error for " die Lunae," Yet, it is unlikely that a scribe writing three days after so
momentous an event could have mistaken the day of the week. It is infinitely
more probable that is writing xxiij. he formed the second " x " so carelessly that it
was mistaken by the enrolling clerk for a *' v." The correct date is thus the
23rd, and the reference is to Friday the 19th. This presumption becomes a
certainty by comparison with the words of the writ to William of Cantilupe,
dated the 21st, and other evidences cited stipra^ p. 40.
^ No specimen of these Letters is known, but a copy is preserved on folio 234,
Red Book of Exchequer. See infra under c. 62 and also R. L. Poole, Eng. Hist.
Rev., XXVIII. 448, .
42 EVENTS LEADING TO MAGNA CARTA
sheriffs, whose writs of instructions dated 19th to 27th June,
to publish the terms of the charters, are preserved in the
Patent Rolls. Each sheriff vv^as instructed to cause all in
his bailiwick to make oath, according to the form of the
Charter, to the twenty-five barons or their attorneys, and
further, to see to the appointment of twelve knights of the
county in full County Court, to declare upon oath all evil
practices as well of sheriffs as of their servants, foresters,
and others.^ This was held to apply chiefly to the redress
of forest grievances.
A week elapsed before these writs, with copies of the
Charter, could be sent to every sheriff. During the same
few days, orders were sent to military commanders to stop
hostilities. A few writs, dated mostly 25th June, show that
some obnoxious sheriffs had made way for better men ;
while Hubert de Burgh became Justiciar in room of Peter
des Roches. On 27th June, new writs directed the sheriffs
and the elected knights to punish, by forfeiture of lands and
chattels, all who refused to swear to the twenty-five Execu-
tors within a fortnight.
The barons were still unsatisfied as to the King's sin-
cerity, and demanded further securities. The interesting
question thus arises, how far they were justified in doubting
John's intentions. Prof. Petit-Dutaillis, founding mainly
on the writs dispatched to sheriffs and constables, credits
John with perfect though perhaps short-lived good faith. 2
He rightly refuses to believe Wendover's unlikely story of
John's immediate retiral to the Isle of Wight, and of the
war preparations he made there in a delirium of fury.^
Proof of John's sincerity is sought in the reputed quarrel
with his Flemish mercenaries, for whom the King's " villain
^ See Appendix.
2 He might here have strengthened his argument by referring to the evidences of
extreme care shown in revising the original Articles of the Barons when translating
them into charter form. This would have been thrown away, if John intended to
break faith. On the other hand, this care, equally with the issue of writs, might
have been a blind.
' See Louis VIII., p. 57, and also Hardy's Introd. to Litt, Pat.^ XXIX., where
the story was disproved by dates of writs issued elsewhere.
RUKNYMEDE AND AFTER 43
peace " meant that his purse would be closed to them and
led them to desert his cause. ^
In brief, according to M. Petit-Dutaillis, John's conduct
was above reproach during June and July, and until the
bad faith of his opponents forced him to protect himself.^
Yet John's punctilious observance, for a short space, of
the letter of his bargain may be equally consistent with
studied duplicity, dictated by urgent need of gaining time,
as with any loyal intention to submit permanently to
restraints which, in his own words, " made him a slave,"
and were to be enforced by " five-and-twenty over kings " ;^
while his negotiations with Rome are difficult to reconcile
with any intention of permanently keeping faith.
Justified or not, the barons demanded that the City and
Tower of London should be placed in their hands as pledges
of good-faith until 15th August, or until the reforms were
completely carried out. John had to surrender the city to
the rebels, but the Tower was placed in the neutral custody
of Stephen Langton. These terms may be read in a
supplementary treaty headed : " Conventio facta inter
Regem Angliae et baro7ies ejusdem regniJ^ * John, equally
distrustful on his side, demanded the security promised at
the renewal of allegiance; but the barons refused to em-
body the terms of their homage in a formal Charter. The
Archbishops of Canterbury and Dublin, with several
suffragans, appealed to as umpires by the King, recorded
a protest narrating the barons' breach of faith. ^
The same prelates, alarmed apparently lest drastic
measures of reform should lead to the total abolition of the
forests, entered a second protest. As mediators, bound to
see fair-play, they declared in writing that the words of the
Charter must be read in a restricted sense : customs need-
ful for preserving the forests should remain in force. ^ The
provisions referred to were, as is now well known, chapters
^See Hint, des dues de Norm.y pp. 149- 151. ^ Louis VIII. ^ p. 57.
^See Norgate, Lackland^ 235, citing M. Paris, II. 61 1.
* New Rymer, I. 133. See Appendix. It is undated, but must be later than the
letters of 27ih June to which it alludes.
'"Rot. Pat., 181. See Appendix. ^ See Rot. Pat. and Ntw Rymer, I. 134.
44 EVENTS LEADING TO MAGNA CARTA
47, 48, and 53 of Magna Carta itself, and not, as Roger of
Wendover states, a separate Poorest Charter. ^ That writer
was led into error by confusing John's Charter with its
reissue by his son. Sir William Blackstone was the first
commentator to correct this mistake. ^
These are not the only pieces of evidence that point to
lack of moderation on the barons' part, revealed even before
the four days' conference was ended. Matthew Paris
narrates how it was found necessary to curb the excesses of
the twenty-five Executors of the Charter by the nomination
of a second body of thirty-eight barons, drawn from both
parties.^
From a contemporary chronicler there comes a strange
tale of the arrogance of the twenty-five : one day when they
went to the King's court "to make a judgment," John, ill
in bed, asked them to come to his chamber as he was unable
to go to them ; but they curtly refused, demanding that the
King, unable to walk, should be carried into their presence.*
John looked for aid to Rome. Three weeks before grant-
ing the Charter, he had begun his preparations for its
repudiation. In a letter of 29th May, addressed to the
Pope, there may still be read his own explanation of the
causes of quarrel, and how he urged, with low cunning,
that the rebels prevented fulfilment of his vow of crusade.
In conclusion, he expressed his willingness to abide by the
Pope's decision on all matters at issue. He followed up
this letter, shortly after 19th June, by dispatching Richard
de Marais to plead his cause at Rome. 5 Delay was doubly
in his favour ; since the combination formed against him
was certain, in a short time, to break up. It was, in the
happy phrase of Dr. Stubbs,^ a mere "coalition," not an
" organic union " — a coalition, too, in momentary danger of
dissolving into its original factors. The barons were with-
out sufficient sinews of war to carry a protracted struggle
to a successful issue.
Soon, both sides to the treaty of peace were preparing
*See R. Wendover, III. 302-318. 2 Q^eat Charter, p. xxi.
' M. Paris, 11. 605-6. ^ Hist, des dues de Normandie, 151.
^NewRynier, I. 129. "Stubbs, Const. Hist., II. 3.
£>
RUNNYMEDE AND AFTER 45
for war. The northern barons, anticipating the King in
direct breach of the compact, began to fortify their castles,
and maltreated the royal officials. 1 John, in equally bad
faith, wrote for foreign allies, whilst he anxiously awaited
the Pope's answer to his appeal. Langton and the bishops
still struggled to restore harmony. The i6th July was
fixed for a new conference. John did not attend; but it
was probably at this Council that in his absence a papal
bull was read conferring upon a commission of three — the
Bishop of Winchester, the Abbot of Reading, and the
legate Pandulf — full powers to excommunicate all " dis-
turbers of the King and Kingdom." No names were
mentioned, but these powers might clearly be used against
Langton and his friends. The execution of this sentence
was delayed, in the groundless hope of a compromise, till
the middle of September, when two of the commissioners,
Pandulf and Peter of Winchester, demanded that the arch-
bishop should publish it; and, on his refusal, they forth-
with suspended him from office (a sentence confirmed by
the Pope on 4th November) .2
Stephen left for Rome, and his absence at a critical junc-
ture proved a national misfortune. The insurgents lost in ^^
him, not only their bond of union, but also a wholesome
restraint. After his departure, a papal bull arrived (in the
end of September) dated 24th August. This is an impor- «>j
tant document in which Innocent, in the plainest terms, ^^
annuls and abrogates the Charter, after adopting all the '^**
facts and reproducing all the arguments furnished by the
King. Beginning with a full description of John's wicked-
ness and repentance, his surrender of England and Ireland,
his Crusader's oath, his quarrel with the barons; it goes on
to describe Magna Carta as the result of a conspiracy, and
concludes, " We utterly reprobate and condemn any agree-
ment of this kind, forbidding, under ban of our anathema,
the foresaid king to presume to observe it, and the barons
and their accomplices to exact its performance, declaring
void and entirely abolishing both the Charter itself and the
obligations and safeguards made, either for its enforcement
1 Walter of Coventry, 222. 2 gee Petit-Dutaillis, Louis VIIL, 61.
46 EVENTS LEADING TO MAGNA CARTA
or in accordance with it, so that they shall have no validity
at any time whatsoever." ^
A supplementary bull, of one day's later date, reminded
the barons that the suzerainty of England belonged to
Rome, and that therefore nothing could be done in tiie^
kingdom without papal consent.^ Thereafter, at a Lateran
Council, Innocent excommunicated the English barons who
had persecuted " John, King of England, crusader and
vassal of the Church of Rome, by endeavouring to take
from him his kingdom, a fief of the Holy See." ^
Meanwhile, the points in dispute had been submitted to
the rude arbitrament of civil war, in which the first notable
success fell to John, who took Rochester Castle by assault
on 30th November. The barons had already made over-
tures to Louis, the French King's son, offering him the
crown of England. Towards the end of November, seven
thousand French troops arrived in London, where they
spent the winter, while John marched from place to place,
meeting, on the whole, with success, especially in the east
of England. John's best ally was once more the Pope,
who did not intend to allow a French Prince to usurp his
vassal's throne. Gualo was dispatched from Rome to
Philip, King of France, forbidding his son's invasion, and
asking protection and assistance for John. Philip, anxious
to break the force of the Pope's arguments by proving some
right to intervene, endeavoured to find defects in John's
title as King of England, and to argue that therefore John
was not in titulo to grant to the Pope the rights of an over-
lord; John had been convicted of treason while Richard
was King, and this involved forfeiture of all rights of suc-
cession. Thus the Pope's claim of intervention was invalid,
while Prince Louis justified his own interference by some
imagined right which, he ingeniously argued, had passed
to him through the mother of his wife.
John had not relied solely on papal protection ; but the
* The bull with the seal attached is in the British Museum (Cotton, Cleopatra E i),
and is carefully printed by Bemont, Chartes, 41. It may also be read in Rymer
and Blackstone.
*The text is given by Rymer. ^See Rymer, and Bemont, Charles, XXV.
RUNNYMEDE AND AFTER 47
fleet, collected at Dover to block Louis with his smaller
vessels in Calais harbour, was wrecked on i8th May, 12 16.
The French Prince, setting sail on the night of the 20th i)
May, landed next morning unopposed. John, reduced to
dependence on mercenaries, dared not risk an engagement.
Gualo, now in England, on 28th May excommunicated
Louis by name, and laid London under interdict. On 2nd
June, the French Prince entered London, received homage
from the Mayor and others, and took oath to uphold good
laws and restore invaded rights. ^ It was probably on this
occasion that Louis confirmed the Charter. ^ Into the
vicissitudes of the war and the royalist reaction, to which
the arrogance of the French troops contributed, it is un-
necessary here to enter. At a critical juncture, when
fortune still trembled in the balance, John's death at Newark
Castle, on the morning of 19th October, 12 16, altered the
situation, rendering possible, and indeed inevitable, a new
arrangement of parties and forces in England. The heir to f\
the throne was an infant, whose advisers found it prudent to v.>
reissue voluntarily, and to accept as their rule of govern-
ment, the essential principles of the Charter that had been
extorted from the unwilling John.
1 Cronique de Merton, cited Petit-Dutaillis, Lotiis VI II., 514.
^Ibid., 115.
PART 11.
FEUDAL GRIEVANCES AND MAGNA CARTA.
I. The Immediate Causes of the Crisis.
Many attempts have been made to show why the storm,
long brewing, broke at last in 12 14, and culminated pre-
cisely in June of the following year. Sir William Black-
stone ^ shows how carefully historians have sought for some
one specific feature or event, occurring in these years, of
such moment as by itself to account for the rebellion
crowned with success at Runnymede. Matthew Paris, he
tells us, attributes the whole movement to the sudden
discovery of Henry I.'s Charter, and most of the chroniclers
assign John's inordinate debauchery as the cause of the
dissensions, dwelling on his personal misdeeds, real
and imaginary .2 " Sordida foedatur foedante Johanne,
gehenna." ^ Blackstone himself suggests a third cause, the
appointment as Regent in John's absence of the hated alien
and upstart, Peter des Roches, and his misconduct in that
office.
Of John's arrogance and cruelty there is abundant
testimony ; ^ but the causes from which Magna Carta took
its rise were more deeply rooted in the past. The very
success of Henry Plantagenet in restoring order in Eng-
1 The Great Charter, p. vii. 2 r, Wendover, II. 535.
' M. Paris, II. 669. Several of the most often-repeated charges of personal
wrongs inflicted by King John upon the wives and daughters of his barons have
been in recent years refuted. See Miss Norgate, y<?/^« Lackland, p. 289.
* See, e.g., the harrowing account of how he starved to death Matilda de Braose
and her son (Davis, Engl, under Normans, 363). For his conduct in Ireland, see
Orpen, Ireland, II. 96-105 ; and in Normandy, Powicke, Loss of Normandy, 190-2.
THE IMMEDIATE CAUSES OF THE CRISIS 49
land, for effecting which special powers had been allowed
to him, made the continuance of these powers unnecessary.
From the day of Henry's death, if not earlier, forces were
at work which only required to be combined in order to
control the licence of the Crown. When the battle of order ..
had^ been won — the complete oyerthrpw of tff^^ rebellion of l^\
n 73-4 may here be taken as the crucial date — the battle of 5f
liberty had, almost necessarily, to be begun.
The wonder is that the crisis was so long delayed.
Events, however, were not ripe for rebellion before John's
accession, and a favourable occasion did not occur previous
to 1 2 15. The doctrine of momentum accounts in politics
for the long continuance of old institutions in a condition
even of unstable equilibrium ; an entirely rotten system of
government may remain for ages until at the destined
moment comes the final shock. John conferred a boon on
future generations, when by his arrogance and his mis- (V)
fortunes he combined against him all classes and interests
in the community.
The chief factor in the coalition that ultimately triumphed
over John was the baronial party, led by those strenuous
nobles of the north, who were goaded into opposition by
their own personal and class wrongs, not by any altruistic
promptings to sacrifice themselves for the common good.
Their complaints, as they appear in the imperishable record
of Magna Carta, are grounded on technical rules of feudal
usage, not upon any broad basis of constitutional principle. Jj
TJne grievances most bitterly resented may be ranged*^
under one or other of two heads-^increase in the weight qf _. 1 ^
feudal obligations and mfringement of feudal jurisdictions j *-
the Crown, while it exacted the fullest measure of services
legally exigible, curtailed those rights and privileges which
had originally balanced the obligations. The barons were
compelled to give more, while they received less. Each of
these heads calls for separate and detailed treatment.^
The grievances of the barons, however, were not the only
wrongs calling for redress. It is probable that the baronial
party, if they had acted in isolation, would have failed in
^ See infra the two sections (II. and III.) immediately following.
D
w
50 FEUDAL GRIEVANCES AND MAGNA CARTA
1 2 15 as they had already failed in 11 73. If the Crown had
retained the active sympathy of Church and common
people, the King might have successfully defied the baron-
age as his father had done before him. John_h^, on_the
contrary, br^oadenedjhe basis of^ op_g^osiiiGn_byL_o^pre^^^^^^
Ithejnercandle classes and the peasantjy . The order-loving
itownsmerT had been willing to purchase protection from
; Henry at the price of heavy taxation : John continued to
exact the price, but failed to furnish good government in
return. Far from protecting the humble from oppression,
he was himself the chief oppressor; and he let loose his
foreign favourites as deputy oppressors in all the numerous
offices of sheriff, castellan, and bailiff. Far from using
theperf ected jiiachinery ., of Exchequer, Curia, and local
admiHistrRtion in the interests of good government, John
valued them merely as instruments of extortion and outrage
— as^ ministers to his lust and greed.
The lower orders were by no means exempt from the
increased taxation which proved so galling to the feudal
tenants .yy When John, durifig his quarrel with Rome,
repaid each new anathema of the Pope by fresh acts of
spoliation against the English Church, the sufferings of
the clergy were shared by the poor. In confiscating the
goods of monasteries, he destroyed the chief provision for
poor-relief known to the thirteenth century. The alienation
of the affections of the great masses of lower-class English-
men thus effected was never wholly undone, even after the
reconciliation of John with the Holy See. Notwithstanding
the completeness and even abjectness of John's surrender,
he took no special pains to reinstate himself in the good
graces of the Church at home. Innocent, secure at the
Lateran, had issued his thunderbolts; and John's counter-
strokes had fallen, not on him, but on the English clergy.
The measures taken, in 12 13 and afterwards, to make good
to these victims some part of the heavy losses sustained,
were inadequate.
After 1 2 13, John's alliance with Rome brought new
dangers in its train. The united action of two autocrats,
each claiming supreme powers, lay and spiritual respec-
THE IMMEDIATE CAUSES OF THE CRISIS 51
tively, threatened to annihilate the freedom of the EngUsh
nation and the English Church. " The country saw that
the submission of John to Innocent placed its liberty,
temporally and spiritually, at his mercy ; and immediately
demanded safeguards."^
This union of tyrants led to another union which check-
mated it, for the baronial opposition allied itself with the
ecclesiastical opposition. The urgency of their common
need brought prelates and barons into line — for the
moment. A leader was found in Stephen Langton, who
succeeded in preventing the somewhat divergent interests
of the two estates from splitting them asunder.
All things were thus ripe for rebellion, and even for
united rebellion ; an opportunity only was required. Such
an opportunity came in a tempting form in 1214; for the.
King had then lost prestige and power by his failure inK)
the wars with France. He had lost the friendship of the
English Church. His unpopularity and vacillating nature
had been thoroughly demonstrated. Further, he had him-
self, in 1 191, when plotting against his absent brother
Richard, successfully ousted the Regent Longchamp from
office, thus furnishing an example of successfully concerted
action against the central government.
The result was that, when the barons began active opera-
tions, not only had they no opposition to dread from church-
man or merchant, from yeoman or peasant, but they might
count on the sympathy of all and the active co-operation of
many. Further, John's policy of misrule had combined
against him two interests usually opposed to each other,
the party of progress and the party of reaction. The
influence of each of these may be clearly read in various
chapters of Magna Carta.
The progressive party consisted mainly of the heads of
the more recently created baronial houses, men trained in
the administrative methods of Henry II., who desired that
his system of government should be properly enforced.
They demanded that the King should conduct the business
of Exchequer and Curia according to the rules laid down by
1 Stubbs, Sg/ea Charters, 270.
52 FEUDAL GRIEVANCES AND MAGNA CARTA
I Henry. Routine and order under the new system were
Iwhat this party desired, and not a return to the unruly days
of Stephen. Many of the innovations of the great Angevin
had now been loyally accepted by all classes of the nation ;
and these accordingly found a permanent resting-place in.
the provisions of the Charter. In temporary co-operation
with this party, the usually rival party of reaction was
willing to combine for the moment against the common
enemy. There still existed, in John's reign, magnates of
the old feudal school, who hoped to wrest from the King's
weakened hand some measure of feudal independence.
They had accepted such reforms as suited them, but still
bitterly opposed many others. In particular, they resisted
the encroachments of the royal courts of law which were
gradually superseding their private jurisdictions. For the
moment, John's crafty policy, so well devised to gain
immediate ends, and so unwise in the light of subsequent
history, combined these two streams, usually ready to
thwart each other, into a united opposition to his throne.
Attacked at the same moment by the votaries of traditional
usage and by the votaries of reform, by the barons, the
trading classes, and the clergy, he had no course left him
but to surrender at discretion. The movement which cul-
minated at Runnymede may thus best be understood as
the resultant of a number of different but converging
forces, some of which were progressive and some re-
actionary.
II. The Crown and Feudal Obligations.
Among the evils calling loudly for redress in England at
the commencement of the thirteenth century, none spoke
with more insistent voice than those connected with feudal
abuses. The refusal of the northern barons to pay the
scutage demanded on 26th May, 12 14, was the spark that
fired the mine. The most prominent feature of the Charter
is its solicitude to define the exact extent of feudal services
and dues, and so to prevent these from being arbitrarily
j increased. A detailed knowledge of feudal obligations
forms a necessary preliminary to the study of Magna Carta.
THE CROWl^ „.,^ xEUDAL OBLIGATIONS 53
The precise relations of the Norman Conquest to the
growth of feudalism in England are complicated, and have
formed the subject of much controversy. The view now
generally accepted, and with reason, is that the policy of
the Conqueror accelerated the process in one direction, but
retarded it in another. Feudalism, regarded as a system
of government, had its worst tendencies checked by the
great upheaval that followed the coming of Duke William ;
feudalism, considered as a system of land tenure, and
as a social system, was, on the contrary, formulated and
developed. It is mainly as a system of land tenure that it
falls here to be considered. Originally, the relationship,
between lord and tenant, dependent upon the double owner- 1
ship of land (of which each was, in a different sense, i
proprietor), implied obligations on both sides : the lord!
gave protection, while the tenant owed services of various!
sorts. It so happened, however, that, with the changes
wrought by time, the legal obligations of the lord ceased to
be of much importance, while those of the vassal became
more and more burdensome. The tenant's services varied
in kind and in extent with the nature of the tenure. It is
difficult to frame an exact list of the various tenures formerly
recognized as distinct in English law : partly because the
classical authors of different epochs, from Bracton to Black-
stone, contradict each other; and partly because of the
obscurity of the process by which these tenures were
gradually differentiated. Sir William Blackstone,^ after
explaining the dependent nature of all real property in
England, thus proceeds : " The thing holden is therefore
styled a tenement, the possessors thereof tenants, and the ^
manner of their possession a tenure." Tenure thus comes
to mean the conditions on which a tenantliblds real estate -
unSer his lord^
The ancient classification differs materially from that in
use at the present day. The modern English lawyer (unless
of an antiquarian turn of mind) concerns himself only with
three tenures : freehold (now practically identical with
socage), copyhold and leasehold. The two last-mentioned
^ Commentaries, II. 59.
54 FEUDAL GRIEVANCES AND MAGNA CARTA
may be rapidly dismissed, as they were of little importance
in the eyes of Littleton, or of Coke : leasehold embraces
only temporary interests, such as those of a tenant-at-will
or for a limited term of years ; while copyhold is the modern
form of tenure into w^hich the old unfree villeinage has
slowly ripened. The ancient writers were, on the contrary,
chiefly concerned with holdings both permanent and free.
Of these, seven at least may be distinguished in the thir-
teenth century, all of which have now come to be represented
by the modern freehold or socage. These seven are
knight's service, free socage, fee-farm, frankalmoin, grand
serjeanty, petty serjeanty, and burgage.
(i) Knight's Service. Medieval feudalism had many
aspects ; it was almost as essentially an engine of war as it
was a system of land-holding. The normal return for
which an estate was granted consisted of the service in
the field of a specific number of knights. Thus the normal
feudal tenure was known as knight's service, or tenure in
^hivalry — the conditions of which must be constantly kept
tn view, since by them the relations between John and his
recalcitrant vassals fell to be determined. When finally
abolished at the Restoration, there fell with knight's
service, it is not too much to say, the feudal system
of land tenure in England. " Tenure by barony "
is sometimes spoken of as a separate species, but
may be more correctly viewed as a variety of tenure in
chivalry.^
(2) Free Socage. The early history of socage, with its
division into ordinary and privileged, is involved in obscuri-
ties which do not require to be here unravelled. The
I services returned for both varieties were not military but
' agricultural, and their exact nature and amount varied
considerably. Although not so honourable as chivalry,
free socage was less burdensome, in respect that two of the
most irksome of the feudal incidents, wardship and mar-
riage, did not apply. When knight's service was abolished
those who had previously held their lands by it, whether
of the Crown or of a mesne lord, were henceforward to hold
^See Pollock and Maitland, Hist. Engl. Law, ist ed., I. 218.
THE CROWN AND FEUDAL OBLIGATIONS 5^
in free socage, which thus came to be the normal holding;
throughout England after the Restoration. ^
(3) Fee-farm was the name applied to lands held in
return for services which were neither military nor agricul-
tural, but consisted only of an annual payment in money. /
The " farm " thus indicates the rent paid, which apparently
might vary without limit, although it was long maintained
that a fee-farm rent must amount at least to one quarter
of the annual value. This error seems to have been founded
on a misconstruction of the Statute of Gloucester.^ Some
authorities 3 reject the claims of fee-farm to rank as a tenure
separate from socage ; although chapter 37 of Magna Carta
seems to recognize the distinction.
(4) Frankalmoin was a favourite tenure with founders of
religious houses. It was also the tenure on which much
of the glebe lands of England was held by the village
priests. The grant was made in liberar}i eleemosinam or
"free alms" (that is, no temporal services were to be;
rendered).* In Scots charters the return formally stipulated :
was preces et lacrymae,
(5) Grand serjeanty was a highly honourable tenure,
sharing the distinctions and the burdensome incidents of
knight's service, but distinct in this, that the tenant, in; ;;/
place of ordinary military duties, performed some specific' ''^'''^<^?^
service, such as carrying the King's banner or lance, or
filled some important office at the coronation. ^ An often-
quoted example of a serjeanty is that of Sir John Dymoke
and his family, who have acted as the Sovereign's cham-
pions at successive coronations from Richard II. to William
IV., ready to defend the Monarch's title to the throne by
battle in the ancient form.
Grand serjeanties were_liable to wardship and marriage.
as well as to relief, but not to payrnent of scutage.^ William
^ See Statute 12 Charles II., c. 24. 2 See Pollock and Maitland, I. 274 n.
=^ Pollock and Maitland, I. 218.
* Littleton, II. viii. s. 133. See, on whole subject, Maitland, Co//. Papers^ II.
205-222.
^Littleton, II. viii. s. 153.
^Littleton, II. viii. s. 158. Cf. Round, Kings Serjeanties, 21.
56 FEUDAL GRIEVANCES AND MAGNA CARTA
Aguilon, we are told by Madox,i "was charged at the
Exchequer with several escuages. But when it was found
by Inquest of twelve Knights of Surrey that he did not hold
his lands in that county by military tenure, but by serjeanty
of finding a Cook at the King's coronation to dress victuals
in the King's kitchen, he was acquitted of the escuages."
(6) Petty serjeanty may be described in the words of
Littleton as " where a man holds his lands of our lord the
king to yield to him yearly a bow or sword, or a dagger or a
knife ... or to yield such other small things belonging to
war." 2 The grant of lands on such privileged tenures was
frequently made in early days on account of some great
service rendered at a critical juncture to the King's person
or interests. Serjeanties, Miss Bateson tells us, "were
neither always military nor always agricultural, but might
approach very closely the service of knights or the service
of farmers. . . . The serjeanty of holding the King's head
when he made a rough passage across the Channel, of
pulling a rope when his vessel landed, of counting his chess-
men on Christmas Day, of bringing fuel to his castle, of
doing his carpentry, of finding his potherbs, of forging his
irons for his ploughs, of tending his garden, of nursing the
hounds gored and injured in the hunt, of serving as
veterinary to his sick falcons, such and many other might
be the ceremonial or meftial services due from a given
serjeanty." ^
The line between grand and petty serjeanties, like that
between the greater and smaller baronies of chapter 14 of
Magna Carta, was at first vaguely drawn. The distinction,
which Dr. Horace Round considers an illustration of " non-
technical classification,"* may possibly have originated in
'^History of Exchequer^ I. 650, c\\!mgPipe Roll oi 18 Henry III.
2 See Littleton, II. ix. s. 159. With this may be compared the definition given
in chapter 37 of Magna Carta, where John speaks of land thus held by a vassal as
'*quam tenet de nobis per servitium reddendi nobis cultellos, vel sagittas vel
hujusmodi."
^Mediaeval England, 249-250. A similar tenure exists in Scotland under the
name of "blench" — wherein the reddendo is elusory, viz., the annual rendering
of such things as an arrow or a penny or a peppercorn, "if asked only" {si petatur
tantum).
* Round, Peerage^ and Pedigree^ 359.
THE CROWN AND FEUDAL OBLIGATIONS 57
the Great Charter. At a later date, however, petty ser-
jeanties, while liable for "relief," escaped the onerous
incidents of wardship and marriage which grand serjeanties
shared with lands held in chivalry. 1 The way was thus
prepared for the ultimate amalgamation of petty serjeanty
with ordinary socage.
(7) Burgage, confined to lands within free boroughs, is
mentioned as a separate tenure by Littleton, ^ and his
authority receives support from chapter 37 of Magna Carta.
Our highest modern authorities,^ however, treat it rather
as a variety of socage. In Scotland, where several of the
English tenures have failed to obtain recognition, burgage
has established itself beyond a doubt. Even the levelling
process consummated by the Act of 1874 has not abolished
its separate existence.*
Of these tenures, originally six or seven, frankalmoin and
grand serjeanty still exist, but rather as ghosts than reali-
ties; the others have been swallowed up in socage, which
has thus become identical with " freehold." ^ This triumph
of socage is the result of a long process : fee-farm, burgage,
and petty serjeanty, always with features in common, were
gradually assimilated in almost all respects, while a statute
(12 Charles II. c. 24) transformed tenure in chivalry also
into socage. The once humble socage has thus risen high, \ \
and now embraces most of the land of England.*' ' i
The interest .of historians^ centres in tenure by knight-' s
service, which is the very kernel of the feudal system^
1 Littleton, II. viii. s. 158. "^Idia., II. x. s. 162.
^Pollock and Maitland, I. 218.
^Littleton and Coke seem almost to countenance two additional tenures, viz.,
scutage or escuage, and castle-guard. Pollock and Maitland consider both as
alternative names for knight's service. (See I. 251 and I. 257.) The latter is
discussed zn/ra under c. 29 of Magna Carta.
^Jenks, Modern Land Law, 14.
6 It has been well described by Pollock and Maitland (I. 294) as "the great
residuary tenure." In Scotland the "residuary tenure" is not socage but
"feu" (resembling the English fee-farm). Holdings in feu are still originated
by charter, followed by registration (the modern equivalent of infeftment or
feudal investiture), thus preserving an unbroken connection with the feudal
conveyancing of the Middle Ages.
58 FEUDAL GRIEVANCES AND MAGNA CARTA
Lack of definition in the middle ages was a fruitful source
of quarrel : for a century and more after the Norman Con-
quest, the exact amount and nature of military services due
by a tenant to his lord were vague and undetermined.
Each Crown tenant (except favoured foundations like Battle
Abbey) held his lands on condition of furnishing a certain
number of fully armed and mounted soldiers in the event
of war. High authorities differ as to when and by whom
the amount of each vassal's service was fixed. The common
view (promulgated by Professor Freeman i) attributes the
allocation of specific service to Ranulf Flambard, the, un-
scrupulous tool of Rufus. Mr. J. H. Round ^ urges con-
vincing reasons in support of the older view which attributes
the innovation to William L Two facts, apparently, are
certain : that within half a century from the Conquest each
military tenant was burdened with a definite amount of
service; and that no written record of the amount was made
at the time of granting : there were, as yet, no written
charters, and thus disputes arose. Probably, such grants
were made in full Curia, and the only record of the condi-
tions would lie in the memory of the Court.
Long before Magna Carta, the various obligations had
been grouped into three classes, which may be arranged in
order of importance, as services, incidents, and aids.
Under each of these three heads, disputes continually
arose.^ The essence of the feudal tie consisted in the lia-
bility to render "suit and service," that is, to follow the
^Norman Conquest, V. 377; Hist, of William Rufus, 335-7.
'^Feudal England, p. 228 et seq.
^All three forms of feudal obligation — service, incidents, and aids — have long
been obsolete in England. The statute 12 Charles II. c. 24 swept away the
feudal incidents along with the feudal system ; centuries before, scutages in lieu
of military service had become obsolete in the transition from the system of
feudal finance to that of national finance, effected by the Crown in the thirteenth
and fourteenth centuries. Feudal aids were also long obsolete, although James I.,
in desperate straits for money, had attempted to revive two of them. In France
the feudal system, with all its burdensome obligations, remained in full vigour
until it was abolished in one night by the famous decree of the National Assembly
of 4th August, 1790. In Scotland, the feudal system of land tenure still exists,
and certain of its incidents {e.g. reliefs and compositions or fines for alienation)
are exacted at the present day.
THE CROWN AND FEUDAL OBLIGATIONS 59
lord's banner in time of war, and attend his court in time
of peace. It will be more convenient, however, to reserve
full consideration of these services until the comparatively
uncomplicated obligations, known as incidents and aids,
have been first discussed.
I. Ff.udal Incidents, In addition to " suit and service,"
the lord reaped, at the expense of his tenants, a number \
of c^sual_2rofits, which thus formed irregular supplements
to his revenue. These profits, accruing, not annually, but
on the occurrence of exceptional events, came to be known
as "feudal incidents." They were gradually defined with
more or less accuracy, and their number may be giv.en as /
six : reliefs, escheats, wardships, marriages, primer seisins, '
and fines for alienation. 1 '~" '*w
(a) Relief is easily explained. The fee, or feudum, or
hereditary feudal estate, seems to have been the result of
a gradual evolution from the old heneficium (or estate held
for one lifetime), and that again from the older precarium
(or estate held during the lord's will). Grants, originally
subject to revocation, gradually attained fixity of tenure for
the life of the original grantee, and, later on, became trans-
missible to descendants : the Capitulary of Kiersey (a.d.
877) is said to be the first authoritative recognition of the
heir's absolute right to succeed. It would seem that even
after the Norman Conquest, this rule of hereditary descent
was not established beyond possibility of dispute.^ The
heir's ri^hM:o succeed Remained subject to onejcondition,
namely, the^payment^oTa^unrTno^ as a " reliefT" TBis V
was an acknowledgment that the new tenant's right tol
ownership was incomplete, until recognized by his superior 1
— a reminiscence of the earlier precarium from which the
feudum had developed. The amount remained long un-
defined, and the lord frequently asked exorbitant sums.^
(h) Escheat, it has been said, " signifies the return of
^Blackstone, Commentaries^ II. 63, arranges these in a different order, and
mentions as a seventh incident "aids," which are here reserved for separate
treatment.
-See Pollock and Maitland, I. 296.
'See infra^ under c. 2, for the process whereby this evil was redressed.
>
M
6o FEUDAL GRIEVANCES AND MAGNA CARTA
an estate to a lord, either on failure of issue from the tenant
or upon account of such tenant's felony." i This lucid
description conveys a good general conception of escheat;
but it is inaccurate in at least two respects. It does not
exhaust the occasions on which escheat occurs, and it errs
in speaking of " the return " of an estate to a lord, when
more accurately, that estate had always remained his pro-
perty, subject only to a burden, which was now removed.
In theory, the feudal grant of lands was always conditional :
when the condition was broken, the grant fell, and the lord
found himself, automatically as it were, once more the
absolute proprietor, as he had been before the grant was
made. Thereafter, he held the land in demesne, unless
he chose to make a new grant to another tenant. The word
" escheat " was applied indifferently to the lord's right to
such reversions, and to the actual lands which had reverted.
In warlike times the right was valuable, for whole families
might become rapidly extinct. Further, when a landholder
was convicted of felony, his blood became, in the phrase of
a later day, attainted, and no one cmald succeed to any
estate through him. If a man failed in the ordeal of water
provided by the Assize of Clarendon in 1166 for those
accused of heinous crimes, his estates escheated to his lord.
A complication arose when treason was the crime of which
the tenant had been convicted; for the king, as the
injured party, had prior rights which excluded those of
the lord : the lands of traitors were forfeited to the Crown.
Even over the lands of ordinary felons the king had rights
,during a period which w^as defined by Magna Carta. 2
i\ Felony and failure of issue were two main^ grounds ^f
y escheat, but not the only ones ; the goods of fugitives from
justice and of those who had been formally outlawed also
escheated, and Glanvill adds another case, ^ namely,^_female
wards guilty of unchastity (an offence which spoiled the
king's market). Failum to oljey the royal_siiinmons~in
time of war or to pay scutage in lieu thereof might also be
a ground of forfeiture.*
^R. Thomson, Magna Charta, p. 236. "^ Infra, c. 32. ^VII. c. 17.
4 Pollock and Maitland, I. 247 and 250, citing Hist. Abingdon, II. 128.
THE CROWN AND FEUDAL OBLIGATIONS 6i
Escheat was thus a valuable right both to the Crown and
to mesne lords. Its effect was simply this : one link in the
chain was struck out, and the links on either side were fitted
together. If the defaulter was a Crown tenant, all his
forrner siib-Ttenants. whether freeholders or villeins, movedT
up one rung in the feudal ladder and held henceforward^
directlv^f the king, who took over the entire complexus of
legal rights previously enjoyed by the defaulter : rents,
crops, timber, casual profits, and advowsons oTchurches
falling vacant; jurisdictions and their profits; services of
villeins; reliefs, wardshi^s^^and marriages of freeholders,
as these became exigible.
The Crown, however, while taking everything the
defaulter might have taken before default, must take
nothing more — so Magna Carta ^ provides. The rights
and status of innocent sub-tenants must not be prejudiced
by the misdeeds of defaulting lords.
(c) Wardships are described in the Dialogus de Scac-
cario as " escheats along with the heir" (escaeta cum {
herede)? ThiT'exlffession does not occur elsewhere, but
it would be impossible to find any description of ward-
ship which throws more light on its nature and conse-
quences. \^Qien_the heir of a deceased tenant was unfitted ;
to__bear arms by reason of his tender years, the lands wereT '
dudng his minority, without an effective owner£_ the lord 30
treated them as temporarily escheated, entered into posses- ,
sion, drew the revenues, and__applied them to his own I
purposes, subject only to the obligatiqn_of maintaining the j
heir in a manner suited to his station in life. Considerable
sums might thus be spent : the Pipe Roll of the seven-
teenth year of Henry II. shows how out of a total revenue
of ;^5o 6s. 8d. from the Honour of "Belveeir," ;^i8 5s.
had been expended on the children of the late tenant.^
Wardship came to an end with the full age of the ward,
that is, in the case of a military tenant, on the completion
ofjiis twenty-first year, " in that of a holder in socage on „.A7
the^completion^fthe fifteenth^and inlne case ot a burgess"
^See Infra, c. 43. ^See Hughes' edition, p. 133.
2 See Dialogus, p. 222 (citing Pipe Roll, p. 27).
y
62 FEUDAL GRIEVANCES AND MAGNA CARTA
when the boy can count money, measure clothj,_and so
(y^JlLli---Wardship of females normally ended atjhe_age
of fourteen, ""becauselHaTa woman of suchjgejnayjiaye
a husband_able to dcT knight's service." Anheiress who
dfd not slucceed to the estate until she was fourteen thus
escaped wardship altogether, but if she became a ward at
a younger age, the wardship continued till she attained
sixteen years unless she married earlier .^
All the remunerative consequences flowing from escheat
I flowed also from wardship — rents, casual profits, advow-
sons, services of villeins, and reliefs. Unlike escheats,
however, the right of the Crown here was only temporary,
and Magna Carta sought ^ to provide that the implied
conditions should be respected by the Crown's bailiffs or
nominees : the lands must not be wasted or exhausted,
but restored to the son when he came of age, in as good
condition as when his father died.
One important aspect ought to be emphasized : Ward-
ship affected bishoprics as well as lay baronies, extending
over the temporalities of a See between the death of one
prelate and the instalment of his successor. It was to the
king's interest to keep sees vacant, while his Exchequer
drew the revenues and casual profits.* This right was
carefully reserved, even in the comprehensive charter in
which John granted freedom of election. s
(d) Marriage as a feudal incident is difficult to define;
for its meaning changed. Originally it seems to have
implied little more than the right of a lord to forbid an
^Glanvill, VII. c. 9. In socage and burgage tenures no wardship was
recognized ; the guardianship went to the relations of the ward, and not to his
feudal lord. Complicated, but equitable, rules applied to socage. The maternal
kindred had the custody, if the lands came from the father's side ; the paternal
kindred, if from the mother's side (Glanvill, VII. c. Ii). In plain language, the
boy was not entrusted to those who had an interest in his death. Cf. infra, cc.
3, 4 and 37.
2 Littleton, II. iv. s. 103. ^See under c. 5.
^ What these were may be read in the Pipe Rolls ^ e.g., in that of 14 Henry II.
when the Bishopric of Lincoln was vacant.
5 See Sel. Chart., 288. Contrast Stephen's Oxford Charter; Sel. Chart.,
1 20- 1. Cf. supra, p. 32, and infra, under c. i.
THE CROWN AND FEUDAL OBLIGATIONS 63
heiress to marry his personal enemy. Such veto was
reasonable, since the husband of the heiress would become
the tenant of the lord. The claim to concur in the choice
of a husband gradually expanded into an absolute right to
dispose of the lands and person of the female ward : the
prize might be a bribe to any unscrupulous gentleman of
fortune who placed his sword at the King's disposal, or it
might go to the highest bidder. The lady passed as a mere
adjunct to her own estates. At fourteen she might be sent
to market, and the only way in which she could protect
herself against an obnoxious husband was by out-bidding
her various suitors.
This right seems, at some uncertain date, to have been
extended from females to males, and instances of sums
thus paid occur in the Pipe Rolls. It is difficult at first
sight to imagine how the Crown found a market for such
wares as male wards; but probably wealthy fathers were j
ready to purchase desirable husbands for their daughters. ]
Thus in 1206 a certain Henry of Redeman paid forty marks
for the hand and lands of the heir of Roger of Hedon,
" ad opus filiae suae,''^ while Thomas Basset secured a prize
in the person of the young heir of Walerand, Earl of
Warwick, to the use of any one of his daughters. 2 This
extension to male heirs is usually explained as founded on
a strained construction of chapter 6 of Magna Carta ; but
the beginnings of the practice can be traced before 1215.3
The lords' right to sell their wards was recognized and
defined by the Statute of Merton, chapter 6. The attempts
made to remedy some of the most serious abuses may be
read in Magna Carta.* Hallam ^ considers that "the
rights, or feudal incidents, of wardship and marriage were
nearly peculiar to England and Normandy," and that the
French kings never " turned this attribute of sovereignty
into a means of revenue." ^
(e) Pruner Seisin, which is usually regarded as a separate
'^Rotuli de oblatis et finibus, p. 354. '^ Rot, Clans., 37, 55.
'^ Pollock and Maitland, I. 305. * See infra, under chapters 6, 7, and 8.
^Middle Ages, II. 429. 6 p. 437.
64 FEUDAL GRIEVANCES AND MAGNA CARTA
incident, and figures as such in Blackstone's list, is perhaps
better understood, not as an incident at all, but as a special
j procedure — effective and summary — whereby the Crown
i could enforce the four incidents already described. It was
an exclusive prerogative of the Crown, denied to mesne
lords. ^ When a Crown tenant died, the King's officers
had the right to enter into immediate possession, and to
exclude the heir, who could not touch his father's lands
without permission from the Crown : he had first to prove
his title by inquest, give security for any balance of relief
or other debts, and perform homage.^ It will be readily
seen what a strong strategic position all this assured to the
King in any disputes with the heir of a dead vassal. If
the Exchequer had doubtful claims against the deceased,
its officials could satisfy themselves before admitting the
heir to possession. If the heir showed any tendency to
evade payment of feudal incidents, the Crown could check-
mate his moves. If the succession was disputed, the King
might favour the claimant who pleased or paid him most ;
or, under colour of the dispute, refuse to disgorge the estate
— holding it in custody analogous to wardship, and mean-
while drawing the profits. If the son happened to be
abroad when his father died, he would experience difficulty
in forcing the Crown to restore the estates. Such was the
experience of William Fitz-Odo on returning from Scotland
in 1 20 1 to claim his father's carucate of land in Barn-
borough. 3 Primer seisin was thus not so much a separate
incident, as a right peculiar to the Crown to take summary
measures for the satisfaction of all claims against a deceased
tenant or his heir. Magna Carta contains no direct refer-
ence to it, but chapters 37 and 53, providing against the
^The Bishop of Durham enjoyed it, so it seems to be stated in a charter of 1303
(Lapsley, Pal. of Durham ^ 133). But this forms no real exception; since the
Bishop, as an Earl Palatine, enjoyed the regalia of a king.
2 See Pollock and Maitland, I. 292. From Statute of Marlborough, c. 16,
primer seisin extended over serjeanty as well as knight's service. Statute of
Merton, c. 7, provided that a ward might refuse a marriage on undertaking to
pay the offered price when he came of age. Under c. 8, double the value might
be exacted for a secret marriage or one in fraud of the lord's right.
'^ Rotuli de oblatis, p. 114.
THE CROWN AND FEUDAL OBLIGATIONS 65
abuse of prerogative wardship, have a bearing on the
subject.!
(/) Fines for alienation occupy a place by themselves.
Unlike the incidents already discussed, they became
exigible not on the tenant's death, but on his parting with
his estate during his lifetime, either as a gift or in return
for a price. How far could he effect this without consent
of his lord? This was, for many centuries, a subject of
heated disputes, often settled by compromises, under which
the new tenant paid a fine to the lord for recognition of his
title. Such fines are payable at the present day in Scotland
(under the name of " compositions ") from feus granted
prior to 1874; ^"^> where no sum has been mentioned in
the Feu Charter, the law of Scotland defines the amount
exigible as one year's rent. Magna Carta contains no
provisions on this subject. Disputes, long and bitter, took
place in the thirteenth century ; but their history is irrele-
vant to the present inquiry .2
11. Feudal Aids. The feudal tenant was expected to
come to the aid of his lord in any special crisis or emer-
gency. At first, the occasions on which these "aids"
might be demanded were varied and undefined. Gradually
they were limited to three. Glanvill,^ indeed, mentions
only two : the knighting of the overlord's eldest son, and
the marriage of his eldest daughter ; but he intends these,
perhaps, as illustrations rather than as an exhaustive list.
Before the beginning of the thirteenth century the recog-
nized aids were the ransoming of the King and the two
already mentioned. * This understanding was embodied in
Magna Carta. ^
^Sir Edward Coke {Co^e upon Littleton^ TJ A) is the original source of much
confusion as to the nature of primer seisin, which he seems to have considered as
a second and additional relief exacted by the Crown, amounting to the whole rent
of the first year. The Popes, he further held (erroneously), were imitating this
practice when they exacted a year's rent from every newly granted benefice under
name of "first fruits." These errors have been widely followed {e.g. Thomson,
Magna Ckarta^ p. 416; Taswell Langmead, Const. Hist.^ 50).
2 See Taswell Langmead, Const. Hist. , pp. 5 1 -2 ; also Pollock and Maitland, II. 326.
3 IX. c. 8.
■^An aid to marry the king's eldest sister might be taken, if not previously
exacted by her father.
"See infra ^ under chapter 12.
66 FEUDAL GRIEVANCES AND MAGNA CARTA
A tradition has been handed down from an early date,
that these aids were voluntary offerings made as a mark of
affection.^ Long before John's reign, however, the obliga-
tion had become fixed by law ; the tenant dared not refuse
to pay the recognized three. But, when the Crown exacted
contributions for any other reason, it required consent of
the commune concilium.
The Great Charter, while confirming this tacit com-
promise, left the amount of aids undefined, merely stipu-
lating that they should be " reasonable." Examples of
such payments, both before and after the Charter, are
readily found in the Exchequer Rolls. Thus, in his four-
teenth year Henry IL took one mark per knight's fee for
his daughter's marriage; Henry HL took 20s., and
Edward L 40s. for a similar purpose. For Richard's
ransom, 20s. had been exacted from each knight's fee
(save those owned by men actually serving in the field) ;
and Henry HL took 40s. in his thirty-eighth year at the
knighting of his son. The Statute of Westminster L^
fixed the " reasonable " aid payable to mesne lords at 20s.
per knight's fee, and 20s. for every estate in socage of ;^20
annual value. This rate, it will be observed, is one-fifth
of the knight's relief.^ The Crown, in thus enforcing
" reason " on mesne lords, seems never to have intended
that the same limit should hamper its own dealings with
Crown tenants, but continued to exact larger sums when-
ever it thought fit.* Thus £2 per fee was taken in 1346
at the knighting of the Black Prince.
A statute of Edward HL^ at last extended to the Crown
the same measure of " reasonableness " as had been applied
^Thus, the Abingdon Chronicle (II. 113) speaks of "auxilia quod barones
michi dederunt"; while Bracton says (Book II. c. 16, s. 8): "Auxilia fiunt de
gratia et non de jure; cum dependeant ex gratia tenentium, et non ad voluntatem
dominorum."
2 3 Edward I. c. 36. ^ Fixed at lOOs. by c. 2 of Magna Carta.
* Cf. Pollock and Maitland, I. 381-2. One entry in the Memoranda Roll of 42
Henry III. (cited Madox, I. 615) seems to admit that the Crown could not exact
more than 20s. ; but in 1258 the baronial opposition would be strong in the
Exchequer as elsewhere.
5 25 Ed. III. Stat. S, c. II.
THE CROWN AND FEUDAL OBLIGATIONS 67
three-quarters of a century earlier to mesne lords. The
last instances of the exaction of aids in England occur as
late as the reign of James I., who, in 1609, demanded one
for the knighting of the ill-fated Prince Henry, and in 1613
another for the marriage of his daughter Elizabeth.
III. Suit and Service. This phrase expresses the
essential obligations inherent in the very nature of the
feudal tie. It may be expanded (as regards tenure in
chivalry) into the duty of attendance at the lord's court,
whether met for administrative or judicial purposes, or for
reasons of mere display, and the further duty of military
service under that lord's banner in the field. Suit had
ceased to be an urgent question before the reign of John.
Indeed, the barons were gradually approaching the modern
conception, which regards it as a privilege rather than a
burden to attend the comynune concilium — the embryo
Parliament — of the King.
It was otherwise with the duties of military service,
which were rendered every year more unwillingly, partly
because of the increased frequency of warlike expeditions,
partly because of the greater cost of campaigning in distant
lands like Poitou, partly because the English barons were
completely out of sympathy with John's foreign policy and
with him. We have seen that the want of definition in
the Conqueror's reign left to future ages a legacy of strife.
William and his barons lived in the present; and the
present did not urgently call for definition. Therefore, the
duration of the military service, and the conditions on
which exemption could be claimed, were originally vague ;
but the return due (servitium dehitum) for each knight's fee
was gradually fixed by custom at the service of one fully
armed horseman during forty days. There were still, how-
ever, innumerable minor points on which disputes might
arise, and these remained even in 12 15. Indeed, although
several chapters of the Charter attempted to settle certain
of these disputed points, others were left as bones of con-
tention to subsequent reigns : for example, the exact equip-
ment of a knight ; the liability to serve for more than forty
days on receiving pay for the extra time ; what exemption
^
68 FEUDAL GRIEVANCES AND MAGNA CARTA
might be claimed by churchmen ; how far a tenant might
compromise for actual service by tendering money ; whether
attendance and money might not both be withheld, if the
King did not lead his forces in person ; and whether service
was due for foreign wars equally as for home ones.^
Difficulties increased as time went on. The Conqueror's
followers had estates on both sides of the Channel : his
wars were theirs. Before John's reign, these simple rela-
tions had become complicated by two considerations. By
forfeitures and the division of inheritances, holders of
English and of Norman fiefs had become distinct. On the
other hand, the expansion of the dominions of the English
kings increased the number of their wars, and the expense
of each expedition. The small wars with Wales and Scot-
land formed sufficient drain on the resources of English
magnates without their being summoned to fight in Maine
or Gascony.
Were the barons bound to follow John in a forlorn
attempt, of which they disapproved, to recover his lost fiefs
from the French Crown ? Or were they bound to support
him only in his legitimate schemes as King of England?
Or were they, by way of compromise, liable for services
in the identical possessions held by William the Conqueror
at the date when their ancestors first got their fiefs — that is,
for wars in England and Normandy alone? So early as
1 198 the Knights of St. Edmunds refused to serve in
Normandy, while offering to pay scutage.2 The northern
barons in 12 13 declared that they owed no service what-
soever out of England.^ This extreme claim put them
clearly in the wrong, since John could produce precedents
to the contrary. When, on his return from the unfortunate
expedition of 12 14, he demanded a scutage from all who
^ Some of these questions might be answered by the terms of special charters :
the Hundred Rolls (1279) relate how Hugh de Plesens must go with the King for
forty days at his own, and thereafter at the King's expense. Rot. Hund.^ II.
p. 710; cf. for France, Etablissements de St. Louis, I. c. 65.
^ Jocelin of Bralcelond, 63, cited by Pollock and Maitland, I. 250 n.
^ See R. Coggeshall, p. 167 ; the barons argued non in hoc ei obncxios esse
secundum munia terrarum suar'um.
THE CROWN AND FEUDAL OBLIGATIONS 69
had not followed him to Poitou, the malcontents declared
that they had no obligation either to follow him out of the
kingdom, or to pay a scutage in lieu thereof.^ Pope
Innocent was probably correct in condemning this contention
as founded neither on English law nor on feudal custom. 2
There is some ground for believing that a compromise was
mooted on the basis that the barons should agree to serve
in Normandy and Brittany, as well as in England, on being
exempted from fighting elsewhere abroad.^
A definite understanding was n^ver arrived at : chapter
16 of Magna Carta provided that existing services were not
to be increased, without defining what these were. This
was to shelve the difficulty : the dispute went on under
varying forms and led to an unseemly wrangle between
Edward I. and his Constable and Marshal, dramatized in
a classic passage by Walter of Hemingburgh.* Strangely
enough, the Confirmatio Cartaruvi of 1297, which was,
in part, the outcome of this later quarrel, omits (like Magna
Carta itself) ^ all reference to foreign service. The omission
from both charters of all mention of the chief cause of
dispute is noteworthy. It must be remembered, however,
that the question of liability to serve abroad had practically
resolved itself into that of liability to scutage, and that
chapters 12 and 14 of the Charter of 12 15 provided an
adequate check on the levy of all scutages; but this is a
subject that requires separate and detailed treatment.
IV. Scutage, The Crown did not always insist on
personal service, but was frequently willing to accept a
commutation in the form of a money payment. The sub-
ject of scutage is one of the most vexed of questions, all
received opinions of yesterday having to-day been thrown
into the melting pot. The theories of Stubbs and Freeman,
once universally accepted, require substantial modifications.
Four propositions may be stated with some confidence : (i)
*W. Coventry, II. 217.
2 See his letter dated ist April, 121 5, in JVew Rymer, I. 128.
' See "unknown charter" in Appendix. * Chronicon^ IT. 121.
'See, however, infra under c. 16.
70 FEUDAL GRIEVANCES AND MAGNA CARTA
that scutage is an ambiguous term with a vague general
meaning as well as a narrow technical meaning; (2) that
the importance of the changes introduced by Henry II. in
1 156 and 1 159 has been much exaggerated; (3) that scutage
was always in the option of the King, never of the barons,
his tenants; and (4) that at a later time, probably during
John's reign, scutage changed its character, and became,
partly through altered circumstances and partly by the
King's deliberate policy, a much more burdensome
exaction. Each of these propositions requires explana-
tions :
(i) The proper technical meaning of scutagium or
" shield-money " is a money payment of so much per
" shield " (that is, per knight's fee) by a tenant in lieu of
actual attendance in the army of his feudal lord : it is, as
Dr. Stubbs explains, ^ " an honourable commutation for
personal service." The word, however, is also more loosely
used for any exaction assessed on a feudal basis, irrespective
of the occasion of its levy ; and, in this wider sense, includes
feudal aids and other payments as well.^
(2) Professor Freeman, Dr. Stubbs, and their adherents
held that one of Henry's most important reforms was the
invention of scutage; that he allowed his Crown tenants
at their discretion to substitute payments in money for the
old obligation of personal service in the field — this option
being granted to ecclesiastics in 1156, and to lay barons in
1 159. Such a theory had a priori much to recommend it.
A measure of this nature, while giving volume and elasti-
city to the resources of the Crown, was calculated subtly to
undermine the basis of the feudal tie; but Henry, far-seeing
statesman as he was, could not discard the ideals of his
own generation : no evidence that he made any sweeping
change is forthcoming. On the contrary, his grandfather,
Henry I., is shown by the evidence of extant charters to
have accepted money in place of the services of knights
when it suited him (notably from church fiefs in ii09),3
and there is no evidence (direct or indirect) to show that the
^ Const. Hist,, I. 632. 2]viadox, I. 619.
'See Round, Feudal England, 262 ff., 532.
THE CROWN AND FEUDAL OBLIGATIONS 71
grandson accepted such commutation when it did not suit
him. Scutage was thus known in England half a century
before 11 56 — the traditional date of its introduction.
(3) Further, neither before nor after the reign of Henry II.
had the individual baron any option of tendering at his
discretion money in place of personal service. The con-
clusions on this subject formulated by Dr. Horace Round
lie implicitly in the examples from the Pipe Rolls stored
in the famous work of Madox. From these it would appear
that the procedure of the Exchequer of the great Angevin
and his two sons might be explained in some such pro-
positions as these :
(a) The option to convert service into scutage lay with
the Crown ; not with the tenants, either individually or
as a body. When the King summoned his army, no baron
could (as Professor Freeman would have us believe) simply
stay away under obligation of paying a small fixed sum
to the Exchequer. On the contrary, Henry and his sons
jealously preserved the right to insist on personal service
whenever it suited them ; efficient substitutes were not
always accepted, much less money payments.
(b) If the individual wished to stay at home he required
to make a special bargain with the King, paying such sum
as the King thought fit to demand and sometimes having
to find a substitute in addition. Exorbitant sums (not
properly " scutages " at all) might thus be extorted from
stay-at-homes ne transfretent or pro remanendo ah exercitu
— phrases which appear in the Pipe Rolls of Richard. A
Crown vassal in John's twelfth year made fine "that he
might send two knights to serve for him in the army of
Ireland." ^ In such cases, each baron made his own
bargain with the Crown : a scutage, on the contrary, " when
it ran in the land " was at a uniform rate.
(c) The tenant-in-chivalry who stayed at home without
first making his bargain was in much worse plight. He
had broken faith, and in strict feudal theory had forfeited
his fief by failing to perform the service for which he held
it. He was " in mercy," and might be glad to accept such
1 Madox, I. 658.
Q
72 FEUDAL GRIEVANCES AND MAGNA CARTA
terms of pardon as a gracious king might offer him.^
Sometimes, quite small amercements were inflicted : the
Abbot of Pershore in 11 96 escaped with 40s :^ But the
Crown sometimes insisted on total forfeiture.^
It was the duty of the Barons of Exchequer to determine
whether lands had thus escheated by default, and also to
determine the amount of " forfeit " to be taken where con-
fiscation was not justified or insisted on. The barons
wished to refer such questions to the judicium pariumA
(4) Scutage tended continually to become more burden-
some :
(a) With new inventions and more complicated fashions
in arms and armour for man and horse, and increased rates
payable for the hire of mercenaries, the expenses of a
campaign steadily increased. It was not unnatural that
the normal rate of scutage should increase in sympathy.
Under Henry the recognized maximum had been two
marks, the exact equivalent of 40 days' wages at the normal
rate of 8d. per diem.^ Usually he was content with a
smaller sum per knight's fee : 20s., 13s. 4d. or even los.
being sometimes taken.
(h) A second method of increasing the yield of scutage
was to readjust the assessment on w^hich it was based,
by increasing the number of contributory knights' fees.
Henry II. in 1166 had invited his unsuspecting barons to
furnish him with details of the number of knights actually
enfeoffed on their lands both before and after the death of
his grandfather; and then treated the latter as a sort of
unearned increment, the benefit of which should be shared
by the Crown. The amount of servitium dehitum as pre-
viously reckoned was increased by the addition of the
number of knights of the novum feoffamentum, that is, of
1 Pollock and Maitland, I. 247, noted this distinction under Edward ; it
existed, as the above-cited instance proves, under John.
^Pipe Roll of Richard I., cited Madox, I. 663.
^ Pipe Roll of 12 John, cited ibid. ^ Cf. infra, under cc. 39 and 21.
5 Interesting details are given by Vinogradoff, English Society, 15 ff. C/1
Round, Feudal England, 277 fF.
THE CROWN AND FEUDAL OBLIGATIONS 73
those created subsequent to the death of Henry I.^ The
basis of assessment thus fixed in 1166 remained unaltered
at John's accession.
(c) The third respect in which scutages tended to become I
more burdensome was in their increased frequency. This
was, in part, a consequence of the growth of the Empire
of the Kings of England, bringing with it a widening of
interests and ambitions, and an increase in the number and
expense of wars. Much depended, however, on the spirit
in which this feudal prerogative was used, on the amount
of consideration given to the needs and interests of the ;
barons. Neither Henry nor Richard seems to have /
regarded it as other than an expedient to be reserved for
special emergencies, not as a permanent source of revenue ^
in normal times. .
Henry II. seems to have levied money in name of scutageV
only when actually at war — on seven occasions in all \
during a reign of thirty-five years ; and only once at a rate I
exceeding 20s., if we may trust Mr. Round,^ and that when
he was putting forth a special effort against Toulouse.
Richard I., rapacious as he was, levied, apparently, only
four scutages during ten years, and the rate of 20s. was
never exceeded even in the King's hour of urgent need,
— in 1 194, when the arrears of his ransom had to be paid
and preparations simultaneously made for war in Nor-
mandy.
If it can be shown that John altered established usages
under every one of these heads, breaking away from all
restraints, and that too in the teeth of the keen opposition
of a high-spirited baronage whose members felt that their
pride and prestige as well as their money-bags were
attacked, a distinct step is taken towards understanding
the crisis of 12 15. Such knowledge would explain why a
storm, long brewing, burst in John's reign, neither sooner
nor later; and even why some of the disreputable stories
told by the chroniclers and accepted by Blackstone and
others, found inventors and believers.
It is here maintained that John did make changes in all
^ Round, Feudal England^ 237-9. "^ Feudal England, 277 seq.
74 FEUDAL GRIEVANCES AND MAGNA CARTA
of these directions; and, further, that the incidence of
this increase in feudal burdens was rendered even more
unendurable by two considerations : — because at his acces-
sion there remained unpaid (particularly from the fiefs of
the northern knights) large arrears of the scutages imposed
in his brother's reign, ^ and because in June, 1212, he drew
the feudal chain tight by a drastic and galling measure.
That John elevated scutage from a weapon reserved for
emergencies into a regular source of revenue, and that he
raised the rate demanded beyond the recognized maximum
of two marks, becomes apparent from a glance at the table ^
of scutages extorted during his reign :
First scutage
of
reign-
-I 198-9 —
-2
marks
per
knight's fee.
Second
»
I 200- I
0
)»
5)
Third
5J
I20I-2
-7
5)
3)
Fourth
5)
1202-3
2
>>
»
Fifth
)>
1203-4
2
»J
)>
Sixth
5>
1204-5
2
5J
5)
Seventh
)>
1205-6
20S.
)»
Eighth
»
1209-10
2
marks
»
Ninth
5J
I2IO-II
2
5)
))
Tenth
?J
I2IO-II
20S.
5>
Eleventli
J5
I213-I4
3
marks
))
It will be seen that, in his very first year, John took a
scutage at two marks per scutum. Next year he wisely
allowed a breathing space; then without a break in each
of the third, fourth, fifth, sixth and seventh years of his
reign, scutages were extorted in quick succession at the
same high rate. Fines, in addition to this scutage of two
marks, were exacted from those who had not made the
necessary compromise for personal service in due time.s
These scutages were collected with increasing difficulty,
and arrears accumulated; but the spirit of opposition
^ Norgate, John Lackland, p. 122.
^Norgate, John Lackland^ p. 123 note, correcting Swereford's lists in the
Red Book of Exchequer. Further corrections are perhaps necessary : R. Wendover
III. 173, mentions a scutage of 2J marks in January, 1204.
^ See Ramsay, Angevin Empire, 390, and authorities there cited.
THE CROWN AND FEUDAL OBLIGATIONS 75
increased even more rapidly. In 1206, apparently, the
breaking point was almost reached.^ Accordingly, in
that year, some slight relaxation was allowed — the annual
scutage was reduced from two marks to 20s. John's
needs, however, were as great as ever, and would prevent
further concessions, unless something untoward happened.
Something untoward did happen in the summer of 1207,
when John quarrelled with the Pope. This postponed his
quarrel with the baronage. John had, for the time being,
the whole of the confiscated property of the clergy in his
clutches. The day of reckoning for this luxury was still
far distant, and the King could meanwhile enjoy a full
exchequer without goading his Crown tenants to rebellion.
For three years no scutage was imposed. In 1209, however,
financial needs again closed in on John, and a new scutage
of two marks was levied ; followed in the next year actually
by two scutages, the first of two marks against Wales, and
the second of 20s. against Scotland. John had no sense of
moderation. These three levies, amounting to a total of
five-and-a-half marks per fee within two years, strained the
tension almost to breaking point.
During the two years following (Michaelmas, 121 1, to
Michaelmas, 12 13) no scutage was imposed. John, how-
ever, although he thus a second time relaxed the tension,
had no intention to do so for long. On the contrary, he
determined to ascertain if scutages could not be made to
yield more in the future. By writs, dated ist June, 1212,
he instituted a strict Inquest into the amount of service
exigible from every estate in England. Commissioners
were appointed to take the sworn verdicts of local juries as
to the amount of liability due by each Crown vassal. Mr.
Round 2 considers that previous writers have unaccountably
ignored the importance of this measure, " an Inquest worthy
to be named in future by historians in conjunction with
those of 1086 and ii66,"3 and describes it as an effort "to
revive rights of the Crown alleged to have lapsed." John
^Cf. Norgate, yij^w Lackland^ 125. "^ Comnmne of London, 273-4.
' Yet, of recent historians, Ramsay {Angevin Empire, 432) treats it briefly, and
Miss Norgate {John Lackland, 163) barely notices it.
76 FEUDAL GRIEVANCES AND MAGNA CARTA
intended by this Inquest, the returns to which were due on
the 25th June, to prepare the necessary machinery for
wringing the uttermost penny out of the next scutage when
occasion for one again arose. That occasion came in 12 14.
Up to this date, even John had not dared to exact a rate
of more than two marks per Icnight's fee; but the weight
of his constant scutages had been increased by the fact that
he sometimes exacted personal services in addition, and
that he inflicted crushing fines upon those who neither went
nor arranged beforehand terms of composition with the
King.i
Thus insidiously throughout the entire reign, the stream
of feudal obligations steadily rose until the barons feared
that nothing of their property would be saved from the
torrent. The normal rate of scutage had been raised, the-
frequency of its imposition had been increased, the con-
ditions of foreign service had become more burdensome,
and the objects of foreign expeditions more unpopular;
while attempts were sometimes made to exact both service
and scutage in the same year. The limit of the barons*
endurance was reached when, under circumstances peculiarly
inauspicious, John, in May, 12 14, demanded a new scutage
at the unprecedented rate of three marks on every fee,"
grounded doubtless on the searching inquest of 12 12.2
This outline of the history of scutage makes plain that
grievances connected with its abuse formed one of the chief
incentives to the insurrection that resulted in the winning
of the Great Charter.
^Miss Norgate (123) describes the exactions supplementing the scutages:
" These scutages were independent of the fines paid by the barons who did not
accompany the King on his first return to Normandy in 1199, of the money taken
from the host as a substitute for its service in 1201, of the equipment and payment
of the 'decimated' knights in 1205, and the fines claimed for all the tenants-in-
chivalry after the dismissal of the host in the same year, as well as of actual
services which many of those who had paid the scutage rendered in the campaigns
of 1202-4 and 1206."
*See Miss Norgate, John Lackland^ 210, and cf. supra^ 31. For a minor
grievance connected with scutage and the writ de habendo scutagio, see infra,
under c. 15. The later history of scutage is outlined in Pollock and Maitland,
I. 254. Cf. infra, under c. 12.
ROYAL JUSTICE AND FEUDAL JUSTICE 77
III. Royal Justice and Feudal Justice.
A well-known aphorism describes the King as " the sole
fountain of justice." It would be an anachronism to "
transport this metaphor into the thirteenth century. In
John's reign there still were, not one, but many competing ^
jurisdictions. It was by no means certain that the King's
Courts were the proper tribunals to which a wronged
individual must repair. On the contrary, the great bulk
of the rural population, the villeins, had no locus standi
except in the court of the manor to which they belonged;
while the doors of the royal Courts had been opened to the
ordinary freeman no earlier than the reign of Henry II.
Royal justice was still the exception, not the rule. Each
man must seek redress, in the ordinary case, in his own
locality. To dispense justice to the nation at large was
no part of the normal business of a medieval King. vj'
I. Rival systems of Law Courts, In the thirteenth
century, there existed not one source of justice, but many .J
t/Rival courts, eagerly competing to extend their own sphere
of usefulness and to increase their own fees, existed in a
bewildering multitude. ^Putting aside for the moment the
Courts Christian, the Borough Courts, the Forest Courts,
and all exceptional or peculiar tribunals, there existed three
great rival systems of jurisdiction which may be named in
the order in which they became in turn prominent in
England. 1
(i) Local or District Courts. Justice was originally a
local product, administered in rude tribunals which partook
more or less of a popular character. Each shire had its
assembly for hearing pleas, known as a " shire-moot " in
Anglo-Saxon days, and as a " comitatus " after the Norman
Conquest; while each of the smaller districts subdividing
the shire, and forming units of administration for purposes
^ Too absolute a line must not be drawn between the three types of court. In
one sense all tribunals were, or tended to become, royal courts. The king's
representatives presided in the "popular courts," and the king received a share
of the fines levied there; while, in Prof. Vinogradoff's words i^English Society^
108), "all the well-known franchises or liberties of the feudal age were chips
from the block of royal authority."
78 FEUDAL GRIEVANCES AND MAGNA CARTA
of taxation, defence, justice, and police, had a moot or
council of its own, serving as a court of law, to which the
inhabitants of the villages brought their pleas in the first
instance. These smaller districts were known as hundreds
in the south, and as wapentakes (a name of Danish deriva-
tion) in the north.
/ The theory generally received is that all freemen were
i originally suitors in the courts of shire and hundred, and
that the whole body of those present, the ordinary peasant
(" ceorl ") equally with the man of noble blood (" eorl "),
took an active part in the proceedings, pronouncing (or,
at least, concurring in) the judgments or dooms there
declared ; but that, as time progressed, the majority of the
Anglo-Saxon ceorls sank to the half-servile position of
villeins — men tied for life to the soil of the manor, and
passing, like property, from father to son. These villeins,
although still subjected to the burden of attendance, and to
some of the other duties of their former free estate, were
deprived of those rights which had once formed the counter-
part of the obligations. Another school of historians, it is
true, denies that the mass of the population, even in very
early times, ever enjoyed an active share in the dispensation
of justice. It is unnecessary here to attempt a solution of
the intricate problems of the courts of shire and hundred ;
or to discuss the still more vexed question how far the small
assembly of each township is worthy to be reckoned a formal
Court of Law.i
(2) Feudal Courts. Centuries before the Norman Con-
quest, the system of popular or district justice found
itself confronted with a rival scheme of jurisdictions — the
innumerable private courts belonging to the feudal lords..^
These private tribunals, known as feudal, manorial, or
seignorial courts, slowly gained ground on the older public
courts of shire, hundred, and wapentake.^
^John's Charter makes no mention of these courts, although c. 25, forbidding
increase of the farms of shires, may have a bearing on the subject. Henry's Charters
of 1 217 and 1225 regulate their times of meeting, Cf. in/ra, Part IV.
2 This account of the relations of the two sets of courts would receive the support
of recent writers, such as Maitland and Round, as well as of the older generation,
ROYAL JUSTICE AND FEUDAL JUSTICE 79
Practically every holder of land in England came to be
also the holder of a court for the inhabitants of that land.
The double meaning of the word " dominus " illustrates the
double position of the man who was thus both owner and
lord.i In the struggle between two schemes of justice, the
tribunals of the feudal magnates triumphed over, but never
abolished their rivals. The earlier popular courts lived on ;
but the system of district justice, which had once embraced
the whole of England, was honeycombed by the growth of
feudal courts. As each village passed under the domina-
tion of a lord, the village-moot became a manorial court
endowed with wider powers and more effective sanctions
for enforcing them. Further, as complete hundreds fell
under control of powerful magnates, the courts of these
hundreds were also transformed into feudal courts : fran-
chises thus took the place of many of the old popular moots.
Still, the older system retained part of the disputed ground,
thanks to the protection of the Crown. Many hundreds
never bowed to the exclusive domination of any one lord,
and the courts of the shires were guarded by the Norman ?
Kings against the encroachment of even the most powerful!
barons.
Although it was the policy of the Norman Kings to,
prevent their barons from gaining excessive powers of juris-
diction, it was by no means their policy to suppress these
jurisdictions altogether. The Conqueror and his sons were
glad that justice should be administered, even in a rough-
and-ready manner, in those districts whither the Crown's
arm was not long enough to reach, and where the popular
courts were likely to prove inefficient. The old system and
the new existed side by side; it was to the interest of the
central government to play off the one against the other.
In later days (but not till long after Magna Carta), each
manorial court had three distinct aspects, according to the
such as Stubbs and Freeman. Mr. Frederic Seebohm may be mentioned as per-
haps the most weighty upholder of the opposite view, which regards the manorial
courts as of earlier origin than those of hundred and shire.
^C/. "landlord."
8o FEUDAL GRIEVANCES AND MAGNA CARTA
class of pleas it was called upon to try. Later writers
distinguish absolutely from each other, the Court Baron,
settling civil disputes between freeholders of the manor;
the Court Customary, deciding non-criminal cases among
the villeins; and the Court Leet, a petty criminal court
enforcing order and punishing small offences. The powers
of these courts might vary, and in many districts the juris-
diction over misdemeanours belonged not to the steward
of the manor, but to the sheriff in his half-yearly Circuits
or " Tourns " through the county. In imperfectly feuda-
lized districts the Tourn of the sheriff performed the same
functions as the Court Leet did within a franchise.
(3) Royal Courts. Originally, the King's Court had
been merely one among many feudal courts — differing in
degree rather than in kind from those of the great earls
or barons. The King, as feudal lord, dispensed justice
among his tenants, just as any baron or freeman dispensed
justice among his tenants, bond or free. No one dreamed,
in the time of the Norman Kings, that the Curia Regis
could undertake the labour of dispensing justice for the
whole nation. The monarchy had no machinery at com-
mand for a task which no Anglo-Saxon King, nor even
■William I., could have undertaken. No attempt in this
indirection was made until the reign of Henry II., who was
jplaced in a position of unprecedented power, partly by
circumstances, but chiefly by his great abilities. Even he,
born reformer as he was, would never have increased so
greatly the labours of government, if he had not seen that
the change would enhance the security of his throne and
the revenue of his exchequer.
From an early date, however, the business of the Monarch
was wider than the business of any other lord. In a dim
way, too, it must have been apparent from the first, that
offences against the established order were offences also
against the King, and that to redress these was the King's
business competent in the King's Courts. The Crown,
further, asserted a right to investigate pleas of special
importance, whether civil or criminal. Still, under William
and his sons, royal justice had made no deliberate attempt
ROYAL JUSTICE AND FEUDAL JUSTICE 8i
to become national justice, or to supersede feudal justice :
the struggle came with the reforms of Henry II. ^
Thus the three great systems of jurisdiction, popular! \ \
justice, feudal justice, and royal justice succeeded eachV-^
other, on the whole, in the order in which they are here
named. Yet the sequence is in some ways logical rather
than chronological. No absolute line can be drawn, show-
ing where one system ended and the next began. The
germs of manorial jurisdiction may have been present from
an early date. Shire-courts and hundred courts alike were —
continually in danger of falling under the domination of
powerful local magnates. Yet, the shire-courts were j
successful in maintaining till the last (thanks to royal
favour) their independence of the manorial jurisdictions ; I
while only a proportion of the hundred courts fell into x
bondage. The royal courts, again, from an early date,
withdrew causes from the Shire Courts and interfered with
manorial franchises. The Courts Baron were silently under-
mined, until they sank into decrepitude without ceasing to
exist. With these caveats, the three systems may be re-
garded, in some measure, as following one another in the
order named : — popular justice, feudal justice, royal justice.
II. Legal Procedure. The procedure adopted in litiga-
tion in Anglo-Saxon and Norman times was similar in
essentials in all three classes of tribunals, and differed
materially from the practice of courts of law at the present
day. Some knowledge of the more glaring contrasts
between ancient and modern procedure will conduce to an
understanding of several obscure provisions of Magna
Carta.
Avoiding technical language, and eliminating special
procedure peculiar to any one court or country, the prin-
cipal stages in a litigation in a modern court of law may be
given briefly as follows : (i) On the complaint of the party
aggrieved a summons, or writ, is issued by an officer of
^ The stages in the process, extending from the reign of Henry I. to that of
Edward I., by which royal justice encroached on feudal justice, may be studied in
Maitland's preface to Sel. Pleas in Manorial Courts^ pp. liii. ff. See also Pollock
and Maitland, I. 181-2.
F
82 FEUDAL GRIEVANCES AND MAGNA CARTA
the Court. Proceedings are opened by the command
addressed to the defendant to appear in Court and answer
what is alleged against him.
(2) In the usual case each party lodges written statements
of his facts and pleas — that is, of the circumstances as they
appear to him (or such of them as he hopes to bring evidence
to prove) — on which he founds his claim or his defence,
and of the legal principles he intends to deduce from these
circumstances. When these statements of facts and pleas
have been revised and adjusted, the complete data are
before the Court ; each party has stated what he considers
essential to his case.
(3) Proof is, in due course, led; that is, each party is
afforded an opportunity of proving such facts as he has
alleged (and as require proof through the denial of his
opponent). This he may do by documents, witnesses, or
oath. Each party has the further privilege of shaking his
opponent's evidence by cross-examination.
(4) The next important stage is the debate, the main
object of which is to establish by legal arguments the pleas
founded on ; to deduce the legal consequences inherent in
the facts which have been proved.
(5) Finally, the Judge gives his decision. He has to
determine, after weighing the evidence led by either party,
what facts have really been established, and how far the
various pleas of plaintiff and defendant respectively are
implied in these facts. Reasoning of such a kind as can
be successfully performed only by a trained legal mind is
thus necessary before the final decree or sentence can be
pronounced by a Judge in a modern court of law.
A trial in Anglo-Saxon and early Norman times stands
in notable contrast to all this in its stages and procedure,
and even more in the spirit which pervades the whole.
Thus, the proceedings, from first to last, were purely oral,
there being no original writ or summons, no written
pleadings, no record kept of the decision except in the
memories of those present. The functions of " the Judges "
were entirely different, and called for no previous training,
since they were not required either to weigh a mass of
ROYAL JUSTICE- AND FEUDAL JUSTICE 83
evidence or to determine the bearing of subtle legal argu-
ments, but merely to see fairplay, and to decide, according
to simple rules, well established by centuries of custom,
by what test the allegations of plaintiff and defendant were
respectively to stand or fall. Finally, the arrangement
of the stages of the litigation was entirely different : it is
with something of a shock that the modern lawyer learns
that in civil and criminal causes alike " judgment " invari- I
ably preceded " trial." Reflection will convince him that 1
each of these words had in the Middle Ages a meaning
diiferent from what it bears to-day. That this is so can be
best understood by following the stages of the old procedure.
(i) The initial difficulty was to obtain the presence of
the defendant in Court, since there existed a strange
reluctance either to compel his attendance or to allow
judgment to pass against him by default. No initial writ
was issued commanding him to appear; almost endless
delays were allowed.
(2) When both parties had been, after many adjourn-
ments, actually brought face to face before the Court, the
statements alike of the claim and of the defence were made
verbally and in set formulae, the slightest slip or stumble
in the words of which involyed compIitCfailirfe. TKis"!^"""
merely one illustration of the tremendously formal and
technical nature of early legal procedure, a trait common to
all primitive systems of jurisprudence.
(3) Before the plaintiff could put the defendant on his
defence, he required to show some presumption of the
probability or bona fides of his case. This he usually did
by producing two friends ready to substantiate his claim,
known sometimes as his " suit " (Latin secta), or his " fore-
witnesses." Their testimony had no reference to the parti-
cular facts of the case; it was not weighed against the
" proof " afterwards led by the defendant ; its object was
merely to warrant the Court in demanding " proof " from
the latter at all.^
^ Sometimes no fore-witnesses were required ; for example, where the claim was
for restoration of stolen cattle, traced by "hue and cry" to defendant's house or byre.
The presumption was here so strong as to render corroborative evidence unnecessary.
84 FEUDAL GRIEVANCES AND MAGNA CARTA
(4) Then came the judgment or "doom," which partook
in no respect of the nature of the judgment of a modern
tribunal. It came before the proof or trial, not after it,
and was therefore called a " medial " judgment. It con-
sisted in decreeing whether or no, on the strength of the
previous procedure, the defendant should be put to his
proof at all; and if so, what " proof " should be demanded.
Now, the exact test to be appointed by the court varied
somewhat, according to circumstances, but long-established
custom had laid down with some exactitude a rule applic-
able to every case likely to occur ; and, further, the possible
modes of proof were limited to some four or five at the
outside. In Anglo-Saxon times, these were mainly com-
purgation, ordeal, witnesses (whose functions were, how-
ever, widely different from those of witnesses in modern
law), and charters. The Norman Conquest introduced for
the new-comers, a form of proof previously unknown in
England — " trial by combat " — which tended, for the upper
classes at least, to supersede all earlier procedures. The
"proof," of whatever kind it might be, thus appointed by
the " judges " for the defendant's performance was techni-
cally known as a " law " (Latin lex) in the sense of a " test "
i)r "trial" or "task," according to success or failure in
which his case should stand or fall.i To pronounce a
" judgment " in this sense was a simple affair, a mere
formality in the ordinary case, where room for dubiety could
hardly be admitted: .thus it was possible for "judgment"
to be delivered by all the members of a feudal court, or all
the suitors present at the hundred or shire-moot.
(5) The crucial stage, this " trial " which thus came after
" judgment," consisted in one party (usually the defendant)
essaying, on the day appointed, to satisfy the court as to
the truth of his allegations by performing the task or " law "
which had been set or " doomed " to him. When this
consisted in the production of a charter, or* of " transaction
witnesses " (that is, the testimony of those officials appointed
in each market-town to certify the conclusion of such
bargains as the sale of cattle), it commends itself readily to
^ See in/j-a under cc. 38 and 39, where /ex terrae is discussed.
ROYAL JUSTICE AND FEUDAL JUSTICE 85
modern approvaL More frequently it took the form of
"an oath with oath-helpers," the plaintiff bringing with
him eleven or twelve of his trusty friends or dependents
to swear after him the words of a long and cumbrous oath,
under risk of being punished as perjurers for any slip in
the formula. Sometimes the decision was referred to the
intervention of Providence by appealing to the ordeal of
the red-hot iron or the more dreaded ordeal of water.
After the Norman Conquest, the trial in all litigations
between men of high rank, took the form oi duellum or
legally regulated combat between the parties. The defen-
dant gained his case if he caused the plaintiff to own him-
self a " craven," or if he held out till nightfall against the
plaintiff's attempts to force him to utter that fateful
word.^
This earlier form of " lex " or trial (which is referred to
in several clauses of Magna Carta) 2 was thus entirely
different from the modern " trial." It may be said without
exaggeration that there was no " trial " at all in the current
meaning of the w^ord — no balancing of the testimony of
one set of witnesses against another, no open proof and
cross-examination, no debate on the legal principles
involved. The ancient " trial " was merely a formal test,
which was, except in the case of battle, entirely one-sided.
The phrase " burden of proof " was inapplicable. The
litigant to whom " a law " was appointed had rather the
" privilege of proof," and usually won his case — especially
in compurgation, and even in ordeal if he had arranged
matters properly with the priest who presided. In one
sense, the final " trial " was determined by the parties
themselves, or by one of them ; in another and higher
sense the facts at issue were left to Providence ; a miracle,
if necessary, would attest the just claim of the innocent.^
' Details may be studied in Neilson's Trial by Combat.
2 See infra^ cc. 38 and 39, where ordeal and compurgation and other forms of
lex are further discussed.
'Cf. Thayer, Evidence, p. 8. "The conception of the trial was that of a
proceeding between the parties, carried on publicly, under forms which the
community oversaw."
86 FEUDAL GRIEVANCES AND MAGNA CARTA
The essentials of this procedure ^ were the same in
Norman as in Anglo-Saxon England, and that in all three
.classes of tribunals — popular, manorial, and royal courts.
Two innovations the Normans did make; they introduced
trial by combat and '^ inquisitio." Among the preroga-
tives of the Norman Dukes was this right to compel the
sworn evidence of reliable men of any district — men
specially picked for the purpose, and put on oath before
answering the questions asked of them. This procedure
was known as inquisitio (or the seeking of information)
from the point of view of the government making the
inquiry, and as recognitio (or the giving of information)
from the point of view of those supplying it. This device
was capable of endless extension to new uses in the deft
hands of the Norman Kings. William employed it in
compiling Domesday Book; while his successors made it
the instrument of experiments in the science of taxation.
It has a double claim to the interest of the constitutional
historian, because it was one of the influences that helped
to mould our Parliamentary institutions; and because
several of the new uses to which it came to be put had
a close connection with the origin of trial by jury. The
recognitors, indeed, were simply local jurors in a rude or
elernentary form. 2
III. Reforms of Henry II. in Law Courts and Legal
Procedure. It was reserved for Henry of Anjou to inaugu-
^ These stages of procedure are fully illustrated by recorded cases. Two of
these, both from the reign of John, may here be cited, (i) " Hereward, the son
of William, appeals "Walter, the son of Hugh, of assaulting him, in the King's
peace, and wounding him in the arm with an iron fork, and giving him another
wound on the head ; and this he offers to prove on his body as the Court shall
appoint. And Walter defends all of it by his body. And it is testified by the
coroners and by the whole county that the same Hereward showed his wounds
at the proper time, and has made sufficient suit. Therefore it is decreed that
there should be 'battle.' . . . Let them come armed, a fortnight from St.
Swithin's day, at Leicester." Sel. Pleas of Crown (Selden Society), p. i8.
(2) "Walter Trenchebof was said to have handed to Inger of Faldingthorpe the
knife with which he killed Guy Foliot, and is suspected of it. Let him purge
himself by water that he did not consent to it. He has failed and is hanged."
Ibid., p. 75-
2 The relation of "recognition" to trial by jury is discussed infra, Part HL,
section 7.
ROYAL JUSTICE AND FEUDAL JUSTICE ^7
rate a new era in the relations of the three classes of courts.
He was the first king deliberately to plan the overthrow
of the feudal jurisdictions by insidiously undermining
them, if not yet by open attack. He was the first king
to reduce the old district courts so thoroughly under the
control of royal officials as to turn them practically into
royal courts. He was the first king also to throw open
the doors of his own courts of law to all-comers, to all •
freemen, that is to say, for the villein had for centuries
-, — "^ -J
still to seek redress in the Court of that very lord of the I
manor who was too often his oppressor. 1
In brief, then, Henry's policy was twofold : to convert
the County Courts completely into Royal Courts, since
in them royal officials now dispensed royal justice
according to the same rules as prevailed at the King's
Curia ; and to_reduce all manorial_Qr private Courts to
insignificance by diverting pleas to his own Curia, and
leaving the rival tribunals to die gradually from inanition.
Both branches of this policy met ultimately with success,
although the event hung in the balance until long after his
death. The barons, though partially deceived by the
insidious nature of Henry's reforms, did what they could
to thwart him; but the current was with the Crown. \
Royal justice steadily encroached upon feudal justic^. One
of the last stands made by the barons has left its traces
in several chapters of Magna Carta.2 These contain what
^ The trend of learned opinion for the moment is towards transferring the chief
share of credit for remedial changes from Henry II. to his grandfather. Prof.
Haskins, too, has shown reason for holding that the younger Henry found precedents
in the procedure of his Angevin father as well as of his Norman grandfather [Amer.
Hist. Rev., VIII. 6i8). There is some evidence also that Henry II. avoided any
violent breaking with the past. Mr. Davis {Engl, under the Normans^ p. 283)
shows Henry and his Justice Glanvill acting in a spirit friendly to the private courts.
It is possible, however, to found erroneous estimates upon such items of evidence.
The true inventor is the man who adapts for common use what was before excep-
tional : Henry II. can afford to be judged by this test. To him, rather than to
Henry I., belongs the credit for revolutionizing the whole system of dispensing
justice. Cf. G. B. Adams {Origin of Engl. Const., 106-7): "It is in his time
that these changes are finally made and the new methods become permanently a
part of the constitution."
2 E.g. 34 and 39.
88 FEUDAL GRIEVANCES AND MAGNA CARTA
seem, at first sight, to be merely trivial alterations of
technical points of court procedure ; but inextricably bound
up with them are principles of wide constitutional impor-
tance. It was Henry's good fortune or policy to disguise
radical reforms until they looked like small changes of
procedure; it follows that the framers of Magna Carta,
while appearing merely to seek the reversal of these trivial
points, were really seeking to return to the totally different
conditions which had prevailed prior to the reforms of
Henry.
The short account of that monarch's system of procedure,
necessary to a comprehension of Magna Carta, falls natur-
ally into two divisions.
(i) Criminal Justice, (a) By his Assizes of Clarendon
and Northampton, Henry reserved important crimes~Tor
the exclusive consideration of his own judges either on
circuit or at his court ;- and he demanded entry for these
judges into all franchises for that purpose. In this part of
his policy, the King was completely successful; heinous
crimes were, in the beginning of the thirteenth century,,
admitted on all hands to be " pleas of the Crown " (that is,
cases reserved exclusively for royal jjurisdiction) ; and
Magna Carta made no attempt to reverse this part of the
Crown's policy: all that was attempted in 12 15 was to
obtain a promise that these functions, now surrendered to
the Crown forever, should be discharged by the Crown's
officials in a proper manner.
(b) Henry's usual good sense, in this matter stimulated
by some notable miscarriages of justice, led hipi to question
the equity of the procedure usually adopted in criminal
pleas : for private " appeal " (or accusation by the injured
party or his nearest surviving relative), he substituted,
whenever possible, communal accusation ; that is, the duty
of indicting suspected criminals before the King's Justices
was no longer left to private initiative, but was laid on a
body of neighbours — the predecessors of the Grand Jury
of later days. Appeals were discouraged and rules laid
down restricting the right of accusation. 2
1 See tn/ra, under cc. 24 and 45. 2 See tn/ra, under chapter 54.
ROYAL JUSTICE AND FEUDAL JUSTICE 89
(c) A necessary complement was the discouragement of
" trial by combat." An ingenious device was invented and
extended to an increasing number of cases; an accused
individual might apply for a writ known as de odio et atia,
and evade the duellum by a reference to what was practically
a jury of neighbours. ^
(2) Civil Justice. Henry's innovations under this head
were equally important. In his reign justice, it is some-
times said, was pigeon-holed. Much attention was
bestowed on the formalities^"~^f" litigation ; while pleas
began to be classified into stereotyped groups, each form
of grievance having its appropriate remedy, to be obtained
only by means of the appropriate writ. '-""^""N
(a) The Writ System. An unflinching rule was estab-1
lished that no case could be brought before the royal couri/
until a writ had been obtained from chancery. This ha^
to be paid for, sometimes at a fixed rate, and sometimes at
whatever sum the Crown demanded. The whole procedure
in the royal courts, which followed the issuing of such a
writ, came to be known as " the writ system." From an
early date, much attention was directed to the devising of
forms of writ applicable to various cases. The system,
somewhat inflexible from the first, had become absolutely:
rigid long before the close of the thirteenth century. If
aTproper writ was not selected, or if no such writ had been
invented, the wronged individual had no remedy in the
King's courts of common law. Registers of writs were
drawn up, copied and enlarged, and transmitted from one
generation to another. 2
(h) Control of Feudal Courts. Whether devised for that
purpose or not, this writ system proved a useful instrument
for diverting the stream of litigation from the barons' courts
to the curia regis. Henry, if we may credit Glanvill,
succeeded in establishing the somewhat astounding rule
that no plea concerning land could be commenced in any
court without the authority of a royal wTit.^ Even if such
^ See znj'ra, under chapter 36. ^ See Maitland, Collected Papers^ II., no- 173.
2 Glanvill xii, 25. For a discussion of the difficulties involved in accepting
Glanvill see Adams, Origin, 96.
90 FEUDAL GRIEVANCES AND MAGNA CARTA
writs were issued as matter of course, the mere need of
asking for them would supply Henry with information
doubly valuable in relation to certain other expedients still
to be explained. That King, applying to his own needs
procedure known to the Carglingian Kings, secured an
effective means of evoking suits regarding freehold from
the seignorial courts to his own. This was done by pro-
cedure initiated by two types of writs: " writs of righ_t "
addressed to the holder of a court, bidding liim do justice
under penalty of interference by the royal court ; and " writs_
izaecip^e" addressed to the sheriff, bidding him require the
holder of a piece of land to hand it over to a claimant or
explain to the King why he has not done so.^
It is probable that even in 12 15 the Crown had not
fully developed the consequences afterwards seen to be
involved in the writ of right, properly so called; but
Henry II. and his sons seem freely to have used the writ
praecipe in such a manner as to cause their barons to lose
their jurisdiction — an abuse struck at by chapter 34 of
Magna Carta.
(c) Royal Pleas and Common Pleas. The mass of new
business made it necessary to increase the staff of judges
and apportion the work. A natural division was that
between ordinary (or common) pleas and pleas of the
Crown. This distinction is recognized in many separate
chapters. 2 Thus two groups of judges were formed which,
in later years, developed into separate courts — the Court
of Common Pleas (known as " the Bench," that is, the
ordinary Bench), and the King's Bench (known earlier
as the court Coram Rege, supposed to be held in the King's
presence).
(d) The Petty Assizes. Special procedure for deter-
mining titles to land or rights of possession was also
invented by Henry to supersede trial by battle. These
Assizes, as they were called, are fully discussed elsewhere. ^
While the Grand Assize is not mentioned in Magna Carta,
its abuse was indirectly struck at by the clause concerning
^ See Brunner, Schwurgerichte^ 78-80. Details are discussed infray under c. 34.
^See infra^ under chapters 17 and 24. ^See infra^ under chapter 18.
ROYAL JUSTICE AND FEUDAL JUSTICE 91
writs praecipe in chapter 34 : the Petty Assizes, however, —
would seem to have won favour with the barons, who in
chapter 18 demanded that regular sessions for hearing them
should be held four times a year.
These were the chief innovations that enabled Henry II.
to effect a revolution in the relations of royal to feudal
justice. As time went on, new writs were continually
devised to meet new types of cases ; and litigants flocked
readily to the King's Courts, leaving the seignorial courts
empty of business and of fees. Nor was this the only
grievance of the barons. When one of their own number
was amerced or accused of any offence involving loss of
liberty or lands, he might be compelled by the Crown,
under Henry and his sons, to submit to have the amerce-
ment assessed, or the criminal proceedings conducted, by
one of the new Benches (by a tribunal composed of some
four or five of the King's officials), in place of the time-
honoured judgment of his peers assembled in the Commune
Concilium (the predecessor of the modern Parliament).
Can we wonder that the barons objected to be amerced/:^
and judged by their inferiors ? ^ Can we wonder that they/ ^
resented the complete though gradual supersession of their
own profitable jurisdictions by the royal courts ?2 or that
they looked with suspicion on every new development of
the royal justice ? Can we wonder that, when they seemed
to have King John for the moment in their power, they
demanded redress of these grievances, as well as of those
connected with increase of feudal burdens ? The cause for
wonder rather is that their demands were not more sweep-
ing : the barons, in their hour of triumph, accepted
cordially one half of the royal innovations.
The chapters bearing on jurisdiction may be arranged in
two groups, some reactionary, and some favourable to
Henry's reforms. On the one hand, no lord of a manor}-
shall be robbed of his Court by the King evoking before 1
the royal courts pleas between two freeholders of the lord's 1
manor; 3 no freeman shall be judged or condemned by the »
^See zn/ra, under chapters 21 and 39. ^ See infra, under chapter 34.
3 c. 34.
92 FEUDAL GRIEVANCES AND MAGNA CARTA
King's officials, but only before the full body of his peers ; ^
earls and barons must be amerced only by their equals. 2
On the other hand, in prescribing remedies for abuses
connected with numerous branches of legal procedure, the
barons accepted by implication this new procedure itself
and the royal encroachments implied therein. For example,
the Crown's right to hold " Common Pleas " was impliedly
admitted, when the barons asked and obtained that these
should be tried in some certain place (that is, at West-
minster). 3 Yet these very pleas must have included many
cases which, prior to Henry II.'s reforms, would have
been tried in a seignorial court. Again, in regulating the
petty assizes, chapters 18 and 19 admit the Crown's right
to hold them. Here, as in chapter 40, the ground of com-
plaint is not that there is too much royal justice, but rather
that there is too little of it : henceforth it must be neither
delayed nor denied. Further, the encroachments made
by Henry H. in 1166 on the private franchises in the matter
of criminal jurisdiction are tacitly accepted by the acquies-
cence in the King's definition of " Pleas of the Crown "
implied in chapter 24.
These, then, are the two groups into which the innova-
tions made by Henry and his sons naturally fell, as viewed
by John's opponents in 1215 : some of them had come to
be warmly welcomed; while others, it was insisted, must
be swept away.
^c. See ;Vz/ra, under chapter 39. ^^ 21. ^c. 17.
PART III.
MAGNA CARTA : ITS FORM AND CONTENTS.
I. Its Prototypes: Earlier Charters.
The traditional view makes Magna Carta the direct
descendant of Henry Beauclerk's Coronation Charter,
which is, in turn, regarded as merely an amplification of the
old coronation oath sworn by the Conqueror and his sons,
in terms borrowed from a long line of Anglo-Saxon kings,
stretching back from Edward Confessor to Edgar, Alfred
and Egbert, until its origin is lost in the mists of anti-
quity. According to this time-honoured view, which
insists on an exclusively Anglo-Saxon pedigree for the
charters of Norman and Angevin kings, the charters of
Henry I. and John were regarded as confirmations to the
nation at large of the essential principles of the old laws
of Alfred and of Edward, thus bridging over, alike in form
and substance, the gulf of the Norman Conquest.
The accuracy of these preconceptions has of late years
been rudely questioned. The simple formula for solving
all problems of English constitutional origins by assuming
an unmixed Anglo-Saxon ancestry, has been challenged
from more sides than one. Magna Carta, like the Con-
stitution itself, is of mixed parentage, tracing its descent (lA
not entirely from Teutonic, but partly from Norman, and
even Danish and Celtic sources. In the first place, John's
Charter derives some of its vital clauses from documents
not couched in charter form. The Constitutions of Claren-
don of 1 164 and the Forma Procedendi of 1194 are as
undoubtedly antecedents of Magna Carta as is the Corona-
94 MAGNA CARTA: ITS FORM AND CONTENTS
tion Charter of Henry itself. The same is true of many
grants made by successive kings of England to the Church,
to London and other cities, and to individual prelates and
barons. In a sense, the whole previous history of England
went to the making of Magna Carta.
Then, again, the exclusively Anglo-Saxon origin of the
antecedents of Henry's Charter is by no means left un-
challenged. A recent American writer, attacking the older
theories as advanced by Bishop Stubbs, has formulated
these three propositions : that Henry's charter was feudal
in character rather than constitutional or national, pro-
mising " a regulated feudal government " purged of Rufus'
misdeeds rather than a return to a " national " type of
government ; that its substance was derived from Norman
innovations rather than from the Confessor's or Canute's
laws ; and that its form was founded on continental models,
possibly on some Norman borough charter, and by no
means on the old coronation oath.^
These iconoclastic theories require to be modified : the
claims of Magna Carta, on its formal side, at least, to an
Anglo-Saxon ancestry have found a powerful advocate in
Mr. W. H. Stevenson, 2 who holds that the Anglo-Norman
charters of liberties " are developments of the Anglo-
Norman writ charter, and that in its turn is . . . merely
the Anglo-Saxon writ translated into Latin." ^
Looking both to the contents and the formalities of
execution of John's Great Charter, the safer opinion would
*See Dr. H. L. Cannon's article, Amer. Hist. J^ev.^ XX. 37. Some of his
theories, however, had been anticipated (see, e.g. Prothero, S. de Montfort, 16),
and others have not been substantiated.
Engl. Hist. Rev.y XXVII. 1-8. Dr. R. L. Poole is also an advocate of the
traditional view: see ibid.^ XXVIII. 444.
^ Ibid.y XXVII. 4. Mr. Stevenson explains further that "the Anglo-Saxon
writ was in its origin a letter from the King to a shire-moot, and this characteristic
clung closely to the Anglo-Norman writ-charter of the twelfth century" (p. 5).
He also shows how the double-faced pendant seal, in the use of which William
and his sons followed the Confessor, was not derived by Edward from the
Normans, who in his day used (like the Kings of France) a seal plaqui. The
whole article throws much light on the diplomatics of the genesis of Magna
Carta.
ITS PROTOTYPES: EARLIER CHARTERS 95
seem to be, that, like the English Constitution, it is of
mixed origin, deriving elements from ancestors of more
races than one ; but that the traditional line of descent from
the oaths and writs of Anglo-Saxon kings, through the
Charter of Henry I., is one that cannot be neglected.
The promises of good government that connect King
John with the old kings of Wessex are thus the outcome of
an essential feature of the ancient monarchy, and of the rules^"
that regulated succession to the Crown. Two rival prin-
ciples, the elective and the hereditary, from an early date,
had struggled for the mastery. In an unsettled state of
society, nations cannot allow the sceptre to pass to an infant
or a weakling. When a king died, leaving a son of tender
age and a brother of mature ability, the magnates of the
kingdom, the so-called Witan, claimed the right to choose
a fitting successor. The exact relations between the elec-
tive and the hereditary principles were never laid down
with absolute precision : the practice usually followed by
the Witenagemot was to select some near kinsman of the
late king competent for the post. The king-elect had still
to be solemnly anointed, and this gave to the Church an
important share in deciding who should be king. Not
later than the days of Edgar, it became the practice for
the officiating archbishop to exact an oath of good govern-
ment from the new sovereign before his final coronation.
The terms of this oath became stereotyped; and, as
administered by Dunstan to King Ethelred, they are still
extant.^ ' 1
This may be analyzed into three promises — peace to x/ I
God's Church and people; repression of violence in men
of every rank ; justice and mercy in all judgments. When
William I., anxious in all things to fortify the legality of
his title, took the oath in solemn form, he created a prece- -'
dent of tremendous importance, although he may have
regarded it at the moment as an empty formality. 2 The
^See Memorials of St. Dunstan (Rolls Series), p. 355.
2 Florence of Worcester and the Worcester version of the Chronicle agree that
the Conqueror took the oath. *'* William of Poitiers and Guy are silent about the
oath" (Freeman, Norman Conquest, III. 561, note).
96 MAGNA CARTA: ITS FORM AND CONTENTS
quasi-elective character of the kingship, the need for -
coronation by the Church, and this tripartite oath were all
preserved.
This was of vital moment, because limits were thereby
placed, in theory at least, to prerogatives that threatened '
to become absolute. The power of the Norman kings
might almost be described as irresponsible despotism,
tempered by fear of rebellion. Three forces, indeed, acted
as curbs : the necessity for consulting the Curia Regis ; —
the restraining influence of the Church; the growth of a
body of public opinion, confined as yet to the upper classes.-
These elements counted for something, but failed to
restrain sufficiently even an average king ; while they were
powerless against a strong ruler, like William I. The
moment at which the Crown might be taken at disadvantage
was during the interregnum that followed a king's death.
Thus, William Rufus, anxious to prevent his elder brother
Robert from making good his claim to the English throne,
succeeded chiefly through the friendship of Lanfranc. To
gain this, he was compelled to make promises of good
government, taking oath in the ancient form. In the same
reign, began the practice of supplementing verbal promises
by sealed charters. No such charter was indeed issued
either by Rufus or his father when they were crowned ; but
the younger William, at a critical period in his reign,
granted a short Charter of Liberties, the text of which has
not come down to us. By a treaty made at Caen in 1091,
Duke Robert and Rufus agreed that each should constitute
the other his heir. Thus, at Rufus' death, Henry was, in
a sense, a usurper, and this made it necessary for him to
bid high for influential support. ^ It is to this doubtful title
that Englishmen owe the first Charter of Liberties that has
come down to us.^
Roger of Wendover relates how " as many charters were
made as there are counties in England, and by the King's
command they were deposited in the abbeys of every county
^Stubbs, Const. Hist., I. 328-9, and authorities there cited.
^ See text in Appendix. For textual criticism see Liebermann, Trans. A\ H. S,,
VIIT. 21 flf.
ITS PROTOTYPES: EARLIER CHARTERS 97
as a memorial," and this is confirmed by an analysis of the
copies still preserved.^
Henry's coronation charter was the price paid for
support in his candidature for the Crown. Its terms
contain, however unconsciously, an indictment of his
brother Rufus' government and, perhaps, in part also of
his father's. The new king was merely " playing to the
gallery " : when his purpose was served, his promises were
broken .2 On the bearing of these promises there is room
for diversity of opinion. Dr. Stubbs' contention that
Hen^y " definitely commits himself to the duties of a
national king " ^ has been rejected, as already explained,
by recent critics. The more modern view is strengthened
by an analysis of the Charter, revealing important conces-
sions to the barons and the Church, while those to the
people at large were few and vague. Of the fourteen
chapters into which it is usually divided, chapter one pro-
ceeds on the narrative that the kingdom had been oppressed
by unjust exactions. Henry, in the first place, makes free
the holy Church of God, " so that I shall neither sell nor
farm out nor, on the death of archbishop, or bishop or
abbot, accept anything from the demesne of the church or
from its feudal-tenants until a successor has been inducted
to it."
It seems doubtful whether the regrettably vague phrase-
ology of the qualifying clause is intended merely to apply
the generalities of the church's " freedom " to specific
instances, or whether it must be taken as a deliberate
restriction. The prohibition of selling has been read as
referring to the simoniacal practice of taking money from
aspirants to episcopal preferments; but more probably it
was meant to prohibit the alienation of the property of a
vacant see, a practice that must have been often resorted to,
if we judge from the efforts at recovery made by successive
archbishops, notably by Becket. This reading is the more
probable from the fact that " selling " is here coupled with
^ See Liebermann, op. cit. On the whole subject of publication of charters by
Henry I., Stephen and John, see Poole, Engl. Hist. Rev.y XXVIII. 444-453.
2 Round, Feudal England^ 227. ^ Const. Hist., I. 331.
G
98 MAGNA CARTA: ITS FORM AND CONTENTS
"farming out," an expedient clearly inapplicable to pre-
latical appointments and referring to the Crown's practice
of granting leases of the lands of vacant sees for nominal
annual returns in consideration of a heavy grassum paid
to the Treasury at the commencement of the lease. The
rest of the clause is best interpreted as a renunciation of the
claim to exact either a " relief " from a prelate on his
appointment or payments in lieu of relief from tenants of
a vacant see or royal abbey. ^
The last clause of the chapter abrogates evil customs
whereby the kingdom was unjustly oppressed, and then
i proceeds to define them — a process that occupies the
V remaining thirteen chapters of the document. Chapter 2
promises that reliefs of feudal tenants should be " just and
legitimate." 2 Chapters 3 and 4 guard against abuse of
the feudal incidents of marriage and wardship.^ Chapter 5
abolishes as an innovation " the common mintage " (an
exaction levied by the mints when the coinage was altered),*
and enjoined the punishment of any one taken with false
money — provisions finding no echo in John's Charter.
Chapter 6 remits a number of arrears, reliefs, and
penalties due to Rufus at his death. Chapter 7 confirms
I crown-tenants in the right to dispose of their personal
' f estate by will, and provides for the division of the property
of intestates among their wives, children, relations, and
vassals, and for the good of their own souls. ^ Chapter 8
seems to promise the total abolition of the Norman system
of forfeitures and amercements (in respect of petty offences,
as opposed to treasons and crimes) and a return to the
Anglo-Saxon system of a fixed tariff of bots and wites.^
^The use of the word ^' donee" is ambiguous, and might grammatically be
strained to make the clause a prohibition of wardship, coupled with an endorse-
ment of relief : the King must take nothing uniz/ the new bishop gets possession.
Another interpretation would stretch the prohibition to include both wardship and
relief, and indeed to include the taking of profits of any sort whatever. It has
also been read as mainly a prohibition against the Crown's permanent appropriation
of " escheats " falling to a see during a vacancy. See Makower, ConsL Hist, of
Church, 17.
2 Cf. infra, under cc. 2 and 3 of 121 5. '^ Cf infra, cc. 3 to 6.
*SeeStubbs, Early Engl. Hist., 113. ^ggg infra, cc. 26 and 27.
"See Pollock and Maitland, IT. 512-3. See also infra, c. 20.
ITS PROTOTYPES: EARLIER CHARTERS
99
Chapter 9 is concerned with the " murdrum " fine — a
payment exacted by the Norman kings from all the inhabi-
tants of a hundred in which a corpse had been found, where
the slayer remained undiscovered and the dead man's
identity as a person of English birth could not be proved.
" Murder " was thus primarily secret slaying, in the sense
that the perpetrator was not known, and, secondarily, the
fine exacted on that account. This heavy fine, whose
original amount is variously given as 40 or 46 marks, was
intended as a protection to Normans against the native
Englishry they oppressed.
Henry remitted all " murder-fines " incurred before his
coronation, and promised that those incurred after that date
should be " justly " paid for " in accordance with the law of
King Edward " — a clause difficult to reconcile with the
recognized opinion that the murdrum was unknown in
England prior to 1066, unless on the supposition that the
draftsman of the Charter of 11 00 was strangely ignorant
of the usages of thirty-four years earlier. Perhaps the
" murder-fine " was not an invention of the Conqueror and
his sons, but an old English institution put by the Normans
to new uses. An alternative suggestion may be hazarded
that here (as perhaps elsewhere in the charter) the reference
to the good laws of Edward was a mere tag or " common
form," meant to please his subjects without committing the
King to anything in particular.
Chapter 10 contains no concession (unless it be an
implied renunciation of Rufus' encroachments), but, on the
contrary, a blunt intimation that Henry, with his barons'
consent, would retain the forests as his father had had them.
The barons' consent may be partly explained by their
expectation to enjoy, as more or less habitually in atten-
dance upon Henry, a share in the pleasures of the hunt
of which the King was " master." By chapter II., Henry
concedes *' propria dono w.eo to knights holding their
lands per loricas [that is, by knight's service] to have the
lands of their demesne ploughs quit of all gelds and of
every [non-military] service, in order that, as being relieved
by so great a relief, so they might effectually provide them-
(^PWWwijhB*Si»i'v*'.'
100 MAGNA CARTA: ITS FORM AND CONTENTS
selves with horses and arms for my service and the defence
' of my kingdom." In thus exempting Crown-tenants hold-
ing by the " hauberk " (that is to say his " barons," in the
\ wider sense of the word) from Danegeld,' on the distinct
understanding that they should keep in readiness an
efficient military equipment, Henry aimed at making hard-
and-fast an old and fluctuating rule that prohibited Crown-
tenants from being subjected to a double set of burdens.
The lands of knights and churchmen, who already served
the King in other ways, were not expected to contribute
Danegeld in respect of their home-farms. Holders of
knight's fees, however, must keep proper weapons and
armour for themselves and their horses — an obligation
involving an expenditure constantly increasing with every
advance in the art of war. The chapter thus recognizes a
contrast between land subject to military service and land
subject to geld; "the inland and warland of old Eng-
lish fiscal arrangements, the dominium and terra geldahilis
respectively of the Geld Roll of 1084." ^ The fact that
Henry's Charter dravv^s so sharp a line between the two,
suggests that the barons may have made this a condition
of their support of his claims against those of Robert.
Henry's promise, however, was never strictly carried out :
the practice continued to fluctuate. Under Flenry II., only
the barons of the Exchequer and a few privileged religious
persons enjoyed exemption. ^ Gradually the distinction
between inland and warland became extinct.
The remaining clauses of the Charter of Henry I. are
mainly of a formal character. Chapter 12 declares a firm
peace for the future throughout his kingdom — thus mark-
ing the end of the interregnum consequent on his brother's
death. Chapter 13, on the strength of which wide-reaching
theories have sometimes been built, seems to be merely an
amplification of the purely formal chapter that precedes it :
it restores the law of Edward, with the reforms his father
^See Prof. VinogradofF, in a review of the first edition of this book, Law
Quarterly Revircu, XXL, 250-7. See also his Growth of the Manor, 226-7, and
his Engl, Society, 191.
^ Dialogus de Scaccario, I. c. II.
ITS PROTOTYPES: EARLIER CHARTERS loi
had effected with the barons' consent. The old law was
vague; the innovations definite and well known. Chapter
14 proclaims terms and conditions of indemnity, extended
to those guilty of acts of spoliation during the interregnum
now brought to an end.
These provisions, taken as a whole, contain little to
justify Henry's claim to rank as a constitutional or national
sovereign. The bulk of the concessions are made to the
barons. The Church, it is true, obtains a definite promise
in chapter one : but the individuals who would chiefly
benefit were newly-appointed prelates, who became feudal
vassals on entering upon the lands of their sees. Chapters
2 and 4 confine their benefits to Crown-tenants and sub-
tenants, and are therefore purely feudal and not " national "
in their range. They may be compared with the clauses
of John's Charter that extend some of its provisions to
sub-tenants. Chapters 12 and 13, with their vague affirma-
tion of a firm peace, and of the old English law, now half-
forgotten (undefined and declared valid only so far as
unaltered by William I.), are the only grants "to the
people at large." The baronial element clearly triumphs/
over the " national," in the tenor and outlook of the famous
coronation charter.^
There are three intermediate links in the chain of charters
connecting those of Henry I. and John, namely, the two
charters of Stephen and that of Henry 11.^ The circum-
stances of the accession of the earlier King were peculiar.
Henry I. had nominated his only child Matilda as his heir :
his nephew, Stephen, and all the English barons had done
homage to her as their future liege lady. Stephen, how-
ever, taking advantage of Matilda's absence and unpopu-
larity, and of the barons' reluctance to be ruled by a woman,
made a bold dash for the Crown. From the moment of
the old King's death, " the Norman barons treated the suc-
^ See Charter in Appendix. For text and textual criticism, see Liebermnan,
Trans. R.H.S.^ VIII. 21-48. On whole subject, see Vinogradoff, Law Quart.,
Rev., as above cited.
2 The discussions on the "unknown charter" {infra, p. 175) would seem how-
ever, in another sense, to leave these three links out of the chain.
102 MAGNA CARTA: ITS FORM AND CONTENTS
cession as an open question " : in these words of Stubbs,^
Dr. Round finds 2 the keynote of the reign. Stephen
was prepared to bid higher for support than Henry had
felt compelled to do : like William of Orange, five centuries
later, he agreed to become " king upon conditions." A
Charter of Liberties and a solemn oath securing " the[^/
liberty of the Church " together formed the price of
Stephen's consecration; and this price was not perhaps
too high when we remember that " election was a matter
of opinion, coronation a matter of fact." ^ The process by
which he built up a title to the Crown culminated in the
Easter of 1136, when he secured the support of Matilda's
half-brother Robert, Earl of Gloucester, whose lead was
quickly followed by influential nobles who, however, per-
formed homage under an important reservation ; their
future loyalty would be strictly conditional on the treat-
ment extended to them by Stephen.
These transactions took place at Oxford;* at the same
time the King issued his second or Oxford Charter, which
combined the provisions of the oath to the Church and of
the vague earlier charter, with the conditions extorted by
Earl Robert and his followers. The opening words con-
tain a laboured attempt to set forth a valid title to the
throne. All reference to predecessors is avoided, and
Stephen declares himself king " by appointment of the
clergy and people, by consecration of the Archbishop and
papal legate, and by the Pope's confirmation. ^
Perhaps its chief provisions are those in favour of the
Church, supplementing a vague declaration that the Church
should be " free " by specific promises that the bishops
should have exclusive jurisdiction and power over ecclesi-
astics and their goods, with the sole right to superintend
their distribution after death. Here was a clear confirma-
»Stubbs, Const. Hist., I. 345. ^Rou^d, Geoffrey de Mandeville, p. i.
^ Round, Geoffrey de Mandeville, p. 6. Dr. Round, ibid., p. 438, explains that
this earlier charter of Stephen was supplemented by the verbal promise recorded
by William of Malmesbury, de libertate reddenda ecclesiae et conservanda.
* Round, Geoffrey, 22.
^Stephen was not justified in this last assertion. See Round, Geoffrey, 9.
ITS PROTOTYPES: EARLIER CHARTERS 103
tion of the right of the Courts Christian to a monopoly of \
all pleas affecting the clergy or their property. Stephen :
also renounced wardship over church lands during
vacancies — a surrender never dreamt of by Henry I. or i
Henry II. Grants to the people at large followed. A
general clause promising peace and justice was supple-
mented by specific concessions of more practical value,
namely, a promise to extirpate all exactions, unjust prac-
tices, and " miskennings " by sheriffs and others, and to
observe good, ancient, and just customs in respect of
murder-fines, pleas, and other causes.
Strangely enough, there is only one provision specially
benefiting feudal magnates, the King's renunciation of all
tracts of land afforested since the time of the two Williams.
The omission of further feudal concessions must not be
attributed either to Stephen's strength, or to any spirit of
moderation or self-sacrifice in the magnates. Each baron
of sufficient importance had already extorted a private
charter, more valued than a general provision in favour of
all and sundry. Such grants often included the right to
maintain a feudal stronghold, whose owner would enjoy
a position of practical independence.
It is instructive to compare these wide promises of
Stephen with the meagre words of the charter granted
by Henry of Anjou at or soon after his coronation. ^
Henry II. omits all mention of Stephen and his charters, \
not because he did not wish to acknowledge a usurper, but •
because of that usurper's lavish grants to the Church.
Henry had no intention to confirm "benefit of clergy" ini
so sweeping a form, or to renounce wardship over vacant|i
sees.
To the Church, as to the barons, Henry confirms only
what his grandfather had already conceded. Even com-
pared with the charter of Henry I., that of the younger
Henry is shorter and less explicit — features that justified
Stephen Langton in his preference for the older document.
If Henry II. granted a short and grudging charter, neither
of his sons, at their coronations, granted any charter at all.
*See Bemont, Ckar/es, 13, and Selecf Charters^ 135.
(;
104 MAGNA CARTA: ITS FORM AND CONTENTS
Reasons for the omission readily suggest tliemselves; the
Crown had grown strong enough to dispense with this
unwelcome formality, partly because of the absence of rival
competitors for the throne, and partly because of the per-
fection to which the machinery of government had been
brought. The utmost the Church could extract from
Richard and John, as the price of their consecration, was
the renewal of the three vague promises contained in the
oath, now taken as a pure formality. John, however, was
not to be allowed to shake himself free from the obligations
of his oath. Stephen Langton, before absolving him in
1 2 13, compelled him to swear it anew.^
Not only were the terms of the ancient oath riveted anew
on John's conscience, but, as has been shown, the corona-
tion Charter of Henry I., exhibited by the Primate in
times of crisis to the nobles, and used in preparing the
schedules that formed the rough drafts of the Articuli
Baronum, was made a curb for royal caprice. It is thus
impossible to neglect the importance of the sequence of
coronation oaths and charters as contributors both to the
form and substance of Magna Carta, although that is only
one of the many lines of descent through which the Great
Charter can trace its ancestry.
II. Magna Carta: its Form and Juridical Nature.
The juridical nature of the document to which John
set his seal at Runnymede will be differently estimated
according as it is judged by present-day or by medieval
standards.
(i) The Modern Point of View. Much ingenuity has
been expended in the effort to discover which particular
category of modern jurisprudence most accurately describes
the Great Charter. Is it an enacted law, or a treaty^ the
royal answer to a petition ; or a declaration of rights^ Is
it a simple pact,. bargain, or agreement between contracting
parties? Or is it a combination of two or more of these?
Something has been said in favour of almost every possible
^See supra, p. 27, and Round, En^. Hist. Rev., VIII. 292.
ITS FORM AND JURIDICAL NATURE 105
view, perhaps more to the bewilderment "than to the en-
lightenment of students of history uYiinterested in legal
subtleties.
The claim of Magna Carta to rank as a formal act of
legislation has been supported on the ground that it was
promulgated in what was practically a co^nmune concilium.
King John, it is maintained, met in a national assembly
all the estates of his realm who had any political rights,
and these concurred with him in the grant. The consent
of all who claimed a share in the making of laws — arch-
bishops, bishops, abbots, earls, and crown-tenants, great
and small — entitles the Charter to rank as a statute. ^-
Against this view, however, technical informalities may
be urged. Both the composition of the Council and the
procedure adopted there, were irregular. No formal writs
of summons had been issued, and, therefore, the meeting
was never properly constituted. Further, the whole pro-
ceedings were tumultuary ; the barons, assembled in mili-
tary array, compelled the consent of John by show of force.
On these grounds, modern jurisprudence, if appealed to,
would reject the claim of the Charter to be enrolled as an
ordinary statute. It may be argued also that Magna Carta,
while something less than a law, is also something more.
A law made by the king in one national assembly might
be repealed by the king in another: whereas the Great
Charter was intended by the barons to be unchangeable.
It was granted to them and their heirs for ever ; and, in
return, a price had been paid, namely, the renewal of their
allegiance.^
V' Magna Carta has also been described as a treaty. Such
is the verdict of Dr. Stubbs.^ " The Great Charter,
although drawn up in the form of a royal grant, was really
a treaty between the King and his subjects. ... It is the
collective people who really form the other high contracting
party in the great capitulation." ^ This view receives some
^ The (/m'd pro qzio was conditional homage, dependent (as we learn from
chapter 63) on observance of the Charter.
'^ Const. Hist., I. 569.
^ Cf. Prothero SiDion de Montfort, 15; Pike, House of Lords, 312.
io6 MAGNA CARTA: ITS FORM AND CONTENTS
support from words contained in chapter 63 of the Charter :
" Juratum est autem tarn ex parte nostra quam ex parte
baronum, quod haec omnia supradicta bona fide et sine
malo ingenio observabuntur." There is, however, a radical
objection. A treaty is a pubHc act between two contracting
powers, who must, to meet the requirements of modern
jurisprudence, be independent States or their accredited
agents; while John and his opponents were fragments of
one State, torn asunder by mutual jealousies.
For other authorities. Magna Carta is a contract, pact*
or private agreement. M. Emile Boutmy is of this
opinion : — " Le caract^re de cet acte est aise a d^finir. Ce
n'est pas pr6cis6ment un traits, puisqu'il n*y a pas ici deux
souverainetes legitimes ni deux nations en presence; ce
n'est pas non plus une loi; elle serait entachee d'irregu-
larit6 et de violence; c'est un compromis ou un pacte."^
Thus considered, the proudest act of the national drama
would take its place in the legal category which includes
the hire of a waggon or the sale of a load of corn. There
are, however, objections to this theory also. It is difficult
to see how the plea of " force," if sufficient (as M. Boutmy
urges) to render null the enactment of a public law, would
not be even more effective in reducing a private agreement.
If Magna Carta has no other basis than the consent of the
contracting parties, it seems safer to describe it as a public
treaty than a private pact.
Other theories also are possible; as, for example, that
the Great Charter is of the nature of a Declaration of
Rights, such as have played so prominent a part in France
and the United States ; while a recent American writer on
English constitutional development regards it as a code,
creating a formal constitution for England — in a rude and
embryonic form, it is true : — " If a constitution has for its
chief object the prevention of encroachments and the
harmonizing of governmental institutions, Magna Carta
answers to that description, at least in part." 2 It would
be easy to cite compromises between these competing
1 J^tudes de droit constikitionnel^ 41.
^Prof. Jesse Macy, English Constitution^ 162.
ITS FORM AND JURIDICAL NATURE 107
theories. Thus, a high authority declares that " the Great
Charter is partly a declaration of rights, partly a treaty
between Crown and people." 1
The essential nature of what took place at Runnymede,
in June, 1215, is plain, when stripped of legal subtleties.
A bargain was struck, between the King and his rebel
magnates, that, in return for a renewal of fealty and homage,
John would grant " to the freemen of England and their
heirs for ever" the liberties enumerated in sixty-three
chapters. No one thought of asking whether the trans-
action thus concluded was a " treaty " or a private " con-
tract." The terms had to be drawn up in legal form, so as
to bear record to the exact nature of the provisions, and
also to the authenticity of John's consent. It was, there-
fore, reduced to writing, and the resulting document was
naturally couched in the form invariably used for all irre-
vocable grants intended to descend from father to son,
namely, a feudal charter, authenticated by the impression
of the granter's seal — just as in the case of a grant of land,
and with many of the clauses appropriate to such a grant.2
John grants to the freemen of England and their heirs
certain specified rights and liberties, as though these were
so many hides of land. 3 The legal effect of such a grant
is hard to determine ; and insuperable difficulties beset any
attempt to expound its legal consequences in terms of
modern law.* In truth, the form and substance of Magna
^ Anson, Law of the Constitution, I. 14. Cf. Report on Dignity of a Peer,
I. 63, which makes it both a contract and a treaty.
2 In strict legal theory the complete investiture of the grantee required that
"charter" should be followed by "infeftment" or delivery (real or constructive)
of the subject of the grant. In the case of such intangible things as political
liberties, the parchment on which the Charter was written would be the natural
symbol to deliver to the grantees.
3 See chapter i. The grant which purports to be perpetually binding on John's
heirs, was in practice treated as requiring confirmation by his son.
* Prof. Maitland, Township and Borough, p. 76, explains some of the absurdities
involved : *' Have you ever pondered the form, the scheme, the main idea of
Magna Charta? If so, your reverence for that sacred text will hardly have
prevented you from using in the privacy of your own minds some such words as
* inept ' or ' childish,' etc."
io8 MAGNA CARTA: ITS FORM AND CONTENTS
Carta are badly mated. Its substance consists of a number
of legal enactments and political and civil rights ; its form
is borrowed from the feudal lawyer's book of styles for
conferring a title to landed estate. i
The results of this part of the inquiry seem, then, to be
mainly negative. It is misleading to describe phenomena
of the thirteenth century in modern phraseology which
would have been unintelligible to contemporaries. Yet,
I if it is necessary to make the attempt, Magna Carta may
! perhaps be regarded as an agreement partaking of the
(natures alike of a statute and a royal grant, of a public-
treaty and a private contract, yet identical with no one of
. these, but (in any view) enacting or proclaiming a number
'.of rules and customs as binding in England, and reducing
them to v/riting in the unsuitable form of a feudal charter
'.granted by King John to the freemen of England and their
heirs. 2
(2) The Contemporary Standpoint. It is perhaps more
profitable to enquire under what category of medieval
jurisprudence Magna Carta would have fallen, if its con-
temporaries had consciously attempted its classification.
In Dr. Vinogradoff 's phrase : " The best way to solve
these problems is perhaps to locate our document in
the pigeon-holes of medieval and not of modern rubri-
cation." ^
Answering his own question, he proceeds to range it,
partly as a unilateral grant by John to his subjects and
partly as of the nature of the medieval expedient known
to the continent of Europe as an " establishment " (stabili-
mentum or etahlissement). No exact definition of a stabili-
mentum need be expected from an age accustomed to a
vague use of words ; but its essence seems to have been
^Pollock and Maitland, I. 150, emphasize this disparity. **In form a donation,
a grant of franchises freely made by the king, in reality a treaty extorted from him
by the confederate estates of the realm, ... it is also a long and miscellaneous
code of laws." Cf. also Idi'd., I. 658.
— 2 See Prof. Adams [Origin, 212), who has a suggestive note on "the diplomatic
form of the Great Charter. "
^ Law Quarterly Review, XXI. 250-7.
ITS FORM AND JURIDICAL NATURE 109
a legislative act, more or less of an institutional and excep-
tional nature, affecting the general welfare of the country,
and thus requiring collective action by all classes or estates.
The elements of authority dispersed among the various
participants in legislative or sovereign power had to be
concentrated round the King, somewhat as the consent of
all first-class States has to be obtained at the present day
for effecting a change in the rules of International Law
observed by civilized nations.^
Legislative acts similar to the etablissements of Capetian
Kings were not unknown in England. The main purport
of the Statute of York (1322), for instance, according to
its latest interpreter,^ would seem to be that consent of
"the community" (or "commonalty," as it is usually
rendered), as well as of the prelates, earls and barons,
should be needed for any change of the nature of an
"establishment," which thus means an alteration in the
framework of government. Magna Carta contemplated
in chapter 61 an institutional innovation, parallels to which
may be found in the more or less oligarchical schemes of
1244, 1258, 1264 and 131 1. The historical importance of
such restrictions upon the method of legislation required
for changing the framework of government, lies in their
bearing on the development of a system of Estates and of
the future Parliament of the three Estates. ^
III. Magna Carta: its Contents and Characteristics.
The rights enumerated in the sixty-three chapters of the
Charter, representing the price paid by John for renewed
allegiance, are fully discussed in the main part of the
present volume : a brief description of their more pro-
minent characteristics, when viewed as a collective whole,
is, therefore, all that is here required.
As to externals, the want of orderly, logical arrangement
has often been noted. As John Reeves * says: "The
^ Cf. Vinogradoff, o/>. czL, who cites an example from a French ordinance
of 1223.
2G. Lapsley, jSn^. Hist Rev. XXVII., p. ii8.
3 Cf. Vinogradoff, op. cit. « History of Law, I. 266.
no MAGNA CARTA: ITS FORM AND CONTENTS
whole is strung together in a disorderly manner, with very
little regard to the subject matter " ; while a recent writer
maintains that " no portion of this famous document
can possibly be described as a good piece of drafting." i
Thirteenth-century standards, however, were different from
our own; and the lengthy document, with its specific
remedies for many abuses, contains evidence of a carefully
weighed scheme and of a deep-rooted conviction of the need
of reform. The barons and royal officials who helped in
framing it were ignorant of the abstract principles of
political science. Their ideas, it has been truly said, 2
" seem to have been concrete and practical, and in their
remedies they went no further than the correction of the
specific abuses from which they suffered." The framers
of the document observed (with few exceptions) great legal
accuracy in defining the traditional rights of the Crown, -
proceeding with praiseworthy moderation and scrupulous
fairness towards John.s
Three closely connected characteristics of the document,'
as a whole, will be brought out in the succeeding analysis :
Magna Carta is feudal, contractual, and (in parts, at least)
reactionary in tone. Professor Adams of Yale, giving
voice to opinions now widely admitted by historians,
emphasizes the crucial place occupied by " the^ feudal
contract " as the basis alike of Magna Carta and of the
medieval English constitutiorT; * and maintains that, from
the narrower point of view of 12 15, the essence of John's
Charter "in spirit, in method, and in principle," was
reaction. 5
In the attempt to analyze the leading provisions, various
principles of classification have been adopted : the chapters
may be arranged according to the functions of the central
government they were intended to limit ; according to their
own nature as progressive, reactionary, or declaratory;
and, finally, according to the classes of the community
which reaped the greatest benefit.
^Harcourt, Steward, 215. ^^^ja^ns, Origin, 250.
3 Adams, ibid., 256. ^Ibid., 150, 169, 203, 232.
^ Ibid., 249.
I
ITS CONTENTS AND CHARACTERISTICS iii
I. Provisions classified according to the prerogatives
affected.
Dr. Gneist ^ arranges the chapters in five groups accord-
ing as they place restraints (i) on the military power of the \r^
Crown, (2) on its judicial power, (3) on its police power, (4) C^
on its financial power, or (5) furnish a legal sanction for
the enforcement of the whole. In spite of Dr. Gneist's
high authority, it is doubtful whether an analysis of Magna
Carta upon these somewhat arbitrary lines throws much
light on its main objects or results. Such a division is
founded on distinctions not clearly formulated in the
thirteenth century, when the various functions of govern-
ment were still undifferentiated. 2
II. Provisions classified according as they are of a j^ro- ^-
^ressive, reactionary, or declaratory nature.
Blackstone,3 writing in 1759, expresses the generally
accepted views : " It is_ agreed by ^11 our historians that ^^
the Great Charter ofKing John was forjthejnost part com-
piled from the ancient customs^ of the realm, or the laws /{J^
of King Edward |Efi_Ccuife_ssor, by which they usually 7^
mean the commmijaw, which was established under our
Saxon princes, before the rigours of feudal tenures and ^
other hardships were imported from the continent." Sub-
stantially the same doctrine has been enunciated only the
other day, by our highest authority. " On the whole, the / ,
charter contains little that is absolutely new. It is restora-/^^^
tive. V John in these last years has been breaking the law ;| ^7
therefore the law must be defined and set in writing." ^ ^
This view seems, on the whole, a correct one : the insur-
gents in 12 15 professed to be demanding nothing new. l.
Yet the Great Charter contained much that was unknown \ Jf\
to the days of the Confessor and had no place in the pro- j
mises of Henry I.
Thus it is not sufficient to describe Magna Carta merely
^msL Engl. Const., Chapter XVIII.
2 Dr. Gneist indeed confesses this, when, in discussing the limitations of the
financial power, he admits that many of these are "already comprised in the
provisions touching the feudal power."
^ Great Charter, vii. * Pollock and Maitland, I. 151.
112 MAGNA CARTA: ITS FORM AND CONTENTS
as a declaratory enactment : it is necessary to distinguish
between the different sources of what it declared. A four-
fold division may be suggested, (i) Magna Carta handed
on some of the usages of the old English law unchanged
by the Conqueror or his successors, now conhrnied/and
pudfiedTTrom abuseSj~~T2)"In definingTeudal incidents and
services, it confirmed many rules of the feudal law brought '"
into England by^ the Normans after 1066. "(3) ~Tt also
embodied many provTslons of which William I. and even -^
Henry I. knew no more than did the Anglo-Saxon kings —
innovations introduced for his own purposes by Henry of
Anjou, but, after half a century of experience, now accepted
loyally even by the most bitter opponents of the Crown.
In the words of Mr. Prothero, " We find . . . the judicial,
and administrative system established by Henry II. pre-"'
served almost intact in Magna Carta, though its abuse was a)
carefully guarded against."^ Finally, (4) in some few
/points, the Charter aimed at going farther than Henry II.
I had intended to go: to mention only two particulars, the
*^ petty assizes are to be taken four times a year, while
/ sheriffs are prohibited from holding pleas of the Crown.
\ History, indeed, has proved that a purely declaratory
enactment is impossible : the mere lapse of time, by pro-
ducing an altered context, changes the purport of any
statute re-enacted in a later age. It is no unusual device
for innovators to render their reforms more palatable by
presenting them disguised as returns to the past. Further,
it is important to bear in mind the nature of the provisions
confirmed. A re-statement of some of the reforms of
Henry II. leads logically to progress rather than to mere^J
stability; while the confirmation of Anglo-Saxon usages
or of ancient feudal customs, fast disappearing under the
new regime, may imply retrogression rather than standing
still. Chapters 34 and 39 of Magna Carta, for example,
really demand a return to the system in vogue prior to the
innovations of Henry of Anjou, when they favour feudal
jurisdictions. Thus, some of the provisions of the Great
Charter which, at a casual glance, appear declaratory,
^ Simon de Montfort, 17.
ITS CONTENTS AND CHARACTERISTICS 113
are, in reality, innovations; while others tend towards
reaction.
III. Provisions classified according to the estates of the
community in whose favour they were conceived.
Here we are face to face with a fundamental question of
immense importance : Does the Great Charter really, as
the orthodox view so vehemently asserts, protect the rights
of the whole mass of humble Englishmen equally with
those of the proudest noble ? Or is it rather a series of
concessions to feudal selfishness wrung from the King by
a handful of powerful aristocrats ? On such questions,
learned opinion is sharply divided, although an over-
whelming majority of authorities range themselves on the
popular side, from Coke down to writers of the present day.
Lord Chatham, in one of his great orations,^ insisted that
the barons who wrested the Charter from John established'^
claims to the gratitude of posterity because they " did not /
confine it to themselves alone, but delivered it as a >
common blessing to the whole people " ; and Sir Edward j
Greasy ^ caps these words with more ecstatic words of his
own, declaring that one effect of the Charter was _^l_to_£iye
and to guarantee full protection for property and person /N
to every huma.n..berng that breathes English air." Staid
lawyers and historians like Blackstone and Hallam use
similar expressions. " An equal distribution of civil rights
to all classes of freemen forms the peculiar beauty of the
charter"; so we are told by Hallam. 3 Bishop Stubbs
unequivocally enunciated the same doctrine. " Clause by
clause the rights of the commons are provided for as well
as the rights of the nobles. . . . This proves, if any proof
were wanted, that the demands of the barons were no selfish
exactions of privilege for themselves." * "The rights
which the barons claimed for themselves," says John
Richard Green, ^ "they claimed for the nation at large."
^ House of Lords, 9th January, 1770. "^History of English Constitution, 151.
^ Middle Ages, \\. 447. ^ Const. Hist., I. 570- 1.
^ Short History, 124. Cf. Gneist, Const. Hist, (trans, by P. A. Ashworth), 253;
"A separate right for nobles, citizens, and peasants, was no longer possible." See
also Gneist, Hist, of Engl. Pari, (trans, by A. H. Keane), 103, and Hannis
Taylor, Engl. Const., I. 380.
H
114 MAGNA CARTA: ITS FORM AND CONTENTS
It would be easy to add to this "cloud of witnesses," but
enough has been said to prove that it has been a common
boast of Englishmen, for many centuries, that the provi-
sions of the Great Charter were intended to secure, and
did secure, the liberties of every class and individual, not
merely those of the feudal magnates.
It is a usual corollary to this theory, to attribute credit
to Stephen Langton for broad-minded statesmanship : the
so-called " Articles of the Barons " are really, it would
seem, articles of the archbishop. In Miss Norgate's words,
the original articles " are obviously not the composition of
the barons mustered under Robert Fitz- Walter, " who could
never have risen to " the lofty conception embodied in the
Charter — the conception of a contract between King and
people which should secure equal rights to every class and
every individual in the nation." i
It is not safe, however, to accept, without a careful con-
sideration of the evidence, opinions cited even from such
high authorities. " Equality " is essentially a modern
ideal : for many centuries after the thirteenth, class
legislation maintained its prominent place on the Statute
Rolls, and the interests of the various classes were by no
means always identical. A vigorous minority has always
protested against the popular view of Magna Carta. " It
has been lately the fashion," Hallam confesses, "to depre-
ciate the value of Magna Charta, as if it had sprung from
the private ambition of a few selfish barons, and redressed
- only some feudal abuses." 2
Two different parts of the Charter have a bearing on this
\ 1 question : chapter i, which explains to whom the rights
were granted; and chapter 61, which declares by whom
they were to be enforced. The liberties were confirmed
"to all freemen of my kingdom and their heirs for ever."
This opens the question — who were freemen in 1215 ? An
enthusiasm that seeks to enhance the merits of Magna
Carta by extending its provisions as widely as possible,
^ Norgate, /okn Lackland, 233.
"^Middle Ages, II. 447. See, e.g. Robert Brady, A Ftill and Clear Answer
(1683).
ITS CONTENTS AND CHARACTERISTICS 115
has led commentators to stretch the meaning of " freeman "
to embrace the entire population of England, including
not only churchmen, merchants, and yeomen, but even
villeins.
Now, homo in medieval law-Latin, was originally
synonymous with baro — all feudal vassals being described
as "men" or "barons." Magna Carta is a feudal grant, iv'
and the presumption is in favour of the technical feudal
meaning. The word, indeed, occasionally bore a looser,
wider sense; but any room for ambiguity seems to be
precluded by the use of the qualifying word " free." No
villein was fully a ^' liber homo." In chapter 34, for
example, the " liber homo" is assumed to be a landowner
with a manorial court. Even a burgess might not be
reckoned for all purposes as " free " ; for the Dialogus de
Scaccario discusses how far a miles or other liber homo
might lose his status by engaging in commerce in order
to make money. ^ The word "freeman," it would appear,' -
as used in the Charter is synonymous with " freeholder " ;
and therefore only a lirnited class could, as grantees or the
heirs of such, make good a legal clsiim to share in the
liberties secured by it.^ To the question, who^ had
authority to enforce its provisions, the Great Charter has
likewise a definite answer, namely, a quasi-committee of
twenty-five barons. It is clear that no support for demo-
cratic interpretations of Magna Carta can be founded on
'^Dialogus, II. xiii. c.
2 In addition to its appearance in the two places mentioned in the text, the
word ''freeman" appears in five other chapters, 15, 20, 27, 30, and 39. The
last three instances throw no light on the meaning of the word. It is different,
however, with chapter 15, where freemen are necessarily feudal tenants of a
mesne lord— that is, freeholders ; and with chapter 20, where, in the matter of
amercement, freeman is contrasted with villanus. Further, where men of servile
birth are clearly meant, they are described generally as probi homines {e.g. in
cc. 20, 29, and 48), and in one place, perhaps, c. 26, as legales homines.
Chapter 44 mentions homines without any qualification. It seems safe to infer
that the Great Charter never spoke of " freemen " when it meant to include the
ordinary peasantry or villagers. In chapter 39 of the reissue of 1217, liber homo
is clearly used as synonymous with "freeholder." In later centuries, it is true,
the "freeman" of the Charter came to be read in an ever less restricted sense,
until it embraced all Englishmen.
ii6 MAGNA CARTA: ITS FORM AND CONTENTS
the choice of executors; since these formed a distinctly
aristocratic body.
Magna Carta, indeed, contains positive evidences which
point to the existence of class legislation. At the begin-
ning and end of the Charter, clauses are inserted to secure
to the Church its " freedom " and privileges. Many
chapters, again, have no value except to landowners; a
few affect tradesmen and townsmen exclusively; while
chapters 20 to 22 adopt distinct sets of rules for the amerce-
ment of the ordinary freeman, the churchman, and the earl
or baron respectively. A distinction is made (for example,
in chapter 20) between the freeman and the villein, and the
latter was carefully excluded from many of the benefits
conferred on others by Magna Carta. ^
(i) The Feudal Aristocracy. A casual glance at the
clauses of the Great Charter shows how prominently feudal
rights and obligations bulked in the eyes of its promoters.
Provisions of this type must be considered chiefly as con-
cessions to the feudal aristocracy — although the relief,
primarily intended for them, indirectly benefited other
classes as well.
(2) Ecclesiastics. The position of the Church is easily
understood when we neglect the privileges enjoyed by its
great men qua barons rather than qua prelates. The
special Church clauses found no place whatsoever in the
Articles of the Barons, but bear every appearace of having
been added as an after-thought, due probably to the
influence of Stephen Langton.2 Further, they are mainly
confirmatory of the separate Charter already twice granted
within the few preceding months.
(3) Tenants and Mesne Lords. When compelling John
to grant Magna Carta by parade of armed might, the
barons were obliged to rely on the support of their own
freeholders. It was necessary that these under-tenants
should receive some recognition of their claims, and con-
cessions in their favour are contained in two clauses
(couched apparently in no generous spirit), chapters 15
and 60. The former limits the number of occasions on
^See zn/ra, under c. 39. ^Cf. sti/>ra, p. 39.
^
ITS CONTENTS AND CHARACTERISTICS 117
which aids might be extorted from sub-tenants to the same
three as were recognized in the Crown's case. Chapter 60
provides generally that all customs and liberties which
John agrees to observe towards his vassals shall be observed ^'^[
by mesne lords, whether prelates or laymen, towards their
sub-vassals. This provision has met with a chorus of
applause from modern writers. Dr. Hannis Taylor ^
declares that, " animated by a broad spirit of generous
patriotism, the barons stipulated in the treaty that every
limitation imposed for their protection upon the feudal
rights of the king should also be imposed upon their rights
as mesne lords in favour of the under-tenants who held of --
them." 2 A vague general clause, however, affords little
protection in a rude age and might readily be infringed
when occasion arose. The barons were compelled to do
something, or to pretend to do something, for their under- /- 7^
tenants. Apparently they did as little as they, with safety
or decency, could.
(4) Something was also done for the merchant and trad^^=^
ing classes. The existing privileges of London were con- "-'
firmed in the Articles of the Barons; and some slight
additions were made. An attentive examination suggests,
however, that these privileges were refined away in the
final form of Magna Carta. The right to tallage London
and other towns was reserved to the Crown, while the rights
of trading granted to foreigners were inconsistent with the
policy of monopoly dear to the hearts of the Londoners.
A mere confirmation of existing customs, already bought
and paid for at a great price, seems a poor return for
support given to the movement of insurrection at a critical
moment, when their adherence was sufficient to turn the
scale. The marvel is that so little was done for them.s
'^English Constitution^ I. 383.
2 Bishop Stubbs, Preface to W. Coventry^ II. Ixxi., represents the barons, in
their fervour for abstract law, as actually supporting their own vassals against
themselves : " the Barons of Runnymede guard the people against themselves as
well as against the common tyrant. "
2 For details, see infra under cc. 12, 13, 33, 35, and 41. Compare with the
corresponding Articles of the Barons (viz. 32, 23, 12, and 31). The alterations,
#
y
ii8 MAGNA CARTA: ITS FORM AND CONTENTS
L -^S) The relation of the villein to the benefits of the
Charter has been hotly discussed. Coke claims for him,
in regard to chapter 39 at least, that he must be regarded
as a liher homo, and therefore as a participant in the
advantages of the clause. ^ This contention, it has been
already shown, is not well founded. Yet the villein, it may
be argued, though excluded from participating in the rights
of freemen, has certain rights secured to him in his own
name. For example, in chapter 20, John promises that he
/' will not so cruelly amerce villeins — other people's villeins
at least — as to leave them utterly destitute.
(The villein was protected, however, not as the acknow-
ledged subject of legal rights, but because he formed a
valuable asset of his lord. 2 This attitude is illustrated by
V a somewhat peculiar expression used in chapter 4, which
\ prohibited injury to the estate of a ward by " waste of men
' or things." For a guardian to raise a villein to the status
of a freeman was to benefit the enfranchised peasant at
^ the expense of his young master. ^
I A Other clauses of John's Charter and of the various reissues
^j/L^show scrupulous care to avoid infringing the rights of
'^property enjoyed by manorial lords over their villeins.
v;.TheKin^ pf^^jHjT^ ^mejcjp. nth er people's villeins harshly,
although those on his own farms might be amerced at his '
discretion. Chapter 16, while carefully prohibiting any^j
\ arbitrary increase of service from freehold "property, leaves^
by inference all villein holdings unprotected. The " farms "
or rents of ancient demesne might be raised by the Crown,*
and tallages might be arbitrarily taken (measures likely to
press hardly on the villein class). The villein was deliber-
ately left exposed to the worst forms of purveyance, from
which chapters 28 and 30 rescued his betters. The horses
and implements of the villanus were still at the mercy of
slightly inimical to the towns, seem to show that the barons were more willing to
sacrifice their allies' interests than their own to John's insistence, when the final
terms were being haggled over.
^ See Coke, Second Institute^ p. 45, "for they are free against all men, saving
against their lord." Contrast ibid., p. 27.
2 Cf. under c. 20 infra. ^ Cf. under c. 4 infra. *■ See under c. 25 infra.
ITS CONTENTS AND CHARACTERISTICS 119
the Crown's purveyors. The reissue of 1217 confirms this
view : while demesne waggons were protected, those of
villeins were left exposed.^ Again, the chapter that takes
the place of the famous chapter 39 of 1215^ makes it clear
that lands held in villeinage are not to be protected from
arbitrary disseisin or dispossession. The villein was left )
by the common law merely a tenant-at-will — subject to^
arbitrary ejectment by his lord — whatever meagre measure j
of protection he might obtain under the "custom of the'
manor " as interpreted by the court of the lord who
oppressed him.
When taken together, the significance of these somewhat
trivial points is clear. The bulk of the English peasantry;-; N^
were protected by Magna Carta merely because they formed | \
valuable assets of their lords. The Charter viewed them as
" villeins regardant " — as chattels attached to a manor, not
as members of an English commonwealth. 3
The conclusion derived from this survey is that the
baronial leaders are scarcely entitled to the excessive lauda-
tion they have sometimes received. The rude beginnings
of features prominent later on (such as the conceptions of
patriotism, nationality, equality before the law, and tender p
regard for the rights of the humble) may possibly be found ^
in germ in parts of the completed Charter ; but the Articles
of the Barons were what their name implies, a baronial
manifesto, seeking chiefly to redress the private grievances /
of the promoters, and mainly selfish in motive.
Yet, when all deductions have been made, the Great
Charter stands out as a prominent landmark in the sequence
of events that have led, in an unbroken chain, to the con- ^ ^f
solidation of the English nation, and to the establishment
of a free and constitutional form of polity upon a basis so
^ See chapter 26 of 12 17. 2 See chapter 35 of 1217.
2 Dr. Stubbs takes a different view. Admitting that there is '* so little notice
of the villeins in the charter," he explains the omission on two grounds : (i) they >
had fewer grievances to redress than members of other classes ; (2) they partici- w
pated in all grants from which they were not specially excluded. ** It was not
that they had no spokesman, but that they were free from the more pressing
■grievances, and benefited from every general provision." Preface to JV. Coventry,
II. Ixxiii.
i2di MAGNA CARTA: ITS FORM AND CONTENTS
enduring that, after many centuries of growth, it still retains
— or, until a few years ago, did retain — the vigour and
buoyancy of youth.
4v J Magna Carta : an Estimate of its Value.
The importance of the Charter for the men of 12 15 did not
lie in what forms its main value for the constitutional
leorists of to-day. To the barons at Runnymede its merit
was that it was something definite and utilitarian — a legal u
document with specific remedies for current evils. To -
English lawyers and historians of a later age it became
something intangible and ideal, a symbol for the essential
principles of the English Constitution, a palladium of
^nglish liberties.
To trace the growth of these modern estimates lies outside
the scope of the present treatise ; but it should be noted that
admiration for John's Charter and its numerous reissues
and confirmations was moi^e measured among contem-
poraries than among its votaries of the seventeenth or nine-
teenth centuries; and that, for a long intervening period, it
suffered almost complete neglect.
There is some reason to suppose that the Carta Liber-
latum or Carta Baronum (as it is usually cited by contem-
porary authorities) was first described as " great " in the
reign of Henry III., and that it was then "great" mainly
in a material sense, a " large " charter as contrasted with a
certain parva carta granted by Henry in 1237.^
When, after many confirmations, the Charter had estab-
lished itself as a permanent part of the law of the land, it
seems to have fallen into the background of men's thoughts.
It played no conscious or conspicuous part in the " consti-
tutional experiments " of the Yorkist kings ; and friends of
popular liberties under the Tudors seem to have made few
appeals to its authority; Shakespeare's King John has
nothing to say of Runnymede or what happened there. 2 It
was during the struggles of Parliament with the first two
Stewart Kings and in part through the influence of Coke,
^See znfra, p. 157. '^See A. F. Pollard, Benry V/I/., 33 ff.
1
I
AN ESTIMATE OF ITS VALUE 121
with his strange combination of black-letter lore and enthu-
siasm for the old Constitution as interpreted by him, that
the Charter, now "great" in a sense higher than material^J^/^'
took hold oF the popular imagination. Thereafter esti-
mates of its worth steadily expanded. In many a time of
national crisis, Magna Carta has been appealed to as a
fundamental law too sacred to be altered — as a talisman ^
containing some magic spell, capable of averting national //
calamity* "^-Kl
Are these modern estimates of its value justified by facts, K
or are they gross exaggerations? Did it really create an I
epoch in English history? If so, wherein did its import-
ance exactly lie?
The numerous factors which contributed towards the ~7 ■
worth of Magna Carta may be distinguished as of two kinds,
inherent and accidental, (a) Its ^atllDi^C value depends ^-^
on the nature of its own provisions. The reforms demanded
by the barons were juSt and moderate : avoidance of
extremes tended towards a permanent settlement. Its aims
were practical as well as moderate; the language in which
they were framed, clear and straightforward. A high [
authority has described the Charter as " an^ Jntenjely 1
practical document."^ This practicability is an English ^
characteristic, and strikes the key-note of almost every great ^
movement for reform which has. succeeded in English ^
history. Closely connected with this is another feature, the
markedly Ipcrnl j^^tiArt^ of the Charter. As Magna Carta, -^
after Coke's day, was rarely absent from the thoughts of
statesmen, a practical and legal direction was thus given to
the efforts of Englishmen in many ages. 2 Therein lies
another English characteristic. While democratic enthusi-
astsin France and America have often sought to found their
liberties on a lofty but unstable basis of philosophical theory
embodied in Declarations of Rights; Englishmen have/.,.,
occupied lower but surer ground, aiming at practical /H^
1 Maitland, Social England, I. 409. '7
2Cf Gneist Const. Hist., Chapter XVIII. : *' By Magna Carta English history
irrevocably took the direction of securing constitutional liberty by administrative
law."
S MAGNA CARTA: ITS FORM AND CONTENTS
remedies for actual wrongs, rather than enunciating theo-
retical platitudes with no realities to correspond.
Further, the nature of the provisions bears witness to the
liroad„.basis on which the edifice was intended to be
^ilt. '/The Charter, notwithstanding the prominence given
to feudal grievances, redressed other grievances as well.|?.
.Another intrinsic merit was that it made definjte what had
►een vague before. Definition is aTaTuablej^rotection for "
'the weak agamst the strong : vagueness favours the tyrant
who can interpret wtiile tie enlorces'the law. Mistyri^its
were now reduced"tD--a:^t5Tigibleiorrnj^jnd ^^ longer
bebroken witn impunitj^. Where previously~TtTrTague-
ness of the law lent itself to evasion, its clear re-statement
in 1215 pinned down the King to a definite issue. He could J-
no longer plead that he sinned in ignorance ; he must either- f
keepTlhe law, or openly defy it — no middle course was
possible.
(b) Part of the value of Magna Carta may be traced to
ejitnnsi^^^uses ; to its vivid historical setting. The im-
portance of its provisions is emphasized by the object-
lessons that accompanied its inauguration. Christendom
was amazed by the spectacle of a King obliged to surrender
at discretion to his subjects. The fact that John was com-
pelled to accept the Charter meant a loss of royal prestige, .<?
and great encouragement to future rebels. What once had
happened, might happen again : the King's humiliation
was stamped as a powerful image on the minds of future
generations.
A separate treatise would be required for any serious
attempt to discuss the various estimates formed of Magna
Carta as viewed in successive centuries and in different
aspects. Some commentators have concerned themselves
mainly with individual clauses; others have treated it as
one whole. Historians look mainly to its immediate effects ;
lawyers and publicists to its ultimate consequences, as it
affected the development of the English law and Constitu-
tion.
(i) Value of Individual Provisions, To judge from the
reforms that attracted the notice of the only contemporary
AN ESTIMATE OF ITS VALUE 123
chronicler ^ whose opinion has come down to us, the clauses
considered of most impp^|ance in his day were those treat- _
ing of the "disparagement", of women, loss of life or ' .-
member for killing beasts of the forest, reliefs, the restora-
tion of seignorial jurisdiction (" hautes justices ") and the
appointment and powers of the twenty-five barons over the
King's government and over the appointment of bailiffs.
Some at least of these clauses are among those usually ^
considered reactionary, and there seems little doubt that the
barons in 12 15 were deeply interested in the restoration of
their feudal franchises, which Henry and his sons were
taking away from them. In the words of the French his-
torian, who was perhaps the first to sound the reaction from
the " traditional " view of Magna Carta : " The barons had I
no suspicion that they would one day be called the founders
of English liberty. . . . They were guided by a crowd of
small and very practical motives in extorting this form of —
security from John Lackland." 2
Of modern writers' estimates of the relative importance
of particular clauses it seems unnecessary to speak, as their
number and variety are great. ^
(2) Its Legal Value. The value of the Charter as a whole,
however, is more than a mere sum of the values of its
separate parts. Its great importance lay, not in the exact
terms of any or all of its provisions, but in the fact that it i
enunciated a definite bod jr of la\^^ claiming to be above the|— 'p
King's will and adrnltted as such by John. As our supreme
authorities say of Magna Carta: • "For in brief it means
this, that the King is, and shall be below the Jaw.'' * The /L
'^ B'ts^oire des dues {A.D. 1220), I4g-i^0. ^
2 Petit-Dutaillis, Louis VIIL, 58. C/. Adams, Origin, 249.
2 Prof. Adams {Origin, 176 n.) condenses its essence into three general rules.
Prof. Maitland {Collected Papers, II. 38), from a temporary angle of observation,
declares that " Magna Carta is an act for the amendment of the law of real
property and for the advancement of justice." John Lilburne {Just Man's
/ustification, p. 11) was also thinking of particular clauses when he wrote,
" Magna Carta itself being but a beggarly thing containing many marks of
intolerable bondage."
^Pollock and Maitland, I. 152. See, however, Petit-Dutaillis, Studies Supple-
vientary, 143 (criticising Pollock and Maitland): "That again, it seems to
124 MAGNA CARTA: ITS FORM AND CONTENTS
King, by granting the Charter, admitted that he was no^
an absolute ruler — that he had a master in the laws he had
often violated, but which he now swore to obey. Magna
Carta has thus been truly said to enunciate " the reign of
law " or " rule of law " in the phrase made famous by Pro-
fessor Dicey. 1
This conception of the existence of a definite body of
clearly formulated rights (now set down in the Charter in
black and white under John's seal), which the King was
bound to observe, was supplemented by the King's accep-
tance of the barons' claim to a right of compulsion. This
was a principle of abiding value, apart from any or all of
the clauses redressing specific grievances. " In the slowly
developing crisis of Henry III.'s reign, what men saw in
the charter in its bearing on their differences with the King
was not a body of specific law, but that the King's action
was bound and limited, and that the community possessed _
the right to coerce^him."^
(3) Its Value for the future Development of the Con-
stitutio7i. Magna Carta marks the commencement of a "^
new grouping of political forces in England; indeed, with-
out such a rearrangement, the winning of the Charter would
have been impossible. Throughout the reign of Richard I.
the unity of interests between King and lower classes had-
been endangered by the heavy drain of taxation ; but the
actual break-up of the old tacit alliance only came in the ^
crisis of John's reign. Henceforward can be traced a r
change in the balance of parties in the commonwealth. No
longer are Crown and people united, in the name of law and -
order, against the baronage, standing for feudal disintegra-
tion. The mass of humble freemen and the Church form a
league with the barons, in the name of law and order,
against the Crown, now the chief law-breaker.
Such an alliance involved the adoption of a new baronial
us, is to assign too glorious a r61e to the baronage of John Lackland and to
its political conceptions, which are childish and anarchical. The English
nobility of that day had not the idea of law at all."
^ A. V. Dicey, Law of the Constitution, Part II,
2 Adams, Origin, 251.
I
t
AN ESTIMATE OF ITS VALUE 125
policy. Hitherto each great baron had aimed at his own
independence, striving to gain new franchises for himself,
and to keep the King outside. This policy, which succeeded
both in France and Scotland, had before John's reign
already failed in England; and the English barons, now-
admitting the hopelessness of the struggle for feudal
independence, substituted a more progressive policy. The
King, whose interference they coufd no longer hope com-
pletely to shake off, must at least be taught to interfere
justly and according to rule; he must walk by law and .
custom, not by the caprices of his evil heart. The barons T"'
sought to control henceforward the royal power they could
not exclude : they desired some share in the national
councils, if they could no longer hope to create little
nations of their own within the four corners of their fiefs.
Magna Carta was the fruit of this new policy.
It has been often repeated, and with truth, that the Great
Charter marks also a stage in the growth of national unity |~""—
or nationality. Here, however, it is necessary to guard y
against exaggeration. It marks merely a stage in a pro-. !
cess, rather than a final achievement. It is necessary
somewhat to discount the Charter's claims to be " the first^^"'''
documentary proof of the existence of a united English
nation " and the often-quoted w^ords of Dr. Stubbs, that
" the Great Charter is the first great public act of the
nation, after it has realised its own identity." ^
A/ A united English nation, whether conscious or uncon-
scious of its identity, cannot be said to have existed in
1215, except under several qualifications. The conception
of " nationality," in the modern sense, is of recent origin,
and requires that the lower as well as the higher classes
should be comprehended within its bounds. Further, the
coalition which wrested the Charter from the royal tyrant
was essentially of a temporary nature, and quickly fell to
1 Cons^. Hist., I. 571. Cf. Ibid., I. 583, "The act of the united nation, the
church, the barons, and the commons, for the first time thoroughly at one." Who
were *' the commons " in 121 5 ? Cf. also Prothero, Simon de Montfort, 18, "The
spirit of nationality of which the chief portion of Magna Carta was at once the
product and the seal."
126 MAGNA CARTA: ITS FORM AND CONTENTS
pieces. Even while the alliance continued, the interests
of the various classes, as has been already shown, were
far from identical. Political rights were treated as the
monopoly of the few ; i and civil rights were far from
.universally distributed. The leaders of the "national"
; movement gave no political rights to the despised villeins,
«^ who comprised more than three quarters of the population
lof England; while their civil rights were almost completely
I ignored in the provisions of the Charter. Magna Carta
marked an important step, in the process by which England
became a nation ; but that step was neither the first nor yet
the final one. 2
In treating of the juridical nature of Magna Carta as
partly of the type of legislation known on the Continent
as an etahlissement, requiring all participants in political
power to be assembled round the King in order that they
might give consent, it has already been suggested that what
took place at Runnymede may have had an influence on
the development of the conception of a series of estates
and therefore on the genesis of the modern Parliament. 3
The Charter's greatest contribution, however, to constitu-
Itional advance lay undoubtedly in its admission (tacitly
limplied in its every clause) that the royal prerogative was
\ limited by the customary feudal rights of the bajfons^if
not of other classes as well).
In a sense there was nothing new in this : the feudal
relation, with its inherent conceptions of mutual, con-
tractual obligations and the rights of diffldatio and rebellion,
needed no official proclamation : it was known to all. But
the formal embodiment of a great mass of feudal custom
j?, in a document, destined to be consulted and reinterpreted
' I in future ages, created, as it were, a bridge between the
J older monarchy, limited by medieval, feudal restraints, and
•--X. the modern, constitutional monarchy, limited by a national
^ See infra, under c. 14.
2 The possibility that the movement leading to the Great Charter may have also
helped forward the growth of the idea of a separate national Church is discussed
infra, under c. one.
^ Supra, p. 109.
■-"t
p'
AN ESTIMATE OF ITS VALUE 127
law enforced by Parliament. This is the main thesis upon
which Professor Adams so emphatically insists, " the un-
intended result " of Magna Carta. ^ In light of it, he claims
to have located the origin of the English Constitution in
Magna Carta, and in these two principles of it which assert
the limitations of the King's prerogative and the barons'j^T^
right to compel him to respect the rights of others.
These estimates of the role played by Magna Carta would
seem to be somewhat excessive and to attempt to find too
simple an origin for a system of which complexity and
compromise between conflicting elements are the very
essence. On the one hand, there is more in the English
Constitution than the mere principle of limited monarchy :
on the other, the main line by which that monarchy has
progressed from medieval to modern ideals has not been
by the method, unsuccessfully attempted in 1215, 1244, 1258,
1265 and 131 1 (to name only the best-known instances), of
subjecting the King to the dictation of a Committee of his
adversaries ; but rather the method of using the counsellors
of his own appointment to curb his own caprice, while
making it progressively difficult for him to appoint any
minister of whom the national council did not approve.
The revolutionary expedient of the Committee of twenty-
five was not destined to be on the direct line of develop-
ment that led, through the doctrine of ministerial responsi-
bility, to the Cabinet system of government that reached
and perhaps passed its highest point of achievement in the
nineteenth century.
(4) lis Moral or Sentimental Value, After every allow-
ance has been made for the great and beneficent influence
of Magna Carta, it may still be doubted whether the belief
of enthusiasts in its excessive importance has been fully
justified. Many other triumphs, almost equally important,
have been won in the cause of liberty ; and statutes have
been passed embodying them. Why then should Magna
Carta be extolled as the palladium of English liberties?
Is not, when all is said, the extreme merit attributed to it
mainly of a sentimental or imaginative nature ? Such
^ Adams, Origin, 250.
128 MAGNA CARTA: ITS FORM AND CONTENTS
questions must be answered partly in the affirmative.
Ij^ Much of its value does depend on sentiment. Yet allj
government is, in a sense, founded upon sentiment — some-
times affection, sometimes fear: psychological considera-i
tions are all-powerful in the practical affairs of life.i
Intangible and even unreal phenomena have played an
important part in the history of nations. The tie that binds
the British colonies at the present day to the Mother
Country is largely one of sentiment ; yet the troopers from
Canada and New Zealand who responded to the call of
Britain in her hour of need produced practical results of
y an obvious nature. The element of sentiment in politics
^ can never be ignored.
It is no disparagement of Magna Carta, then, to confess
that part of its power has been read into it by later genera-
tions, and lies in the halo, almost of romance, that has
■gathered round it in the course of centuries. It became a "~
battle cry for future ages, a banner, a rallying: point,
a stimulus to the imagination. For a King, thereafter, s
openly to infringe the promises contained in the Great
Charter, was to challenge public opinion — to put himself T"
palpably in the wrong. For an aggrieved man, however
humble, to base his rights upon its terms was to enlist the
sympathy of all. Time and again, from the Barons' War
against Henry III. to the days of John Hampden and
Oliver Cromwell, the possibility of appealing to the words
of Magna Carta has afforded a practical ground for opposi-
f t'lon ; an easily iirtenigTBTe pfmcTpTFtolfigh for ; a fortified -^
'^ posftloh to Traid^^gainst the enemies of nafional freedom.
To explain tEe exact way TiiwFichtHTs^^^^^^ document
— dry as its details at first sight may seem — has fired the
popular imagination, is a task that lies rather within the
sphere of psychology than of history, as usually conceived.
How^ever difficult it may be to explain this phenomenon,
there is no doubt of its existence. The importance of the
Great Charter has increased, as traditions, associations, and
aspirations have clustered more thickly round it.
Thus Magna Carta, in addition to its legal and political
value, has a moral value of an equally emphatic kind.
MAGNA CARTA: ITS DEFECTS 129
Apart from and beyond the salutary effect of the useful
laws it contains, its moral influence has contributed to aii")
advance in the national spirit, and therefore in the national y
liberties. Such considerations justify enthusiasts, who
hold that the granting of Magna Carta was the turning-
point in English history. '
V. Magna Carta. Its defects.
The great weakness of the Charter was the absence of an I --'''
adequate sa.nction. The only expedient for compelling the { '-^1
King to keep his promises was clumsy and revolutionary ;
quite worthless as a working scheme of government. ^..
Indeed, it was devised not so much to prevent the King^ /
from breaking faith as to punish him when he had done so. /
In other words, instead of constitutional machinery to turn /
the theories of Magna Carta into realities, " a right of /
legalized rebellion " was conferred on an executive com-
mittee of twenty-five of the King's enemies.^
This is the chief defect, but not the only one. Many
'minor faults and omissions may be traced to a similar root.
Constitutional principles are conspicuously absent. The
importance of a council or embryo parliament, framed on
national lines; the right of such a body to influence the
King's policy in normal times as well as in times of
crisis ; the doctrine of ministerial responsibility ; the need
of distinguishing the various functions of government,
legislative, judicial, and administrative — all these cardinal
principles are completely ignored. Only five of the sixty-
three chapters can be said to bear directly on the subject)^
of constitutional (as opposed to purely legal) machinery,
and these do so only incidentally, namely, chapters 14, 21, j
39, 52, and 61. *
The Commune Concilium, is indeed mentioned; and its
composition and mode of summons are defined in chapter
14. But this chapter appears as an afterthought — an
appendix to chapter 12: it has no counterpart in the
Articles of the Barons. The rebel magnates were interested
^ See tn/ra, c. 6r, for details.
I
130 MAGNA CARTA: ITS FORM AND CONTENTS
in the narrow question of scutage, not in_the_wide possi-
Dilities involvt^d m the existence of a national counciL
j TEe~CommwnF C'ona/twm was draped into the Charter,
not on its own merits, but merely as a convenient method-
f of preventing arbitrary increase of feudal exactions. This
is further proved by the omission of the Council from the
reissue of 12 17, when an alternative way of checking the
increase of scutage had been devised.
If the framers of John's Magna Carta had possessed the
grasp of constitutional principles, with which they have
been sometimes credited, they would have seized the oppor-
tunity afforded them by .the mention of the Common
Council, in chapters 12 and 14, to define carefully the
A powers they claimed for it. On the contrary, no list of its
^'functions is drawn up; nor do the words of the Charter
contain anything to suggest that it exercised powers other
than that of consenting to scutages and aids. Not a word
is said of any right to a share in legislation, to control or^
even to advise the Executive, or to concur^in choosing
the great ministers of the Crown. Neither deliberative,
administrative, nor legislative powers are secured to it,
while its control over taxation is strictly limited to scutages
. and aids — that is to say, it only extends over the exactions
Y that affected the military tenants of the Crown. It is true
that chapters 21 and 39 may possibly be read as confirming
the judicial power of the Council in a certain limited group
of cases. Earls and barons are not to be amerced except
by their peers (per pares suos)^ and the natural place for
these " equals " of a Crown vassal to assemble for this
purpose would be the Commune Concilium. This, how-
ever, is matter of inference; chapter 21 makes no mention
of the Council ; and it is equally possible that its require-
ments would be met by the presence among the officials
of the Exchequer of a few Crown tenants.^ Similar reason-
ing applies to the provisions of chapter 39 (protecting
persons and property of freemen, by insisting on the
necessity of a " trial by peers ") so far as they affect earls
and barons.
^ This is the view of Pike, J7ouse of Lords ^ 204. See infra^ c. 21.
MAGNA CARTA: ITS DEFECTS 131
It is clear that the leaders of the opposition in 12 15 did
not consider the constitutional powers of a national Parlia-
ment to be the best safeguard of the rights and liberties
theoretically guaranteed by the Charter. They relied
rather on the revolutionary powers of the twenty-five
barons to be appointed under chapter 61.
The same inability to devise practical remedies may be
traced in minor clauses of the Charter. ^ When John pro-
mised in chapter 16 that no one should be compelled to do
greater service than was due, no attempt was made to
provide machinery to define such service; while chapter 45,
providing that only men who knew the law and meant to
keep it, should be made justiciars, sheriffs and bailiffs,
laid down no criterion of fitness, and contained no sugges-
tion of the way in which so laudable an ambition might
be realized.
Thoughtful and statesmanlike as were many of the pro-
visions of Magna Carta, and wide as was the ground they
covered, important omissions can be pointed out. Some
crucial questions seem not to have been foreseen ; others,
for example the liability to foreign service, were deliber-
ately shelved 2 — thus leaving room for future misunder-
standings. The praise, justly earned, by its framers for
the care and precision with which they defined a long list
of the more crying abuses, must be qualified in view of ... '
the failure^tojproyide procedure to prevent their recurrence.
Men had not yet learned the force of the maxim, so closely \
identified with all later reform movements in England, that j
a right is valueless without an appropriate remedy to
enforce it.^
1 Magna Carta has been described, in words already quoted with approval, as
"an intensely practical document," Maitland, Social En gla7id, I. 409; but this
requires qualification. If it was practical in preferring condemnation of definite
grievances to enunciation of philosophical principles, it was unpractical in omitting
machinery for giving effect to its provisions.
2 Except in so far as affected by cc. 12 and 16.
3 Mr. Prothero estimates more highly the constitutional value of Magna Carta :
" The constitutional struggles of the following half-century would to a great extent
have been anticipated had it retained its original form. " Simon de Mont/ori, 14.
:32 MAGNA CARTA: ITS FORM AND CONTENTS
V./j Magna Carta : Value of Traditional Interpretations.
The Great Charter has formed a favourite theme for
orators and politicians, partly from its intrinsic merit,
'7 partly from its dramatic background, but chiefly because
ij^^ it has been, from the time of its inception to the present
'*^'^ day, a rallying cry and a bulwark in every crisis that
threatened to endanger the national liberties.
The uses to which it has been put, and the interpreta-
tions read into it, are so numerous and varied, that they
would require a separate treatise to describe them all. Not
only was Magna Carta frequently reissued and confirmed,
but its provisions have been asserted and reasserted times
without number in Parliament, in the courts of justice, and
in institutional works on jurisprudence. Its influence has
thus been threefold ; and any attempt to explain its bearing
on the subsequent history of English liberties would require
to distinguish between these three separate and equally
important aspects: — (i) It proved a powerful weapon in
/ the hands of politicians, especially of the parliamentary
leaders in the seventeenth century, when waging the battle
of constitutional freedom against the Stewart dynasty.
(2) Its legal aspect has been equally important as its
^ political one : in the course of legal debate and in judicial
opinions, it has been the subject of many and conflicting
interpretations, some of them accurate and some erroneous.^
(3) Finally, it has been discussed in many commentaries
either exclusively devoted to its elucidation or treating of
it incidentally in the course of general expositions of the
law of England.
In light of the part played by Magna Carta throughout
centuries of English history, it is not surprising that an
increasing veneration has tended at times to overstep all
bounds. It is unfortunate, however, that it has been more
frequently described in terms of inflated rhetoric than of
sober methodical analysis. 2 Nor has this tendency to un-
* As early as 1231 the ** carta de Runemede " was cited in a plea. See Bracton's
Notebook^ No. 513. See also No. 1478, dating from 1221 ; others in Index.
2 Extravagant estimates of its value will readily suggest themselves. Sir James
Mackintosh {History of England^ I. 218, edn. of 1853) declares that we are
VALUE OF TRADITIONAL INTERPRETATIONS 133
thinking adulation been confined to popular writers ; judges
and institutional authors, even Sir Edward Coke, have too
often lost the faculty of critical and exact scholarship when
confronted with the virtues of the Great Charter. There
is scarcely one great principle of the modern English
constitution calculated to win the esteem of mankind,
which has not been read by commentators into Magna
Carta. The political leaders of the seventeenth century 1
discovered among its chapters every reform they desired J
to introduce into England, disguising revolutionary pro- ]
jects by dressing them in the garb of the past.
Instances of constitutional principles and institutions
erroneously credited to the Great Charter will be expounded
under appropriate chapters of the sequel. It will be suffi-
cient, in the meantime, to enumerate trial by jury ; Habeas
Corpus; abolition of arbitrary imprisonment; prohibition
of~monopolfes'X~'the close tie between taxation and repre-
sentation; equality before the law; a matured conception
of nationality : all these, and more, have been discovered
in various clauses of the Great Charter.^
'^ If these tendencies to excessive and sometimes ignorant
praise have been unfortunate from one point of view, they
have been most fortunate from another. The legal and
political aspects must be sharply contrasted. If the vague
and inaccurate words of Coke have obscured the bearing
of many chapters, and diffused false notions of the develop-
ment of English law ,'^ the service these very errors have
done to the cause of constitutional progress is measureless.
If political bias has coloured interpretation, the ensuing
benefit has accrued to the cause of national progress in its
widest and best developments. ^■-
" bound to speak with reverential gratitude of the authors of the Great Charter.
To have produced it, to have preserved it, to have matured it, constitute the im-
mortal claim of England upon the esteem of mankind. Her Bacons and Shake-
speares, her Miltons and Newtons, etc., etc."
^Edmund Burke ( IVor/es, II. 53) credits Magna Carta with creating the House
of Commons! "Magna Charta, if it did not give us originally the House ofJ" ^^y-/"
Commons, gave us at least a House of Commons of weight and consequence." As f
will be shown in the sequel, chapter 14 of the Great Charter (the only one bearing
on the subject) is in reality of a reactionary nature, confining the right of attendance
at the commune conciliuvi to the freeholders of the Crown.
MAGNA CARTA: ITS FORM AND CONTENTS
'hus the historian of Magna Carta, while bound to
correct errors, cannot afford to despise traditional interpre-
tations. The meanings read into it by learned men have
had a potent effect whether they were historically well or
ill founded. The stigma of being banned by the Great
Charter was something to excite dread. If the belief pre-
vailed that an abuse was really prohibited by Magna Carta,
the most arbitrary king had difficulty in finding judges to
declare it legal, or ministers to enforce it. The prevalence
of such a bf^ljpf ^^as the main point; whether it was well
or ill founded was, for political purposes, immaterial.
(The greatness of Magna Carta lies not so much in what
it was to its framers in 12 15, as in what it afterwards
became to the political leaders, to the judges and lawyers,
and to the entire mass of the men of England in later ages.
VII. Magna Carta. Its traditional relation to Trial by-
Jury.
One persistent error, adopted for many centuries, and
even now hard to dispel, is that the Great Charter
guaranteed trial by jury.^ This belief is now held by all
competent authorities to be unfounded. Not one of the
three forms of a modern jury trial had taken definite shape
in 1 2 15, although the root principle from which all three
subsequently grew had been in use since the Norman
Conquest.2
Jury trial in each of the three forms in which it is known
to modern English law (the grand jury, the^etty criminal
jury, and the jury of civil pleas^TTs able to trace an unbroken
pedigree (though by three distinct lines of descent) from
the same ancestor, that principle known as recognitio or
inquisitio, which was introduced into England by the
Normans, and was simply the practice whereby the Crown
1 The source of this error was the identification of jury trial with the judicium
J>arium oic. 39. q.v.
'^Yox the origin of the jury see Brunner, Schurgerichte (1871) : Haskins, Avi.
His. Rtv.^V\\\. 613 fiF., traces the steps made towards the civil jury in Normand)',
particularly under Henry's father, Geoffrey.
RELATION TO TRIAL BY JURY 135
obtained information on local affairs from the sworn testi-
mony of local men. While thus postulating a foreign
origin, we are afforded some consolation l)y "remembrance
of a fact which modern authorities are inclined to neglect,
namely, that the soil was prepared by Anglo-Saxon labour
for its planting. 1
The old English institution of frithborh-;;^the practice of
binding together little groups of neighbours for preservation
of the peace — and the custom of sending representatives of
the villages to the Hundred Courts, had accustomed the
natives to corporate action, and formed precedents for
asking them to give evidence on local matters jointly and
on oath. Further, one form of the jury-— the jury of
accusation — is clearly foreshadowed by the directions given
to the twelve senior thegns of each Wapentake by a well-
/known law of Ethelred. Yet the credit of establishing the
jury system as a fundamental institution in England is
undoubtedly due to the Norman and Angevin kings,
although they had no clear vision of the consequences of
what they did. The uses to which Inquisitio was put by
William and his sons in framing Domesday Book, collect-
ing information, and dispensing justice, have already been
discussed.2 It was reserved for Henry II. to start the
institution on a further career of development : he thus laid
the foundations of the modern jury system jiQt merely in
one of its forms, but in all three of them.
(i) In reorganizing machinery for the suppression and
punishment of crime by the Assizes of Clarendon and
Northampton, he established the principle that criminal
trials should (in the normal case) begin with indictnient of
the accused by a representative body of neighbours sworn
^ The theory now generally accepted that the origin of trial by jury must be
sought in procedure introduced by Norman dukes, not in any form of popular
Anglo-Saxon institutions, is ably maintained by Pollock and Maitland, I. 119,
and by the late Professor J. B. Thayer, Evidence^ p. 7. Undoubtedly their
conclusions are in the main correct ; but trial by jury may have had more than
one root, and appreciation of the Norman contribution need not lead to neglect
of the Anglo-Saxon. See, e.g. Hannis Taylor, English Constitution^ I. 308 and
I. 323 ; Vinogradoff, Growth of the Manor ^ 193 : * something more than a
Norman device."
* See supra, p. Z(>.
136 MAGNA CARTA: ITS FORM AND CONTENTS
to speak the truth. i This was merely a systematic enforce-
ment of one of the many forms of inquisitio already in use :
criminal prosecution was not to be begun on mere suspicion
or irresponsible complaints. The jury of accusation (or
presentment), instituted in 1 166, has continued in use ever
since, passing by an unbroken development into the grand
jury of the present day. 2
(2) Py insisting that ordeal should supersede compurga-
tion as the test of guilt or innocence, Henry unconsciously
prepared the way for a second form of jury. When the
fourth Lateran Council in the very year of Magna Carta
forbade priests to countenance ordeal by their presence or
blessing, a death-blow was dealt to that form of procedure
or "test," since it depended for its authority on supersti-
tion. A canon of the Church had struck away the pivot
on which Henry had made his criminal system to revolve.
Some substitute was urgently required and so the petty
jury (or its rude antecedent) came into existence. The man
publicly accused as presumably guilty was asked if he
would stand or fall by reference to the verdict of a second
jury of neighbours. This second verdict, then, was the
new " test " or " law " substituted, if the accused man
agreed, for his old right of proving himself innocent by
ordeal. By obscure steps, on which those best entitled to
speak with authority are not yet agreed, this jury, giving
a second and final verdict, gradually developed into the
criminal jury of twelve, the petty jury of to-day, which has
had so important an influence on the development of con-
stitutional liberties in England, and even on the national
character. Another expedient of Henry's invention aided
the movement towards the criminal jury, namely, the writ
de odio et atia by applying for which a man " appealed "^
^ See Pollock and Maitland, I. 131. It was part of Henry's policy to substitute
indictment by a representative jury for the older appeal by the wronged individual
or his surviving relatives. The older procedure, however, was not completely
abolished : its continuance and its unpopularity may be traced in chapter 54 of
Magna Carta, c/.v.
'Chapter 38 of Magna Carta, according to a plausible interpretation of an
admittedly obscure passage, seems to insist on the necessity of such an accusation
by the jury : — *' non . . . sine testibus Jidelibus ad hoc itidnctis."
RELATION TO TRIAL BY JURY 137
of a crime might substitute what was practically a jury's
verdict for the " battle " which had previously followed
" appeal " as matter of course. ^
(3) The Civil Jury owes its origin to quite a different
set of reforms, though inaugurated by the same reformer.
Among the evil legacies from Stephen's reign, not the
least troublesome were the claims advanced by rival
magnates to estates and franchises which had been
bestowed with lavish prodigality by Matilda and Stephen.
Henry realized the urgent need of protecting vested
interests by a more rational expedient than trial by combat.
Here again he had recourse to a new development of
" inquisition." In such cases an option was given to the
tenant (the man in possession) to refer the question at issue
to the verdict of local recognitors.
This new expedient was applied at first only to a few
special cases. It was used to settle claims of ultimate title
— the out-and-out ownership of land — and then it was
known as the Grand Assize ; it was also used to settle
a few well-defined groups of pleas of disputed possession,
and then it was known as a Petty Assize (of which there
were three distinct varieties) .2 The King by a high-
handed act of power deprived the demandant of that remedy
which was his right by feudal law, the resort to the legal
duel. It was because the new procedure was founded on
a royal Ordinance, that the name " Assize " was applied
to it.
By consent of hoik parties, however, disputes of almost
every description might be similarly determined; being
referred (under supervision of the King's judges) to the
verdict of local recognitors, usually twelve in number, who
were then known as a jurata (not an assisa). While the
assisa was narrowly confined to a few types of cases, the
'jurata was a flexible remedy capable of indefinite expansion,
and thus soon became the more popular and the more
important of the two. Sometimes the twelve recognitors,
summoned as an assisa by the King's command, were by
^ For details see tn/ra under chapter 36, and supra, p. 89.
'The three Petty Assizes are mentioned by name in c. 18, q.v.
138 MAGNA CARTA : ITS FORM AND CONTENTS
consent of both litigants turned into a jurata to try a
broader issue that had unexpectedly arisen. This explains
the phrase, assisa vertitur ad juratam. The assisa and
jurata, always closely connected and resembling each other
in essential features, can both claim to be ancestors of the
modern civil "jury," — the name of the more popular
institution having survived. Magna Carta, in providing
for the frequent holding of the three petty assizes, marked
a stage in the development of the Civil Jury; while, by
enforcing the criminal procedure of Henry Plantagenet,
and guarding it from abuse, the Charter had also a vital
bearing on the genesis of the Grand Jury and the Petty
Jury alike.
These tentative measures, however, still vague and un-
consolidated, must not be identified with the definite
procedure into which at a later date they coalesced : Magna
Carta did not promise " trial by jury " to anyone.
PART IV.
HISTORICAL SEQUEL TO MAGNA CARTA
I. Reissues and Confirmations of the Great Charter.
King John had accepted the reforms contained in Magna
Carta unwillingly and insincerely; but the advisers of
his son accepted them in good faith. ^Three reissues of "^
the Charter were granted in 1216, in 12 17, and in i225,j) -
and these were followed by many confirmations. The i\
scheme of this Historical Introduction is restricted to such v
facts as have direct bearing on the genesis and contents
of John's Charter. Yet no account of Magna Carta would
be complete without some notice of the more important
alterations effected in its text during the reigns of later
kings.
(I.) Reissue of 1216. On 28th October, 1216, Henry of
Winchester was crowned at Gloucester before a small
assemblage. 1 The young King took the usual oath as
directed by the Bishop of Bath, and he also performed
homage to the Pope's representative Gualo ; for the King
of England was now a vassal of Rome.^ At a Council held
at Bristol, on nth November, William Marshal, Earl of
Pembroke, was appointed Rector regis et regni; and, next
day, the Charter was reissued in the King's name. This
was a step of extreme importance, marking the acceptance
by those in power for the time being of the programme of
the baronial opposition.
^ See Annals of Waverley, p. 286.
2 For the question of the Regency and the position of England as a fief of Rome,
see Norgate, Minority, 10-62 ; Turner, Trans. R.H.S. (1904), 268 ff. In a plea
roll of 1237 (Bracton's Notebook, No. 12 19) Gualo is described as ^* quasi tutor
dontini regis et custos regni.^^
140 HISTORICAL SEQUEL TO MAGNA CARTA
The Charter in its new provisional form was really a
manifesto issued by the moderate men who rallied round
the throne of the young King; it may be viewed in two
aspects, as a declaration by the Regent and his co-adjutors
of the policy on which they accepted office, and as a bid
for the support of the barons who still adhered to the faction
of the French prince. Its issue was, indeed, dictated by
the crucial situation created by the presence in England of
Prince Louis of France, supported by a foreign army and
by a large faction of the English barons who had sworn
homage to him as King. It was, therefore, framed in
terms meant to conciliate such of the opposition as were
still open to conciliation.
Yet the new Charter could not be a verbatim repetition
of the old one. Vital alterations were required by altered
circumstances.^ It was no longer an expression of reluc-
tant consent by the government of the day to the demands
of its enemies, but a set of rules deliberately accepted by
that government for its own guidance. The chief tyrant
against whom the original provisions had been directed
was dead, and certain forms of tyranny, it was confidently
hoped, had died with him. Restraints now placed on the
Crown's prerogatives would only hamper the free action of
the men who framed them, not of their political opponents.
The Regent, while willing to do much for the cause of
conciliation, could not afford to paralyze his own efficiency
at a time when foreign invaders were in possession of one-
half of England, from which it would require a supreme
effort to dislodge them. It was imperative that the govern-
ment should retain a free hand in exacting feudal services
and levying scutages.
Miss Norgate argues,2 somewhat unconvincingly, that
the omission of chapters 12 and 14 was a concession to
Gualo and the Holy See. Rome had regarded these pro-
^ The cause for wonder is rather how few changes required to be made. *' It is,
however, by no means the least curious feature of the history, that so few changes
were needed to transform a treaty won at the point of the sword into a manifesto of
peace and sound government." Stubbs, Const. Hist., II. 21.
^Minority, 15.
REISSUES AND CONFIRMATIONS 141
visions as dangerous innovations of so marked a kind as
to justify the annulling of the Charter of 12 15, and papal
sanction could be obtained in 12 16 only by their jettison.
William Marshal, however, is not likely to have required
external pressure : he naturally preferred to leave his own
hands untied.
Yet the issue of the Charter under papal sanction, how-
ever obtained, was of material value to Henry's cause.
It had the immediate effect of bringing over eleven bishops
to the young King's side. M. Petit-Dutaillis ^ sums up
the situation in two propositions : the French invasion
saved the Great Charter, and then papal support saved
England from Louis. ^
The Charter of 1216^ is notable for its omissions, which
may be arranged under five groups.* (i) Restraints placed
in 1 2 15 on the taxing power of the Crown now disappeared.
The chapters which forbade the King to increase the
" farms " or fixed rents of the counties and hundreds, those
which defined the King's relations with the Jews, and those
which restricted the lucrative rights derived from the
rigorous enforcement of the forest laws, were discarded.
An even more important omission was that of chapter 12,
which abolished the Crown's rights to increase feudal con-
tributions arbitrarily, without consent of the Common
Council.^
(2) No reference is made to John's charter of May, 12 15
to the Church, granting liberty of elections, although the
vague declaration that " the English Church should be
free " was retained. Chapter 42, allowing liberty to leave
the kingdom, and to return without the King's consent (a
privilege chiefly valuable to the clergy in their intercourse
with Rome) was entirely omitted : and the same is true of
^ See Petit-Dutaillis, Louis, 130-I. ^Ibid., 181.
^ In the Appendix, an attempt is made to show at a glance the main differences
between the various Great Charters.
*This classification takes no account of alterations merely verbal or inserted to
remove ambiguities, e.g, cc. 22, 28, and 30 of the original Charter.
*See, however, Adams {Origin, 261 ; 220), who maintains that the omission
was not intended to leave the Crown a freer hand (whatever might be the practical
result).
142 HISTORICAL SEQUEL TO MAGNA CARTA
chapter 27, which had placed in the church's hands the
supervision of all distributions of chattels of men who had
died intestate.^
(3) A great number of provisions of purely temporary-
interest disappeared, among them those providing for dis-
bandment of mercenary troops and dismissal from office of
obnoxious individuals.
(4) A number of omissions of a miscellaneous nature may
be grouped together; for example, chapter 45, by which
the Crown restricted itself in the choice of justiciars and
other officers ; the latter half of chapter 47, relating to the
banks of rivers and their guardians; and some of the pro-
visions affecting the forest laws.
(5) These alterations implied, incidentally rather than
deliberately, the omission of such constitutional machinery
as had found a place in John's Great Charter. The twenty-
five Executors fell with the other temporary provisions ;
while chapter 14, which defined the composition and mode-
of summons of the Commune Concilium, was omitted with
chapter 12, to which it had formed a supplement. 2
Magna Carta as granted by Henry is thus concerned
with matters which lie within the sphere of private law,
and contains no attempt to devise machinery of govern-
^ Are these omissions mainly accidental ? Are they the result of some influence-
at work hostile to English ecclesiastics ? Or, are they due to the personal wishes,
and ambitions of Gualo ? The Legate may have preferred to keep the patronage
of vacant sees in his own hands rather than confirm the rival rights of chapters.
It is notable that when John made his peace with Rome, no suggestion of '* free
elections " was hinted at, whereas that concession was the essence of his charters to-
the English Church. Again, freedom of intercourse with Rome would facilitate-
appeals from the Legate to the Pope, and so diminish Gualo's authority. In the-
months to follow, Gualo exercised almost despotic power over the Church, excom-
municating all who supported Louis. On 27th October, 1217, he entered London,,
"went to the church of St. Paul, . . . and he put in new canons ; and the old ones
who had chanted the service in defiance of him he deprived of all their benefices. "^
Hist, des dues, 206. See also Adams, Origin, 258. Plonorius had conferred oa
Gualo authority to appoint to vacant sees and benefices ; see Bouquet, XIX. 623.
2 Minute points of difference, which are numerous, will be discussed under
appropriate chapters of the Commentary. Cf. Norgate, Minority, 10-14; Adams,
Origin, 256-7, who holds these changes to strengthen the theory *' that in the-
original charter the barons intended to state the law accurately and were not trying-
to take unjust advantage of the King."
REISSUES AND CONFIRMATIONS 143
ment or to construct safeguards for national liberties.
The King's minority implied a constitutional check, in the
necessary existence of guardians, but when Henry III.
attained majority, Magna Carta, deprived of its original
sanctions, would, with the disappearance of the Regency,
tend to become an empty record of royal promises. The
machinery of government remained exclusively monarchic ;
the King, once out of leading-strings, would be restrained
only by his own sense of honour and by the fear of armed
resistance — by moral forces rather than legal or constitu-
tional. The logical outcome was the Barons' War.
The importance of the omissions is minimized by two
considerations, (a) Many of the original provisions were
declaratory, and their omission in 12 16 by no means
implied that they were then abolished. The common law
remained what it had been previously, although it was not
deemed advisable to emphasize those particular parts of it
in black and white. In particular, throughout the reign of
Henry, the Comtyiune Concilium was always consulted
before a levy was made of any scutage or aid. (b) It is
stated in the new charter that the omitted clauses were
reserved for further consideration. In the so-called
" respiting clause " (chapter 42) six topics were thus
reserved because of their " grave and doubtful " import :
levying of scutages and aids; rights of Jews and other
creditors; the liberty of going from and returning to Eng-
land; the forest laws; the "farms" of counties; and the
customs relating to banks of rivers and their guardians.
This respiting clause amounts to a definite engagement by
the King to consider at some future time (probably as soon
as peace had been restored) how far it would be possible
to re-enact the omitted provisions.^
A practical difficulty confronted the advisers of the young
^Dr. Stubbs propounds the theory that this reissue of 1216 represents a com-
promise whereby the central government, in return for increased taxing powers,
allowed to the feudal magnates increased rights of jurisdiction. He gives, however,
no reasons for this belief, either in Select Charters, p. 339, or in his Constitutional
History, II. 27. The Crown reserved a freer hand in taxation, but there seems no
evidence that feudal justice gained ground against royal justice in 1216, not already
gained in 1215.
144 HISTORICAL SEQUEL TO MAGNA CARTA
King. Neither law nor custom afforded precedents for the
execution of documents during a king's minority. The
seal of a king was not available for his heir : the custom
was to destroy the matrix when a death occurred, to prevent
its being put to improper uses.^ Henry was made to
explain that, in the absence of a seal of his own, the Charter
had been sealed with the seals of Cardinal Gualo and of
William Marshal, Earl of Pembroke, " rectoris nostri et
regni nostri,"
In the Red Book of the Dublin Court of Exchequer there
is a copy of an Irish version of this Charter,^ bearing to
be executed at the same place and date as the English one
(Bristol, 1 2th November, 1216). It is possible that it was
not issued till some months later. After the coronation,
the Marshal wrote to Geoffrey de Marsh, Justiciar of Ireland,
promising to send a confirmation to the Irish of the liberties
just granted to Henry's English subjects. ^ It was not till
6th February, 1217, that this promise was fulfilled by the
sending of an Irish version of the Charter, in the King's
name, as a reward to his Irish subjects for their fidelity .*
If this is the original transcribed into the Red Book, it
would seem to have been antedated by nearly three months ;
while its terms suggest that little trouble was taken to adapt
the purport of the English Charter to Irish needs : four
perfunctory alterations substitute the freedom of the Irish
Church for that of the English Church; the liberties of
Dublin for those of London ; prohibit weirs in the Liffey,
instead of the Thames and Medway ; and make the " quarter
of Dublin," not of London, the measure of corn. The value
^ It is unnecessary to invent a catastrophe to account for the loss of John's seal.
Blackstone [Great Charter, xxix.) says, " King John's great seal having been lost
in passing the washes of Lincolnshire."
2 On pp. 69-73. Text is given in Early Statutes of Ireland (Rolls Series,
H. F. Berry), 5-19, and in Gilbert's Hist, and Mun. Docs, of Ireland, 65-72.
^ New Rymer, I. 145.
■* Rot. Pat., I. 31. Cf Norgate, Minority, p. 93 : "On 6th February, 1217, a copy
of the Charter was sent to Ireland with a letter in the King's name addressed to all
the King's faithful servants in Ireland, expressing his desire that . . . they and their
heirs should, of his grace and gift, enjoy the same liberties which his father and he
had granted to the realm of England." This was the Marshal's policy.
REISSUES AND CONFIRMATIONS 145
of the grant must have lain rather in the principle involved
than in the phraseology of particular clauses.
(II.) Reissue of 1217. The effect of the new Charter in
England was disappointing : apart from the bishops, only
four submissions were made to Henry in three months.^
The vicissitudes of the war need not be traced : on 19th
May, 1217, the royalists gained a decisive victory at the
battle known as the " Fair of Lincoln " ; and, on 24th
August following, Hubert de Burgh destroyed the fleet
on which Louis depended. The French prince was glad
to accept honourable terms. Negotiations, beginning on
the 8th, resulted, on 12th September, 12 17, in the Treaty
of Lambeth or Kingston. " The treaty of Lambeth is,
in practical importance, scarcely inferior to the charter
itself." 2 It marked the final acceptance by the advisers of
the Crown of the substance of Magna Carta as the per-
manent basis of government for England in time of peace,
not merely as a provisional expedient in time of war. Its
terms were equally honourable to both parties : to the
Regent and his supporters, because of the moderation they
displayed; and to Louis who, while renouncing all claim
to the English Crown, did so only on condition of a full
pardon to his lay allies, and a guarantee of the principles
they fought for. He strove in vain to make better terms
for the clergy, who were left exposed to Gualo's vindictive
greed.3
It must have been an impressive scene when, on 12th
September, at an eyot in the Thames near Kingston,
between rival armies lining opposite banks, Louis (who
had already granted a confirmation of the substance of
John's Charter) and Henry, laying their hands on the
Gospels, swore with the Legate and the Marshal to restore
to the barons of England and all other men of the realm
their rights and heritages, with the liberties formerly
demanded.* Henry promised to pay to Louis 10,000
marks nominally as an indemnity for his expenses, an
^ Davis, En^/. under Normans y 392. ^gj-ybbg^ Const. Hist.^ II. 25.
3 Petit-Dutaillis, Louis VIII., 171.
* Wendover, IV. 31-32 ; cited Norgate, Minority, 59, where full details are given.
146 HISTORICAL SEQUEL TO MAGNA CARTA
amount partly raised by a scutage of two marks " ad
Angliam deliberandam de Francis." i Louis, on his side,
restored all cities, lands and property taken by him in
England. One version of the treaty mentions particularly
the Rolls of Exchequer, charters of the Jews, charters of
liberties made in the time of King John, and all other
exchequer writings. 2 The restoration of rights and liber-
ties by Henry was the main provision of the treaty, and
this was fulfilled on 6th November, 12 17, by the issue of a
revised Charter of Liberties and a separate Forest Charter. ^
The issue of these two Charters put the coping stone to
the general pacification. After the havoc wrought by two
years of civil war, the moment had come for a declaration by
the Regent of his policy for ruling an England once more
at peace. Not only was he bound in honour to this course
by the Treaty of Lambeth, but the opportunity was a good
one for fulfilling the promise made in chapter 42 of the
Charter of 12 16. Accordingly the respiting clause of that
document disappeared, and some new clauses took its
place. The matters then reserved for further discussion
^Pipe Rolls, 2 and 3 Henry III., cited Petit-Dutaillis, 177. Miss Norgate
{Minority, 85) gives the rate per incuriavi as "two shillings."
'^Martene and Durand, Thesaurus Novus Anecdotoriwi (17 17), I. 858, cited
Norgate, Minority, 59. Blackstone thinks that under this clause the original of
the Articles of the Barons, captured by Louis with other national archives,
was restored and deposited at Lambeth Palace until the seventeenth century.
See Great Charter, xxxix.
^The Charter of Liberties of 1217, found among the archives of Gloucester
Abbey and now in the Bodleian Library at Oxford, still bears the impression of
two seals — that of Gualo in yellow wax, and that of the Regent in green. See
Blackstone, Great Charter, p. xxxv. The existence of the separate Forest Charter
was only surmised by Blackstone, Ibid., p. xlii ; but, shortly after he wrote, an
original of it was found among the archives of Durham Cathedral. For an account
of this and of its discovery, see Thomson, Magna Charta, pp. 443-5. This
Forest Charter bears the date 6th November, 12 17, and that, in itself, affords
presumption that the Charter of Liberties (undated) to which it forms a supplement
was executed at the same time. M. Bemont accepts this date ; see his Charles,
xxviii., and authorities there cited. Blackstone, Great Charter, xxxix., gives the
probable date as 23rd September. Dr. Stubbs gives 6th November in Const.
Hist., II. 26; and both dates alternatively in Sel. Chart., 344. Prof. Lawlor,
Engl. Hist. Rev., XXII. 514-6, contended for two independent issues, one of each
date ; but Prof. Powicke's researches, Eng. Hist. Rev., XXIV. 232, prove that there
is only one genuine charter of that year, dated as in the text.
REISSUES AND CONFIRMATIONS 147
as " gravia et dubitahilia " had now been reconsidered and
were either finally abandoned, or accepted with or without
amendment. Of the six topics "respited" in 12 16, one
(concerning forests and warrens) was dealt with in the
Forest Charter which took the place of chapters 36 and 38
of 1 2 16 and of the omitted provisions of 12 15; two others
(concerning scutage and enclosure of rivers) formed the
subjects of special chapters (44 and 20 respectively) ; while
the remaining three (the rights of Jews, free egress from
and ingress to England, and the " farms " of shires) were
not mentioned, although some of the grievances involved
may have been indirectly affected by certain newly added
chapters (e.g. that which regulated the times of meeting of
shire and hundred courts) or by the " saving clause " in
chapter 42.
To take the chief alterations in the order in which
they occur, ^ chapter 7 of 12 17 defines further a widow's
rights of dower; chapters 13, 14 and 15 alter the procedure
for taking the three petty assizes ; chapter 16 makes it clear
that the King's villeins do not share in the protection from
harsh amercement ; chapter 20, as already mentioned, treats
of river enclosures ; chapters 23 and 26 treat of purveyance,
the former extending the term of payment allowed to Crown
officials, the latter exempting entirely the carts of people
of the better classes — " parsons," knights and ladies. The
two provisions, taken together, speak eloquently against
the " democratic " interpretation of the Charter. Chapter
34 further limits or defines Crown bailiffs' rights in regard
to legal tests or " trials " where there is no evidence except
their own unsupported testimony; chapter 38 makes clear
a previously doubtful point concerning the King's rights
over escheats. Chapters 39, 42, 43, 44 and 46 will immedi-
ately receive separate discussion ; while chapter 47 ordains
" of common counsel " the demolition of all " unlicensed ''
strongholds built or rebuilt since the outbreak of the war
between John and his barons.
1 Details are discussed zn/ra, under appropriate chapters of John's charter.
The points in which this reissue differs from earlier and later charters are shown
in the Appendix, in the footnotes to the text of 1225.
148 HISTORICAL SEQUEL TO MAGNA CARTA
Chapter 44, generally regarded as replacing chapter 12
of 1 2 15, declares that scutages should be taken in future
as they had been wont to be taken under Henry 11. If, as
has already been suggested, the scutage question was the
immediate cause of the revolt of 12 15, the importance and
difficulty of this subject are obvious. Professor Adams ^
thinks that the leaders in 121 7, at their wits' end for a
solution, fell back on a vague, non-committal formula as
" an effort of despair." Yet the old rates of scutage could
still be read in the Exchequer Rolls, and the practice of
a reign that had closed only twenty-six years before must
have been familiar to many others besides the aged Marshal
who set seal to the Charter. In reality John's innovations
were now swept away ; these included the habit of making
an annual tax of what was meant for special emergencies,
the assessment under the Inquest of 12 12, the demand for
scutage and service cumulatively, and, above all, the high
rate of three marks per knight's fee.^
The essence of the barons* demands in 1217 must un-
doubtedly have been the return to the normal maximum
rate of 2 marks. The substitution of this reference to the
usage of Henry for the discarded chapters 12 and 14 of
John's Charter (which made " common consent " necessary
for all scutages, whatever the rate) was a natural com-
promise; and the barons in agreeing to it were justified
in thinking, from their own medieval point of view, that
they were neither submitting to unfair abridgments of their
rights, nor yet countenancing reactionary measures hurtful
to the growth of liberty .^ Yet when this alteration is
viewed by modern eyes, in the light cast by the intervening
centuries of constitutional progress, the conclusion suggests
itself that, unconsciously, retrograde tendencies were at
^ Origin^ 260.
2 Pollock and Maitland, I. 25011., suggest that this chapter absolved under-
tenants from the obligation of personal attendance in the army.
2 Mr. Hubert Hall {Eng. Hist. Rev.y IX. 344) takes a different view, considering
that a reduction of scutages to the old rate of Henry II. was impossible; he speaks
of " the astounding and futile concession in c. 44 of the charter of 12 17." The
clause is neither astounding nor futile if we regard it as a promise by Henry III.
that he would not exact more than two marks per fee without consent^ and if we
REISSUES AND CONFIRMATIONS 149
work. All mention of the Commune Concilium — that pre-
decessor of the modern Parliament, that germ of all that has
made England famous in the realm of constitutional laws
and liberties — disappears. If (as it w^as once the fashion to
maintain) the control of taxation by a national assembly, the
conception of representation, and the indissoluble connec-
tion of these two principles with each other, really found
place in Magna Carta in 12 15, they were ejected in 12 16,
and failed to find a champion in 1217 to demand their
restoration.
A modern statesman, with a grasp of constitutional
principles, would have seized the occasion of the revision
of the Charter, to define the functions of the Great Council
with precision and emphasis. He would not lightly have
thrown away the written acknowledgment implied in
chapters 12 and 14 of 12 15 — in the germ, at least — of the
right of a national council to control the levying of taxes.
The magnates in 12 17 were content, however, to abandon
abstract principles; they were selling, not indeed their
birthright, but their best means of gaining new rights from
the Crown, for " a mess of pottage."
Such considerations, however, must not be pressed too
far; no one seriously thought in 1217, any more than in
1 2 16, of dispensing with future meetings of the feudal
tenants in Commune Concilium. Great Councils con-
tinued to meet with increasing frequency throughout the
reign of Henry III., and the consent of the magnates was
habitually asked to scutages even at a lower rate than
that which had been normal in Henry II.'s reign. Some-
times such consent was given unconditionally; sometimes
in return for a new confirmation of the Charters; some-
times a demand was met by absolute refusal — the first dis-
tinct instance of which seems to have occurred in January,
1242.1
Chapters 39, 42 and 43, treating of topics not mentioned
farther note that it was the practice of his reign to ask such consent from the Com-
nmne Concilium for scutages even of a lower rate. A levy of los., for example,
was granted by a Council in 1221. See Stubbs, Const. Hist., II. 33.
^M. Paris, 581-2; Sel. Chart., 369.
150 HISTORICAL SEQUEL TO MAGNA CARTA
in John's Charter, fall (strictly considered) outside the
scope of this treatise, but a short account of their main
provisions may prove useful here. Chapter 42, from
its possible connection with the omitted chapter 25 of
12 15, may be taken first. The shire court is not to meet
oftener than once a month ; less often, where local custom
so ruled it. No sheriff or bailiff is to make his tourn
through the hundreds oftener than twice a year — after
Easter and after Michaelmas respectively — and only in the
accustomed places. Careful provision is made for holding
view of frankpledge at Michaelmas, with due regard to
" liberties " upon the one hand, and to the King's peace and
keeping the tithings full upon the other. Finally, the
sheriff is not to make " occasions," but shall content himself
with what he used to have for holding view of frankpledge
in Henry of Anjou's time — a reference, it would seem, to
that " Sheriff's aid " which was the cause of a famous
quarrel in 1163 between Henry and his recently appointed
Archbishop, Thomas a Becket.^
Chapters 39 and 43 link themselves rather with the future
than the past, showing that new problems were thrusting
themselves to the front since the days of John — topics round
which much controversy was to rage. These chapters
anticipate the principles underlying two famous measures
of Edward's reign : the statutes of Quia Emptores 2 and
of Mortmain. 3 Chapter 39 forbade for the future that any
freeman should give away or sell so much of his land as
would not leave sufficient to furnish the service due from
the fief to the feudal lord.
Chapter 43 marks the growing hostility against the
accumulation by the monasteries of wealth in the form of
landed estates. The times were not ripe for a final solution
of this problem, and the charter only attempted to remedy
one of the subsidiary abuses of the system, not to abolish
the main evil. An ingenious expedient had been devised
by lawyers to enable tenants to cheat their lords out of
some of the lawful feudal incidents. Religious houses
1 Sel. Chart. ^ 129. 2 jg Edward I., also known as Westminster III.
^ 7 Edward I., also known as the Statute de religiosis.
I
REISSUES AND CONFIRMATIONS 151
made bad tenants, since, as they never died, the lord of
the fief was deprived of wardship, relief, and escheat.
This was not unfair, provided the transaction was bona fide.
Sometimes, however, collusive agreements were made,
whereby a freeholder bestowed his lands on a particular
house, which then subinfeudated the same subjects to the
original tenant, who thus got his lands back, but now
became tenant of the church, not of his former lord. The
lord was left with a corporation for his tenant ; and all the
profitable incidents would, under the new arrangement,
accrue to the church. Such expedients were prohibited,
under pain of forfeiture, by chapter 43 of the reissue of
1217; and this prohibition was interpreted liberally by the
lords in their own favour.^
The only remaining provision that calls for comment is
the " saving clause " in chapter 46, intended, perhaps, to
cover the gaps left in the Charter as conceived in 12 15, by
the decision not to restore some of the duhitahilia of 1216 :
this chapter reserves to archbishops, bishops, abbots,
priors, templars, hospitallers, earls, barons, and all other
persons, cleric and lay, the liberties and free customs they
previously had. The vagueness of this provision deprived
it of value.
These were the main alterations made in 121 7 in the
tenor of the Great Charter. This reissue is of great
importance, since it represents practically the final form
taken by the Charter. On 22nd February, 12 18, copies
of the Great Charter, in this new form, were sent to the
sheriffs to be published and enforced. In the writs accom-
panying them, the special attention directed to the clause
against unlicensed castles shows the importance attached
to their demolition.^ These remained in 1217, as in 1154,
a result of past civil war, and a menace to good govern-
ment in the future. It was the aim of every efficient ruler
to abolish all fortified castles — practically impregnable in
the thirteenth century when artillery was unknown —
except those of the King, and to see that royal castles were
under command of castellans of approved loyalty. John
1 See Pollock and Maitland, I. 314. ^See Rot. Claus., I. 377.
152 HISTORICAL SEQUEL TO MAGNA CARTA
had placed his own strongholds under creatures of his
own, who, after his death, refused to give them up to his
son's Regent. The attempt to dislodge these soldiers of
fortune, two years later, led to new disturbances in which
the famous Falkes de Breaut^ played a leading part.^ The
destruction of " adulterine " castles and the resumption of
royal ones were both necessary accompaniments of any
real pacification.
Attempts have been made to estimate the motives and
forces at work in these considerable changes in the text of
the revised Charter. Attention to minute points of detail
in practice and phraseology are rightly held to indicate a
return towards more normal conditions under which " pro-
blems of everyday government " and the more accurate
statement of the law receive attention. 2 The new Charter,
in its desire to profit by the actual experience of the past
two years, has some analogy to a modern amending
statute. Other alterations, however, of a more fundamental
nature would seem to have been deliberately made; and,
as changing the old customs of the realm, they are of a
legislative character in the strictest sense. Evidence of
pressure from the baronage, in pursuance of their own
selfish interests, can be traced in some at least of these
innovations ; but, on the other hand, the destruction of their
" adulterine " castles shows that there were limits to their
power.
The sincerity with which Magna Carta, thus amended,
had been accepted by those in power is shown by the issue,
seven months later, of letters to the sheriffs ordering them
to publish the Charter in their shires and see that it was
put in force; while orders were also given to respect the
franchises of the city of London. ^
(III.) Reissue of 1225.4 Henry's second Charter, like his
first, had been authenticated by the seals of the Legate
and the " Rector." The objection to providing a seal of
Henry's own w^as that it might be used to prejudice the
royal prerogatives by alienating Crown lands and fran-
iStubbs, Consf. Hist., II. 32. "^Qi. Adams, Origin, 258-260.
^ New Rynier, I. 147, 150. ^See text in Appendix.
REISSUES AND CONFIRMATIONS 153
chises during the King's minority. But, shortly before
Gualo left England, his task as Legate well done, instruc-
tions were given to a goldsmith to prepare a royal seal of
silver, 5 marks in weight. Apparently the first use to
which it was put was to attest letters patent, issued after
Michaelmas, 12 18, warning all men that no grant in per-
petuity was to be sealed with it till the King came of age.^
The full twenty-one years would not be completed until
ist October, 1228; but by letters dated 13th April, 1223,
Pope Honorius declared his ward to be of full age under
certain reservations. A few months earlier (30th January,
1223) consternation had been created by writs issued in
the King's name to the sheriffs for a sworn inquest as to
the customs and liberties enjoyed by John in the various
shires, before the war; and Henry's advisers thought it
prudent to issue second writs on 9th April ordering that
the results of the inquest should be held back till 25th June,
and disclaiming all intention of raising up " evil customs." -
It was not, apparently, until December, 1223, that the
Pope's declaration of the partial ending of Henry's non-
age was given effect to, with consent of the Council; and
on 13th January, 1224, Henry was asked by Stephen
Langton for a new confirmation of the Charters.^ In
the ensuing debate, William Brewer answered for the
King : " The liberties you ask ought not to be observed
of right, because they were extorted by force," words which,
coming from a royal favourite, were sufficient to justify
suspicion. When the Archbishop had rebuked this rash
^See Norgate, Minority, 102; Stubbs, Const. Hist., II. 30. Annals of
Waverley, 290, speak of a reissue of the charters about this date ; but this
probably results from confusion with what happened a year earlier.
"^ New Ryiner, I. 168; Rot. Claus., I. 569.
^ R. Wendover, IV. 84, who dates the demand a year earlier. Miss Norgate's
chronology is here followed [Minority, 215 n.). The request would be a natural
corollary to the King's coming of age. There may have been special reasons for
uneasiness, e.g. the suspicions aroused by the recent inquest, the resumption of
royal castles from their former wardens, and the Crown's need of increased sources
of taxation. See Adams, Origin, 281 n. ; Turner, Trans. R.H.S., I. 205 ff.
Miss Norgate {Ibid. 215) suggests that Langton desired some modification of the
terms of the charter of 121 7.
154 HISTORICAL SEQUEL TO MAGNA CARTA
adviser : " William, if you loved the King, you would not
endanger thus the peace of his realm," the young King
said : " We have all sworn these liberties, and what we
have sworn we are bound to keep."i
No formal charter seems at this time to have been
granted; but the barons' opportunity came in December
of the same year, when Henry's necessities forced him to
demand a contribution of one-fifteenth of moveables. A
bargain on these terms was struck, and on nth February,
1225, the Charter of Liberties and the Forest Charter were
both reissued. 2 The new Forest Charter was practically
identical with that issued in 1217; while the alterations
in the new Charter of Liberties were the result of a deter-
mination to place on record the circumstances in which it
had been granted. In the preamble Henry stated that he
acted " spontanea et bona voluntate nostra " and all refer-
ence to consent was omitted, although many magnates
appear as witnesses. These alterations were intended ro
emphasize the fact that no pressure had been brought to
bear, and thus to meet the objection urged by Brewer in
1224, that the Charter had been extorted by force.^
The " consideration " also appears in the concluding
portion of the Charter, where it is stated that in return for
the foregoing gift of liberties along with those granted in
the Forest Charter, the archbishops, bishops, abbots,
priors, earls, barons, knights, free tenants, and all others
of the realm had given a fifteenth part of their moveables
to the King.
^R. Wendover, /did.
2 Miss Norgate {Minority, 262), for reasons not fully explained, speaks of this
purchasing of admitted rights by payments of hard cash as an "irretrievable
blunder." Does she not neglect, however, the effect of the legal doctrine of
* ' valuable consideration " and the force underlying Brewer's argument that earlier
charters were voidable because granted under duress ?
3 Dr. Stubbs thinks that in avoiding one danger, a greater was incurred. " It
must be acknowledged that Hubert, in trying to bind the royal conscience, forsook
the normal and primitive form of legislative enactment, and opened a claim on the
king's part to legislate by sovereign authority without counsel or consent." {Consi.
Hist., 11. 37.) This seems to exaggerate the importance of an isolated precedent,
the circumstances of which were unique. The confirmation was something far
apart from an ordinary " legislative enactment." It had been asked and paid for.
REISSUES AND CONFIRMATIONS 155
The prominence given to this feature brings the trans-
action embodied in the reissue of 1225 (as compared with
the original grant of 12 15) one step nearer the legal cate-
gory of "private bargain." In another important new
clause — founded probably on a precedent taken from
chapter 61 of John's Charter — Henry is made to declare :
" And we have granted to them for us and our heirs, that
neither we nor our heirs shall procure any thing whereby
the liberties in this charter shall be infringed or broken ;
and if any thing shall be procured by any person contrary
to these premises, it shall be held of no validity or effect."
This provision was clearly directed against future papal
dispensations; the clause, however, was diplomatically
made general in its terms.
One original copy of this third reissue of the Great
Charter is preserved at Durham with a still perfect impres-
sion of Henry's recently made seal in green wax, though
the parchment has been " defaced and obliterated by the
unfortunate accident of overturning a bottle of ink." 2 A
second original is to be found at Lacock Abbey, in Wilt-
shire. The accompanying Forest Charter is also preserved
at Durham. 3
This third reissue brings the story of the genesis of
the Great Charter to an end. It marked the final form
assumed by Magna Carta; the identical words were then
used which afterwards became stereotyped and were con-
firmed, time after time, without further modification. It
is this Charter of 1225 which (in virtue of the confirmation
of Edward I.) still remains on the statute book.*
Henry, however, was not yet, in 1225, fully of age; and
suspicions seem still to have been entertained as to what
would be his attitude when he became of full age for all
purposes. It w^as apparently in January, 1227, that the
1 A few minor alterations, such as the omission of the clause against unlicensed
castles (now unnecessary) and some verbal changes need not be mentioned. A
list of these is given by Blackstone, Great Charter^ 1.
2 See Blackstone, Ibid.^ xlvii. to 1. ^ Ibid.
* One slight exception should be noted. In one point of detail a change had
occurred between 1225 and 1297 ; the rate of relief payable from a barony had been
reduced from ;^I00 to 100 marks. See infra, under chapter 2.
156 HISTORICAL SEQUEL TO MAGNA CARTA
Council authorized the King to issue writs to his sheriffs
that all grants of lands, tenements, or liberties, to be held
valid, must be confirmed under Henry's seal. Writs in
these terms went forth on 21st January. This was tanta-
mount to an official declaration that the minority was
ended. ^
Under feudal theory, the close personal relations between
lord and vassal had to be renewed when a death occurred :
every new King exacted payments for confirmation of
earlier grants, and Henry's previous recognitions had been
provisional. The King was enunciating no general doc-
trine of contempt for vested interests : his abuse of power
lay in the exorbitant sums charged for charters confirming
earlier, informal "precepts." 2 There is no substantial
ground for the opinion, once widely held,^ that the King
intended to annul the Great Charter, and that, accordingly,
it was not in force from 1227 to 1237. Nor, in the instruc-
tions to the sheriffs, is there a word said about the Forest
Charter. Henry, indeed, dared not openly repudiate
either of the Charters, which had received full papal
authority.
Yet he was far from scrupulous in observing the letter
of their provisions : there was good warrant for the com-
plaint contained in article 7 of the Petition of 1258,* that
Henry broke his bargain, by extending the forests beyond
the boundaries to obtain which the fifteenth had been paid.
The process was begun by the issue of letters close, on 9th
February, 1227.5 Henceforward, Henry's attitude towards
the charters was a settled one : he confirmed them with a
^ A bull of Gregory IX., dated 13th April, 1227, confirmed this. See Blackstone,
Great Charter^ li., and Stubbs, Const, Hist., II. 39.
2 See Powicke, Eng. Hist. Rev., XXIII. 221.
^ R. Wendover, IV. 140, is apparently the source of the error. See Norgate,
Minority, 266 n.
^Sel. Chart., 383.
^See Rot. Claus., II. 169. The best account is in Turner's Select Pleas of the
Forest, pp. xcix. to cii. , who gives a full and convincing account of Henry's pro-
cedure and motives. " The king neither repudiated the Charter of the Forest nor
annulled the perambulations vi^hich had been made in his infancy. He merely
corrected them after due inquiry." See also Adams, Origin, 283 n.
REISSUES AND CONFIRMATIONS 157
light heart when he could obtain money in return, and
then acted as though they did not exist.
(IV.) Confirmations (1237 to 1297). After the close of
Henry's minority history is concerned not with reissues of
the Charter but with confirmations. Matthew Paris refers
to the circumstances under which the first of these was
executed on 28th January, 1237 : as the express condition
of a grant of " a thirtieth part of the kingdom, to wit of
all moveables," Henry promised that thenceforward the
'' lihertates Magnae Cartae " should be inviolably observed. ^
This Charter differs fundamentally from those of 12 15,
1216, 1217 and 1225. It does not rehearse the substance of
any one of the " liberties " it confirms, but contents itself
with a brief reference : " We have granted and by this our
charter confirmed ... all liberties and free customs con-
tained in our charters which we caused to be made to our
subjects during our minority, to wit as well in magna carta
nostra as in carta nostra de foresta." 2 Even with the long
list of witnesses, occupying half of its extent, this document
is a small one when compared with the voluminous parch-
ments of earlier grants. It has been suggested ^ that the
marked contrast in size may have given rise to the practice
of alluding to the earlier charter (whether of John or Henry)
as Magna Carta, in distinction from the new parva cartaA
In support of the suggestion, it may be argued that the
phrase " Magna Carta " is never used by Roger of Wen-
dover, and that its first appearance in the narrative of
Matthew Paris is in the passage just quoted, suh anno
1237, " carta lihertatum " being the usual description. The
words " Magna Carta " appear a second time in his account
of a famous debate in 1242,^ where pointed reference is
^M. Paris, 435; Sel. Chart., 326-7.
2 Its facsimile is given in Statutes of the Realm ; its text in Sel. Chart. ^ 365-6.
^ByDr. GeoxgQ 'iiitWson, Jtcridical Reviezu, XVII. 137.
* Henry I.'s charter was also described as "Magna Carta" but not till the
thirteenth century. Leibermann, Trans. R.H.S., VIII. 21.
'^M. Paris, 581-2; Sel. Chart., 369-370. Bracton's Notebook (see its Index)
mentions the Charter eight times under various descriptions, but never as the Great
Charter.
158 HISTORICAL SEQUEL TO MAGNA CARTA
made to the bargain struck in 1237, when Henry conceded
the liberties contained in " Magna Carta " in return for the
thirtieth of moveables " et hide fecit eis quandam parvam
cartam suam." The antithesis is here emphatic.
The adoption of this parva carta means that the Charter
had become stereotyped as it stood in 1225, and no longer
moved with the times. For ten years previous it had, like
a living thing, adapted itself to changing needs and
grievances. The new style possibly corresponds with a
new attitude on the part of both King and barons. Henry
had abandoned any intention of repudiating the Charter
or even of infringing its specific promises as to wardships,
reliefs or the like : his practice was to evade its spirit,
while observing its letter. The opposition, on their part,
may unconsciously have come to consider the Charter's
value to lie, not in its specific clauses, but in its assertion
of the existence of a fixed body of law to which successful
appeal could be made against the King's caprice. Changes
in the texture of that law are no longer reflected in re-
affirmations of the Charter; but must be sought for in a
series of supplementary documents such as those of 1258,
1297, 1300, 131 1, 1406 and 1628.
After 1237 little is heard of the charters until 1253, when
complaint was raised of infractions, particularly in regard
to the privileges of the Church. Both charters were
republished, and on 13th May, the sentence of excom-
munication, which had accompanied the reaffirmations of
1225 and 1237, was repeated in a peculiarly impressive
manner.^
In 1265 Simon de Montfort, during his brief period of
power, exacted from Henry and his son a new confirma-
tion, dated 14th March, notable for its clause empowering
" all of the realm to rebel against us and use their utmost
resources and efforts to our hurt " in imitation of chapter 61
of John's grant. After Simon's overthrow and death, the
King and the young Edward, of their own initiative,
affirmed the charters by chapter five of the Statute of
Marlborough (1267). Of the confirmations of Edward's
iBlackstone, Greai Charter, 70-72 ; Stubbs, Sel. Chart., 373.
REISSUES AND CONFIRMATIONS 159
reign, it is only necessary to mention the emphatic Con-
firmatio Cartarum of 1297, accompanied by an Inspeximus
of the issue of 1225, granted under conditions that are
well known. It contains new clauses which impose restric-
tions on the taxing power of the Crown ; and these, to
some extent, take the places of those chapters (12 and 14)
of the original grant of John, which had been omitted in
all intervening grants.
Of later confirmations, Coke ^ has counted 15 under
Edward III., 8 under Richard II., 6 under Henry IV. and
one under Henry V. Of these, only the statute of 1369
(42 Edward III. c. i) requires special notice : it commands
that " the Great Charter and the Charter of the Forest be
holden and kept in all points, and if any statute be made
to the contrary that shall be holden for none." Parliament
in 1369 thus sought to deprive future Parliaments of the
power to effect any alterations upon the terms of Magna
Carta. Yet, if Parliament in that year had the power to
add anything, by a new legislative enactment, to the ancient
binding force of the Great Charter, it follows that succeed-
ing Parliaments, in possession of equal powers, might
readily undo by a second statute what the earlier statute
had sought to effect. If Parliament had power to alter the
sacred terms of Magna Carta, it had power to alter the less
sacred statute of 1369 which declared it unalterable. The
terms of that statute, however, are interesting as perhaps
the earliest example on record of the illogical theory that
the English Parliament might so use its present legislative
supremacy as to limit the legislative supremacy of other
Parliaments in the future. 2
II. Magna Carta and the Reforms of Edward I.
The Great Charter, alike from its excellences and its
defects, exercised a potent influence throughout the two
succeeding reigns. It is hardly too much to say that the
failure of Magna Carta to provide adequate machinery for
^ Second Institute, p. i.
2 Many further details will be found in Bemont, Chartes, xxx.-lxx., and
authorities there cited.
i6o HISTORICAL SEQUEL TO MAGNA CARTA
its own enforcement is responsible for the protracted
struggles and civil war that made up the troubled reign
of Henry III.; while the difference of attitude assumed by-
Henry and his son respectively towards the scheme of
reform it embodied, explains why one reign was full of
conflicts and distress, while the other was prosperous and
progressive. The fundamental difference between the
policies of Henry and Edward lies in this, that while
Henry, in spite of numerous nominal confirmations of
Magna Carta, never loyally accepted the settlement it
contained, Edward acquiesced in its main provisions
honestly on the whole, with a sincere intention to carry
them into practice.
At the same time, the attitude of Henry III. indicates
an advance upon that of John. Henry, on attaining
m.ajority, had confirmed the charters freely and on his
own initiative, and found himself thereafter unable openly
to repudiate the bargain he had made. Yet the settlement
between Crown and baronage was nominal rather than
real : the King was bound by bonds of parchment which
he could break at pleasure. In the absence of sanctions
for its enforcement, the Charter became an empty expres-
sion of good intentions : no constitutional expedient
existed to obviate a final recourse to the arbitrament of
civil war. Thus, part of the blame for the recurring and
devastating struggles of the reign of Henry must be
attributed to the defects of the Great Charter.
The whole interest of the reign, indeed, lies in the
attempts made to evolve adequate machinery for enforcing
" the liberties." Experiments of many kinds were tried in
the hope of turning theory into practice. The system of
government outlined in the Provisions of Oxford of 1258,
for example, reproduced the defects of the scheme contained
in chapter 61 of the Great Charter, and added new defects
of its own. The baronial committee was not designed to
enter into friendly co-operation with Henry in the normal
work of government, but rather to supersede entirely certain
of the royal prerogatives. No glimmering was yet apparent
of the true solution afterwards adopted with success : it
THE REFORMS OF EDWARD I, i6i
was not yet realized that the best way to control the Crown
was through the agency of its own Ministers.
If Simon de Montfort had any vague conception of the
real remedy for the evils of the reign, his ideals were over-
ruled in 1258 by the more extreme section of the baronial
party. Earl Simon, indeed, had one opportunity of putting
his theories into practice : during the brief interval between
the battle of Lewes, which made him supreme for the
moment, and the battle of Evesham, which ended his
career, he enjoyed an unfettered control ; and some authori-
ties find in the provisional scheme of the closing months
of 1264, traces of the constitutional expedient afterwards
successfully adopted as a solution of the problem. In one
respect, the Earl of Leicester did influence the development
of the English constitution ; he furnished the first precedent
for a true Parliament, reflecting interests wider than those
of Crown tenants and free-holders, when he invited repre-
sentatives of the boroughs to take their places by the side
of representatives of the counties in a national council
summoned to meet in January, 1265. His schemes of
government, however, were not fated to be realized by
him in a permanent form : the utter overthrow of his
faction followed his decisive defeat and death on 4th
August, 1265.
The personal humiliation of Simon, however, assured
the ultimate triumph of the cause he had made his own.
Prince Edward, from the moment of his brilliant victory
at Evesham, was not only supreme over his father's
enemies, but also within his father's councils. He found
himself in a position to realize some of his political ideals ;
and he adopted as his own, the main constitutional con-
ceptions of his uncle Earl Simon, who had been his friend
and teacher before he became his deadliest enemy.
Edward Plantagenet, alike when acting as chief adviser
of his aged father and after he had succeeded to his throne,
not only accepted the main provisions of the Great Charter,^
^ The best proof of this will be found in a comparison of Magna Carta with the
statute of Marlborough, and the chief statutes of Edward's reign, notably that of
Westminster I.
i62 HISTORICAL SEQUEL TO MAGNA CARTA
but adopted also a new scheme of government which formed
their necessary counterpart. The very fact of the adoption
of Earl Simon's ideals by the heir to the throne altered
their chances of success. All such schemes had been fore-
doomed to failure so long as they emanated from an
opposition leader, however powerful ; but their triumph was
assured when accepted by the monarch himself. Under
the protection of Edward I. — the last of the four great
master-builders of the constitution — the Commune Con-
cilium of the Angevin kings grew into the English Parlia-
ment. This implied no sudden dramatic change, but a
long process of adjustment, under the guiding hand of
Edward.
The main features of his scheme may be briefly sum-
marized : Edward's conception of his position as a national
king achieving national ends, the funds necessary for
which ought to be contributed by the nation, led him to
devise a system of taxation which would fill the Exchequer
while avoiding unnecessary friction with the tax-payer.
In broadening the basis of finance, he was led to broaden
the basis of Parliament; and thus he advanced from the
feudal conception of a Commune Concilium, attended only
by Crown tenants, towards the nobler ideal of a national
Parliament containing representatives of every community
and every class in England. The principle of representa-
tion (foreshadowed in a vague way for centuries before the
Conquest in English local government), now found a home,
and, as it proved, a permanent home, in the English
Parliament.
The powers of this assembly widened almost auto-
matically, with the widening of its composition. To its
original function of taxation, legislation was soon added.
The functions of hearing grievances and of proffering
advice had, even in the days of the Conqueror, belonged
to such of the great magnates as were able to make their
voices heard in the Curia Regis; and similar rights
were gradually extended to the humbler members of the
augmented assembly. The representatives of counties and
towns retained rights of free discussion even after Parlia-
THE REFORMS OF EDWARD I. 163
ment had split into two Houses. These rights, fortified
by command of the purse strings, tended to increase, until
they secured for the Commons some measure of control
over the executive functions of the King, varying in extent
and effectiveness v^ith the weakness of the King, with his
need of money, and with the political situation of the hour.
The new position and powers of Parliament logically
involved a corresponding alteration in the position and
powers of the smaller but more permanent council or Con-
cilium Ordinariuin (the future Privy Council). This had
long been increasing in power, in prestige, and in inde-
pendence, a process quickened by the minority of Henry
ni. The Council was now strengthened by the support
of a powerful Parliament, usually acting in alliance with
the leaders of the baronial opposition. The Council was
recruited from Parliament, and the appointment of King's
ministers was influenced by the proceedings in the larger
assembly. 1
The Council thus became neutral ground on which the
conflicting interests of King and baronage might be dis-
cussed and compromised. Wild schemes like that of
chapter 61 of Magna Carta or that typified in the Committee
appointed by the Mad Parliament of 1258, were now
unnecessary. The King's own ministers, backed by Parlia-
ment, became an adequate means of enforcing the constitu-
tional restraints embodied in royal Charters. The problem
was thus, for the time being, solved. The bargain made
at Runnymede between the English monarch and the
English nation found its counterpart and sanction, before
the close of the thirteenth century, in the conception of a
King ruling through responsible ministers and in harmony
with a national Parliament. Edward Plantagenet, though
merely the unconscious instrument by whose agency the
new conception was for a time partially realized, yet merits
the gratitude of posterity for his share in the elabora-
tion of a working scheme of government which took the
^The doctrine that the Commune Concilium should have some voice in the
appointment of Ministers had been acted upon on several occasions even in the
reign of Henry III. See Stubbs, Const. Hist., II. 41.
i64j^ HISTORICAL SEQUEL TO MAGNA CARTA
place of the clumsy expedients designed as constitutional
sanctions in 1215. The ultimate triumph of the principles
underlying Magna Carta was assured not through any
executive committee of rebellious barons, but through the
constitutional machinery devised by Edward Plantagenet.
PART V.
MAGNA CARTA: ORIGINAL VERSIONS, PRINTED
EDITIONS AND COMMENTARIES.
I. Manuscripts of Magna Carta and Relative Documents.
The barons who had forced the Great Charter on King
John were determined that its contents should be widely
known and permanently preserved. It was not sufficient
that the great seal should be formally impressed upon one
parchment. Those who compelled John to submit were
not content even with the execution of its terms in dupli-
cate or in triplicate : copies w^ere to be distributed throughout
the land, to be preserved in important strongholds and
among the archives of the chapters of cathedral churches.^
I. The extajit original versions. Of the many sealed
copies, four have escaped the destroying hand of time :
(i) The British Museum Magna Carta, number one — for-
mally cited as " Cotton, Charters XIII. 31A." The recent
history of this document, which is possibly the original
copy delivered to the barons of the Cinque Fortes, is well
known. It was discovered in the seventeenth century,
among the archives of Dover Castle, by the Warden, Sir
Edward Dering, and by him presented to Sir Robert
Cotton. 2 In the fire of 23rd October, 1731, this Charter was
rendered in parts illegible, while the yellow wax of the seal
was melted. It is possible that the accident has added to
the prestige of this particular copy of Magna Carta.
1 For methods of publishing Great Charters see R. L. Poole, Eng: Hist. Rev.
XXVIII. 444 (July, 1913) ; and infra under c. 62.
2 The accompanying letter, dated loth May, 1630, is also preserved in the British
Museum, as "Cotton, Julius, C. III. Fol. 191."
i66 MAGNA CARTA
Like the three others still extant, it is written continu-
ously, though with many contractions, in a neat, running,
Norman hand. Some omissions seem to have been made
in the body of this version and to have been supplied at
the foot. These are five in number.^ It is possible to
regard them as corrections of clerical omissions due to
carelessness or hurry in engrossing the deed ; but the fact
that one of the additions is distinctly in the King's favour
raises a presumption that they embodied additions made
as afterthoughts to what had been originally dictated to
the engrossing clerk, and that they were inserted at
the King's suggestion before he would adhibit the great
seal.
The importance of this document was recognized, and a
facsimile was prepared by John Pine, a well-known en-
graver, some eighteen months after the great fire. The
engraving bears a certificate, dated 9th May, 1733, that the
copy is founded on the original, which had been shrivelled
up by the heat ; but that, where two holes had been burned,
the words obliterated had been replaced from the other
version (to be immediately described) preserved in the
Cottonian collection.
(2) The British Museum Magna Carta, number two —
cited as "Cotton, Augustus, II. 106." ^ The early history
of this document is unknown, but it came into the posses-
sion of Mr. Humphrey Wyems, and by him was presented
to Sir Robert Cotton on ist January, 1628-9. Unlike the
other Cottonian copy, this one is happily in an excellent
^ These are carefully noted among the variations described by the editors of the
Charters of Liberties forming Part I. of the first volume of the Statutes of the Realm.
These addenda are (i) at the end of c. 48, ^^ per eosdem, ita quod nos hoc sciamus
prius, vel justiciarius noster, si in Anglia nonfuerimus^ " providing that the King
should receive intimation of all forest practices branded as " evil " before they are
abrogated ; (2) ; two small additions, near the beginning of c. 53, («:), *' et eodem
modo dejusticia exhibenda," and {b) *^ vel remansuris forestis'^ ; (3) in c. 56, these
four words, *' in Anglia vel in PVallia" ; and (4) in c. 61 the words '■^ in per-
petuufn" after ^^ gaudere." In the 2nd British Museum MS. three of these
addenda appear at the foot, viz. (i), {2a) and {2b) ; but the words of (3) and (4)
are incorporated in the body of that MS.
* Reproductions of this are sold at the British Museum for 2s. 6d.
MANUSCRIPTS AND RELATIVE DOCUMENTS 167
state of preservation ; but there is no trace left of any seal.^
Three of the five addenda inserted at the foot of the copy-
previously described are found in a similar position here;
but the substance of the two others is included in the body
of the deed. On the left-hand margin, titles intended to
be descriptive of several chapters occur in a later hand.
Thus for the preservation of two original copies of the
national charter of liberties the nation is indebted to Sir
Robert Cotton. Several authors ^ gravely record how Sir
Robert discovered " the palladium of English liberties " in
the hands of his tailor at the critical moment when scissors
were about to transform it into shapes for a suit of clothes.
This detail is a fable, since both manuscripts of Magna
Carta in the Cottonian collection are otherwise accounted
for.
(3) The Lincoln Magna Carta, This copy is under the
custody of the Dean and Chapter of Lincoln Cathedral,
where it has lain for many centuries. The word " Lin-
colnia " is endorsed in a later hand in two places on folds
of the parchment. It has no corrections or additions
inserted at the foot, but embodies in their proper places
all those which occurred in the versions already discussed.
Further, it is executed with flourishes and in a more
finished manner : the inference is that it took longer to
engross. The Record Commissioners, in preparing the
Statutes of the Realm, considered this version of superior
authority to the others and have chosen it for their engrav-
ing published in 1810 in that valuable work, and also in
the first volume of their edition of Rymer's Foedera in
1816.3
(4) The Salisbury Magna Carta — preserved in the archives
^ " The fold and label are now cut off, though it is said once to have had slits in
it for two seals, for which it is almost impossible to account ; but Dr. Thomas
Smith, in his Preface to the Cottonian Catalogue, Oxford, 1695, f"oJio. states that
they were those of the barons" (Thomson, Magna Carta, 425). The facsimile
published by the Trustees of the British Museum shows slits for three seals.
2 See Isaac D'Israeli, Curiosities of Literature, I. 18, and Thomson, Magna
Carta, 424.
^ The engraving was executed to their order by James Basire.
i68 MAGNA CARTA
of the Cathedral there. The early history of this manu-
script has not been traced, but its existence was known
at the close of the seventeenth century.^ Sir William
Blackstone, in April, 1759,^ instituted a search for it, but
without success — his inquiries being met with the state-
ment that it had been lost some thirty years before, during
the execution of repairs in the Cathedral library. As its
disappearance had taken place during the tenure of the see
by Gilbert Burnet, his political adversaries accused him of
appropriating it — an undoubted calumny, as will be here-
after explained. The document had not been re-discovered
in 1800 when the royal commission published a report of
Its inquiries for national records.^ Two sub-commissioners
visited Salisbury in 1806 in search of it, but obtained no
satisfaction. It seems, however, to have been re-discovered
within the next few years, since it is mentioned in a book
published in 1814,^ and it is now exhibited to the public
by order of the Dean and Chapter of Salisbury Cathedral.
It resembles the Lincoln copy both in its fine, leisurely
penmanship and also in the absence of additions at the
bottom of the parchment.^
II. Comparison of the Originals, Prior to Sir William
Blackstone's work, extraordinary confusion seems to have
prevailed concerning the various Charters of Liberties.
Not only was John's Magna Carta confused with reissues
by Henry ; but these latter were known only from an official
copy of the Charter of 1225 contained in the Inspeximus
^ See James Tyrrell, History of England ^ Vol. II. 821 (1697-1704).
"^ Blackstone, Great Charter, p. xvii.
'See Report (1800), p. 341.
* Dodsworth, Historical Account of the Cathedral^ 202.
^ It is unnecessary to treat in detail of the copies of the charter not authenticated
by John's Great Seal, though some of these are of value as secondary authorities.
The four most important are {a) a copy appearing in the Register of Gloucester
Abbey, ijb) the Harleian MS., British Museum No. 746 (which also contains the
names of the twenty-five Executors in a hand probably of the reign of Edward I. ).
{c) in the Red Book of the Exchequer. There is also {d) an early French version,
printed in D'Achery, Spicilegium^ Vol. XII. p. 573, together with the writ of 27th
September addressed to the Sheriff of Hampshire. See Blackstone, Great Charter y.
p. xviii., and Thomson, Magna Carta, pp. 428-430.
MANUSCRIPTS AND RELATIVE DOCUMENTS 169
of the twenty-eighth year of Edward I. Neither Madox^
nor Brady ^ was aware of the existence of any one of the
four originals; and no mention is made of them in the
first edition of Rymer's Foedera, which appeared in 1704.
Mr. Tyrrell, indeed, seems to have known of the second
original in the British Museum and also of the Salisbury
version.^ Mr. Care* showed no clear knowledge of the
various manuscripts, though he mentioned the existence of
several. Even Blackstone in 1759 collated only the two
Cottonian copies, since he failed to find that of Salisbury,
and was unaware of the existence of the Lincoln manu-
script.^
As these four versions are practically identical in their
substance, no important question seems to be involved in
the discussion as to whether any one of them has greater
authority than the others. The R ecord Commissioners con-
sidered that the Lincoln copy was the first to be completed
(and therefore that it possessed special authority), because
it contained no insertions at the foot of the instrument.
Yet it seems more plausible to argue that this very
immunity from clerical errors, or from additions made
after engrossment, proves that it was of later and less
hurried execution than the others, and therefore of less
^Thomas Madox, Fii-ina Burgi (1726). On p. 45, Madox refers only to the
Inspeximus of Edward I.
2 Robert Brady, Complete History of England, p. 126 of Appendix to Vol. I.
(1685), takes his text of the Charter from Matthew Paris ** compared with the
manuscript found in Bennet College Library," i.e. Corpus Christi, Cambridge.
'James Tyrrell, History of England (1697-1704). In p. 9 of Appendix to
Vol. II. p. 821, Tyrrell prints a text of John's Charter founded on that of M. Paris,
collated with those two originals.
* Henry Care, English Liberties in the Freeborn stihjects'' inheritance ; containing
Magna Charta, etc. (1719), p. 5. The first edition, with a somewhat different
title, is dated 1691.
5 Strangely enough, Sir Thomas Duffus Hardy, so recently as 1837, in publishing
his Rotuli Chartarum (Introduction, p. ii. note 5) declared that no original of
John's Charter existed ; "notwithstanding all the care taken by multiplication of
copies, it is singular that no contemporary copy of King John's Magna Carta has
yet been found." The Lincoln MS. he dismissed as "certainly not of so early a
date." He further reasserts the fallacy, exposed by Blackstone eighty years
earlier, that John had issued a separate Carta de Foresta.
170 MAGNA CARTA
authority, if any distinction is permissible. Mr. Thomson
has much ground for his contention, in speaking of the
fire-marked version, that " the same circumstances may
probably be a proof of its superior antiquity, as having
been the first which was actually drawn into form and
sealed on Runnymede, the original whence all the most
perfect copies were taken." ^
In all printed texts of Magna Carta, the contents are
divided into a preamble and sixty-three chapters. There
is no warrant for this in any one of the four originals : the
" chapters " are a modern invention, made for convenience
of reference.
III. Articles of the Barons. Of hardly inferior interest
is the parchment which contains the heads of agreement
made between John and the rebels on 15th June, 12 15.
This is now in the British Museum, cited officially as
"Additional MSS. 4838." The seven centuries that have
passed over it have left surprisingly few traces ; it is legible
throughout, and still bears the impression of John's seal
in brown wax. It was apparently deposited in Lambeth
Palace, where it remained until the middle of the seven-
teenth century. Archbishop Laud, when threatened with
impeachment, thought it prudent to set his papers in order;
and on i8th December, 1640, he dispatched for that purpose
to his episcopal palace, his friend Dr. John Warner,
Bishop of Rochester. A few hours later, Laud was com-
mitted to custody of Black-Rod, and an official messenger
was sent by the House of Lords to seal up his papers.
Bishop Warner had, meanwhile, escaped with the Articles
of the Barons. He kept this till he died, and at his death
it passed to one of his executors named Lee, and from him
to his son Colonel Lee, who presented it to Gilbert Burnet.
When the Salisbury Magna Carta disappeared, Burnet
was suspected of appropriating it. What gave apparent
weight to these misrepresentations of his political opponents
was that special facilities had been granted him to search
public records in the prosecution of his historical labours,
and that he actually had in his possession — quite lawfully,
^ Thomson, Magna Carta, 422.
MANUSCRIPTS AND RELATIVE DOCUMENTS 171
as we know — the Articles of the Barons, which was con-
fused by the carelessness of early historians with Magna
Carta itself. The calumny was so widely spread that
Burnet thought it necessary formally to refute it, explain-
ing that he had received the Articles as a gift from Colonel
Lee : — " So it is now in my hands, and it came very fairly
to me." 1
Bishop Burnet left it as a legacy to his son Sir Thomas
Burnet; and on his death it passed to his executor David
Mitchell, whose permission to print it Blackstone obtained
in 1759. It was purchased from Mr. Mitchell's daughter
by another historian, Philip, second Earl of Stanhope, who
presented it to the British Museum in 1769. It is now
exhibited along with the two Cottonian copies of Magna
Carta. The Record Commissioners have reproduced it in
Statutes of the Realm in 1810, and also in the New Rymer
in 1816.2
The document begins with this headline : " Ista sunt
Capitula quae Barones petunt et dominus Rex concedit,"
Then the articles follow in 49 paragraphs of varying length,
separate, but unnumbered, each new chapter (unlike the
chapters of Magna Carta, which run straight on) beginning
a new line. The numbers, which appear in all printed
editions, have no warrant in the original.^
IV. The so-called " unknown Charter of Liberties." At
Paris is preserved a copy of what looks like a charter
granted by John, but irregular in its form. This document
is preserved among the Archives du Royaume in the Section
Historique and numbered J. 655.^ A copy of this copy
was discovered at the Record Office in London by Dr.
Horace Round in 1893. Before that date it seems to have
been practically unknown to English historians, although
it had been printed by a French writer thirty years earlier.^
As the interpretation of this document has proved a
^See Burnet's Own Time, I. 32 (edition of 1724).
2 Reproductions are sold by the British Museum at 2s. 6d.
'■^ Cf. st4pra, p. 39, and Blackstone, Greai Charter, xvii,
^ See the account by Mr. Hubert Hall, Eng. Hist. Rev.^ IX. 326.
^Teulet, Layettes du Trisor des Chart es, I. p. 423 (1863).
172 MAGNA CARTA
puzzle attracting many to attempt its solution, it may be
well to give a brief analysis of its tenor.^ The text of the
supposed Charter is preceded, in the manuscript (which is
in a French hand of the early quarter of the thirteenth
century), by a copy of the Charter of Liberties of Henry I.,
from which it is separated by this sentence, in Latin :
" This is the Charter of King Henry, by which the barons
seek their liberties, and these following are granted by King
John," words which invite comparison with the heading of
the Articuli Baronum, and suggest that the document under
description may have formed a link between Henry L's
charter and these Articuli.
The first clause runs in the third person (concedit rex
Johannes) and grants that he will arrest no man without
judgment, nor accept anything for justice, nor commit
injustice. The remaining eleven clauses are all in the first
person singular (whereas regular charters run in the plural).
The second clause restricts relief ; the third regulates ward-
ship ; the fourth, marriage ; the fifth, testate and intestate
succession; the sixth, the rights of widows. The seventh,
opening with the word " adhuc " (as though later additions
were now made to provisions previously written), concedes
that Crown vassals need not go on military service outside
of England except in Normandy and Brittany; and seems
further to suggest, in certain circumstances, a diminution
of the servitium dehitum. Clause 8 limits scutage to one
mark unless by counsel of the barons.
Clause 9, again beginning with adhucy agrees to give up
the forests made by Henry H. and Richard. Clause lo
(also with its adhuc) grants remission, in several particulars,
of the strictness of the forest laws. Clause ii prohibits
Jews from taking interest during a debtor's minority ; and
clause 12 concedes that no one shall lose life or limbs for
the killing of a deer.
At least seven solutions have been attempted of the pro-
blems raised by this manuscript, (i) Dr. Round, in com-
municating his discovery to the English Historical Review,
maintained that the document was a copy, in a mangled
^ See text in Appendix.
MANUSCRIPTS AND RELATIVE DOCUMENTS 173
form perhaps, of a charter actually granted in the year
1 2 13 by King John to the northern barons, containing
concessions which they had agreed to accept in satisfaction
of their claims. ^ (2) Mr. Prothero preferred to view it
as an abortive proposal made by the King early in 12 15
and rejected by the barons.^ (3) Mr. Hubert Hall dis-
missed the document as a forgery, describing it as " a
coronation charter attributed to John by a French scribe
in the second decade of the thirteenth century " — probably
between November, 12 16, and March, 12 17, when King
Philip desired to prove that John had committed perjury
by breaking his promises, and had thereby forfeited his
right to the Crown of England.^
(4) In the first edition of this work, published in 1905,
the tentative suggestion was made that the document might
be a copy of the actual " schedule " which we know from
Roger of Wendover ^ to have been drawn up by the barons
prior to 27th April, 12 15, and at that date forwarded to
John with the demand, under threat of civil war, that he
should forthwith set his seal to it. In this view the
schedule would be merely a precursor of the Articles of
the Barons, with which it had been previously identified.
The fact that this " schedule " was hurriedly drawn up by
unskilful hands was suggested as an explanation of the
peculiar features of the " unknown charter " emphasized by
Mr. Hall ; its archaisms, its erroneous royal style running
in the singular, and its transition from the third to the first
person. (5) Mr. Davis, ^ in rejecting this theory, main-
tained that the document contained the jottings made by
some one present while negotiations were actually in
progress between the barons and John's representatives at
some date between the drawing up of the Articuli Baronum
and the sealing of the Great Charter, presumably, there-
fore, between 15th and 19th June, 12 15.
(6) Mr. Petit-Dutaillis 6 modifies Mr. Davis's theory
^Eng. Hist. Rev., VIII. 288-294. ^ Ibid., IX. 117-121.
^Ibid., IX. 326-335. * Wendover, III. 298, and cf. supra, 33.
^Eng, Hist. Rev., XX. 719 ff. ^Studies SMpplemeritary, 120 ff.
174 MAGNA CARTA
materially. The conference, at which the unofficial note-
taker was present, must have taken place shortly before
the framing of the Articuli Baronum, and the note-taker
himself may have been an emissary of Philip Augustus,
possibly a spy of humble origin, collecting information in
furtherance of Philip's designs on England. (7) The
most recent, detailed, and ingenious theory is that of Dr.
Ludwig Riess of Berlin, ^ who thinks that a copy of the
first Henry's Charter was sent to John for convenience of
reference when the latter, amid the misfortunes of the ill-
starred campaign of 1 2 14, was trying to make terms with
the rebellious northern barons, and that jottings subse-
quently made on the blank space at the foot of the parch-
ment, as to concessions granted by John, constitute the
so-called " unknown charter."
Successive clauses of the document tell the story of its
genesis — and a romantic story it is. When the northern
barons met the demand of 26th May, 12 14, for a scutage,
by the counter demand for a confirmation of Henry Beau-
clerk's Charter, John's Regent, Peter des Roches, wrote
to the King, then in Poitou, for instructions, enclosing a
transcript of Henry's Charter, to which he had appended
a jotting to remind John of the promise already made on
28th August, 1 2 13, through Stephen Langton. This note
forms, in Dr. Riess's theory, clause one of the much dis-
cussed document. Thereafter a period of haggling ensued
between John and the distant rebels, with Peter and
perhaps also the archbishop as intermediaries, the King
making a careful memorandum from time to time of each
concession wrung from him by the obduracy of the barons.
The King is thus the author of clauses 2 to 12 inclusive,
couched in the informal first person singular, each new
group opening with the word " adhuc."
The original document, which thus represented the stages
of unsuccessful negotiations extending over several months,
was captured, so it is inferred, by the French. After a
copy had been made for preservation at Paris, the original
was sent by Philip to the barons that they might embarrass
"^ Historische Vierteljahrschrift, 1 910, 449-458.
MANUSCRIPTS AND RELATIVE DOCUMENTS 175
John by confronting him with concessions in his own
handwriting which he now desired to repudiate. When
Henry's Charter was produced by Stephen Langton at
Bury St. Edmunds on 4th November, 12 14, it was the royal
jottings appended to it, not the familiar, century-old
charter itself, that produced the sensation which modern
writers have found so hard to explain.
Such is Dr. Riess's brilliant effort at historical recon-
struction : the main difficulties to its acceptance are that
it involves too many unproved assumptions; that John,
before the failure of his schemes, was unlikely to authorize
substantial concessions, or to make careful memoranda of
them as though he meant to keep his promises; and that
five months, between May and October, would not suffice
for the conduct of protracted negotiations between John in
Poitou and the malcontents scattered through the north of
England.
It is beyond doubt, however, that offers and counter-
offers, of which the schedule of Easter was only one, passed
to and fro, between March and June of the year 12 15.
The negotiations of which our document contains a
record may have taken place between the respective dates
of the " schedule " and the Articuli. It would be easy to
explain the presence of a copy in the French archives on
the assumption that the original was among " the charters
of liberties" surrendered by Louis in 1217. This trifling
amendment would meet some of the objections to Dr.
Riess's theory, which in all essentials seems to be the most
convincing yet suggested. In any view, the " unknown
charter " would appear to be a link between the Charter of
1 100 and the Articuli.
It would clearly be inadvisable to found conclusions upon
a document, the nature and authenticity of which form the
subject of so many rival theories; but even if further
investigation proves it to be a forgery, a forgery of con-
temporary date may still throw light on otherwise obscure
passages in genuine charters. Instances of this will be
found in the sequel.
176 MAGNA CARTA
II. Previous Editions and Commentaries.
I. Printed Editions of the Text. Prior to 1759, even the
best informed writers on English history laboured under
much confusion in regard to the various charters of liberties.
Few seem to have been aware that fundamental differences
existed between the charter granted by John and the
reissues of Henry. Much of the blame must be borne by
Roger of Wendover, who, in his account of the transactions
at Runnymede, incorporated, in place of John's Charter,
the text of the two charters granted by Henry. ^
Early editions of " Magna Carta," then, are not of John's
Charter at all, but give the text of Edward's Inspeximus
of Henry's reissue of 1225. The very earliest of these to
be printed was apparently that published on 9th October,
1499, by Richard Pynson, the King's printer.^ The same
document was followed in numerous editions by Pynson,
Redman, Berthelet, Tottel, Marshe and Wight, from 1499
to 1618. It was not until Blackstone's day, however,
that John's Charter appeared in print. Of the numerous
editions that have since appeared, only four call for
separate notice.
(i) In 1759 appeared Sir William Blackstone's scholarly
work entitled The Great Charter and the Charter of the
Forest, containing accurate texts of all the important issues
of the Charters of Liberties carefully prepared from the
original manuscripts so far as these were known to him.^
(2) In some respects the Record Commissioners have
improved even on Blackstone's work, in their edition of
the Statutes of the Realm, published in 1810. A special
section of the volume is devoted to Charters of Liberties,
where not only the grants of John and Henry III., but
also the charters which led up to them, and their subse-
quent confirmations, have received exhaustive treatment.
iR. Wendover, III. 302-318.
2 This date is given by Bemont, Chartes, Ixxi., but Robert Watt in his
Bibliotheca Britamtica, Thomson, Magna Carta, 450, and Lowndes, Biblio-
grapher's Manual, 1449, all give the date of the earliest edition as 1514.
'The substance of this admirable edition, now unhappily scarce, has been
reproduced in the same author's Tracts (1762).
PREVIOUS EDITIONS AND COMMENTARIES 177
(3) A carefully revised text, Magna Carta regis Johannis,
was published by Dr. Stubbs in 1868; and the various
charters are also to be found, arranged in chronological
order, in his well-known Select Charters, first published
in 1870.
(4) In 1892, M. Charles Bemont published carefully
edited texts of the charters of 12 15 and 1225, printing as
footnotes to the latter the variants of 12 16 and 1217.
II. Commentaries and Treatises. Within five years of
the peace made at Runnymede, a minstrel attached to
Robert of Bethune, one of John's familiars, included an
incomplete but not inaccurate summary of the Charter in
his Histoire des dues de Normandie et desrois d'Angleterre,
supposed to have been composed in 1220.^ This first rude
commentary has already been alluded to.^ Posterity would
gladly have bartered it, such as it is, for a few words of
explanation from one who was well able to speak but pre-
ferred to keep silence. The discreet biographer of William
the Marshal excuses himself from drawing upon his inti-
mate sources of information : he must pass over, he says,
the war which was in England between the King and his
barons, for there were too many incidents which it would
not be honourable to recount.^
Later in the century, comes the mysterious medieval
lawbook known as the Mirror of Justices, complaining of
" the damnable disregard " of Magna Carta and containing
a .chapter on that document with some claims to rank as
a commentary, although it represents the opinions of a
political pamphleteer rather than those of an unbiassed
judge. The date of this treatise is still the subject of
dispute. It has been usual to place it not earlier than
the years 1307-27, mainly because it makes mention of
" Edward II." Prof. Maitland, however, dates it earlier,
maintaining on general grounds that it was " written very
soon after 1285, ^^d probably before 1290." * He explains
1 Published in 1840 (edited by F. Michel). "Supra, p. 123.
3 G. le Mart<chal, 1 503 1 ft.
* See The Mirror of Justices (edited for the Selden Society by W. J. Whittaker),
Introduction (by Maitland), xxiii. to xxiv.
M
178 MAGNA CARTA
the reference to " Edward II." as applying to the monarch
now generally known in England as Edward I., but some-
times in his own reign known as Edward II., to distinguish
him from an earlier Edward still enshrined in the popular
imagination, namely, Edward Confessor. Mr. Maitland
is not disposed to treat this work of an unknown author
too seriously, and warns students against " his ignorance,
political bias, and deliberate lies."^
Reference has already been made to the comparative
neglect of Magna Carta in the fifteenth and sixteenth
centuries, and to the influence of Coke in reviving interest
in its provisions. Of the commentaries that have subse-
quently appeared, it is not, perhaps, necessary to mention
more than the following thirteen, (i) The elaborate treatise
of Sir Edward Coke, King James's deposed Chief Justice,
comprising the second of his four Institutes, was published
in 1642 under direction of the Long Parliament, the House
of Commons having given the order on 12th May, 1641.2
Although this commentary, like everything written by
Coke, was long accepted as a work of great value, its
method is in reality uncritical and unhistorical. " The great
lawyer reads into Magna Carta the entire body of the
common law of the seventeenth century j-^ of which he w^as
admittedly a master. He seems almost unconscious of the
changes wrought by the experience and vicissitudes of four
eventful centuries. The clauses of Magna Carta are merely
occasions for expounding the law as it stood, not at the
beginning of the thirteenth century, but in Coke's own day.
In the skilful hands of Sir Edward, the Great Charter is
made to attack abuses of James or Charles, rather than
those of John or Henry. In expounding the judicium
parium, for example, he explains minute details of proce-
dure before the Court of the Lord High Steward, and the
nature of the warrants to be issued prior to arrest of any
one by the Crown ; while in the clause of Henry's Charter
which secures an open door to foreign merchants in Eng-
land " unless publicly prohibited," he discovers a declara-
^See T/ie Mirror of Justices^ xxxvii. Cf. xlviii.
2 See Dictionary of National Biography, XL 243.
PREVIOUS EDITIONS AND COMMENTARIES 179
tion that Parliament shall have the sole power to issue
such prohibitions, forgetful that " Parliament " did not
exist in 12 15, and that the regulation of trade was then
an exclusive prerogative of the Crown.
(2) In 1680 Edward Cooke, barrister, published a small
volume entitled Magna Charta made in the ninth year of
King Henry III, and confirmed by King Edward I, in the
twenty-eighth year of his reign. This contained a trans-
lation of Henry's Magna Carta with short explanatory
notes founded mainly on the commentary of Sir Edward
Coke. Mr. Cooke declared that his object was to make
the Great Charter more accessible to the public at large,
since, as he said, " I am confident, scarce one of a hundred
of the common people, know what it is."
(3) Sir William Blackstone's Introduction to his edition
of the charters, published in 1759, as already mentioned,
contains valuable information as to the documents he edits ;
but he explicitly disclaims all intention of writing a Com-
mentary. He is careful to state " that it is not in his
present intention, nor (he fears) within the reach of his
abilities, to give a full and explanatory comment on the
matters contained in these charters." ^
(4) Daines Harrington published in 1766 his Observations
upon the Statutes from Magna Charta to 21 James I. This
book contains some notes on the Charter also founded
chiefly upon Coke's Second Institute ; his original con-
tributions are not of outstanding value.
(6) In 1772 Prof. F. S. Sullivan issued a course of lectures
under the title An Historical Treatise on the Feudal Law,
with a Commentary on Magna Charta : " I shall therefore
proceed briefly to speak to Magna Charta, and in so doing
shall omit almost all that relates to the feudal tenures, which
makes the greatest part of it, and confine myself to that
which is now law."^
(7) John Reeves' invaluable History of English Law,
which appeared in 1783-84, marked the commencement of a
new epoch in the scientific study of the genesis of English
law. Treating incidentally of Magna Carta, he shows
^ Introduction, p. ii. ^ P. 375 of work cited.
i8o MAGNA CARTA
wonderful insight into the real purport of many of its pro-
visions, but the state of historical knowledge when he wrote
rendered serious errors inevitable.
(8) In 1829 Richard Thomson published an elaborate
edition of the charters, combined with a commentary which
makes no serious attempt to supplement the unhistorical
explanations of Coke by the results of more recent investi-
gations. His work is a storehouse of information which
must, however, be used with caution.
(9) In many respects, the most valuable contribution yet
made to the elucidation of the Great Charter is that con-
tained in M. Charles Bemont's preface to his Charles des
Liberies Anglaises, published in 1892. Although he has
subjected himself to the severe restraints imposed by the
slender size of his volume and by a rigid desire to state
only facts of an undisputed nature, leaving theories strictly
alone; he has done much to help forward the study of
the charters, insisting upon the close mutual connection
between the various Charters of Liberties. It is doubtful,
however, whether by this very insistence upon the continuity
of this one series of documents, he does not lay himself
open to the misconception that he takes too narrow a view
of the scope and relations of the Charter. Magna Carta's
antecedents must not be looked for exclusively among
documents couched in the form of charters, nor its results
merely in their subsequent confirmations. It is impossible
to understand it aright, except in close relation to all the
varied aspects of the national life and development. Every
Act appearing on the Statute Rolls is, in a sense, an Act
amending Magna Carta; while such enactments as the
Statutes of Marlborough and Westminster I. have as
intimate a connection with John's Great Charter as the
Confirvialio Cartarum or the Arliculi super Cartas have.
This is a truth which M. Bemont recognizes, though the
scheme of his book led him to emphasize another aspect
of his subject. His object was not to explain the numerous
ways in which the Charters of Liberties are entwined with
the whole of English history, but merely to furnish a basis
for the accurate study of one of their most important
PREVIOUS EDITIONS AND COMMENTARIES i8i
features. His book is indispensable, but is not intended
to form, in any sense, a commentary on Magna Carta.
(id) a brilliant article by Mr. Edward Jenks appeared
in The Independent Review for November, 1904, whose
title, The Myth of Magna Carta, indicates the iconoclastic
lines on which it proceeds. He argues that the Charter
was the product of the selfish action of the barons pressing
their own interests, and not of any disinterested or national
movement ; that it was not, by any means, " a great land-
mark in history " ; and that, instead of proving a material
help in England's advance towards constitutional freedom,
it was rather " a stumbling block in the path of progress, "
being feudal and reactionary in its intention and effects.
Finally, for most of the popular misapprehensions con-
cerning it, he holds Coke responsible.
(11) In The Magjia Carta of the English and of the
Hungarian Constitution (1904), Mr. Elemer Hantos ably
analyzes the numerous and interesting parallels between
John's Charter and the Bulla Aurea of Andreas II.,
dating from 1222, and furnishes a brief commentary on
both.
(12) M. Charles Petit-Dutaillis, in his Etude sur la vie
et le regne de Louis VIII. (1894), was one of the first of
modern historians to deprecate exaggerated estimates of
the value of Magna Carta, insisting that " the barons had
no suspicion that they would one day be called the founders
of English liberty." ^ More recently, in his Studies and
Notes supplementary to Stuhbs' Constitutional History ^
he has included a brief but valuable discussion of the Great
Charter.
•^ (13) The whole of Prof. G. B. Adams' The Origin of the
English Constitution (191 2) is virtually a discussion of the
Great Charter, and abounds in valuable suggestions for
estimating its tenor and value, and for elucidating its
various clauses. It does not aim at being an exhaustive
1 P. 57 of work cited.
2 This is the title of the English translation by Mr. W. E. Rhodes (1908) of the
Appendices to the first volume of a French version of Stubbs' Const. Hist.^
published in 1907.
i82 MAGNA CARTA
treatise, but is intended to supplement rather than super-
sede existing commentaries.^
^ Of the books and articles containing incidental references to Magna Carta, it
is unnecessary to speak ; those containing comments on isolated chapters or
particular aspects are mentioned infra in their appropriate places. The late Mr.
Harcourt's His Grace the Steward and Trial of Peers contains a vigorous com-
mentary on chapter 39, and his article **The Amercement of Barons by their
Peers" {Eng. Hist. Rev., XXII. 732), on chapter 21. The first edition of the
present work (published, 1905) evoked a number of valuable contributions to
various aspects of the subject ; among these may be mentioned Vinogradoff, Law
Quart. Rev.y XXI. 250-7 ; Liebermann, Historische Vierteljahrschrift, 1907,
231-5; Bemont, Revue Historique, 1907, 122-4; Petit-Dutaillis, Le Moyen Age,
1906, 277-282 ; H. W. C. Davis, Eng. Hist. Rev. (1905), XX. 719-726 ; Neilson,
Juridical Reviezv, June, 1905, 128-144. See 2\%o Jurid. Rev., March, 1905, 61 ;
and Law Notes (New York), August, 1905, 94-6 for some legal decisions, Scotch
and American respectively.
TEXT, TRANSLATION, AND
COMMENTARY
MAGNA CARTA.
PREAMBLE.i
Johannes Dei gratia rex Anglie, dominus Hibernie, dux
Normannie et Aquitannie, et comes Andegavie, archi-
episcopis, episcopis, abbatibus, comitibus, baronibus,
justiciariis, forestariis, vicecomitibus, prepositis, ministris
et omnibus ballivis et fidelibus suis salutem. Sciatis nos
intuitu Dei et pro salute anime nostre et omnium ante-
cessorum et heredum nostrorum, ad honorem Dei et exalta-
tionem sancte Ecclesie, et emendacionem regni nostri, per
consilium venerabilium patrum nostrorum, Stephani Can-
tuariensis archiepiscopi tocius Anglie primatis et sancte
Romane ecclesie cardinalis, Henrici Dublinensis archi-
episcopi, Willelmi Londoniensis, Petri Wintoniensis,
Joscelini Bathoniensis et Glastoniensis, Hugonis Lincoln-
iensis, Walteri Wygorniensis, Willelmi Coventriensis, et
Benedicti Roffensis episcoporum ; magistri Pandulfi
domini pape subdiaconi et familiaris, fratris Aymerici
magistri milicie Templi in Anglia; et nobilium virorum
Willelmi Mariscalli comitis Penbrocie, Willelmi comitis
Sarresburie, Willelmi comitis Warennie, Willelmi comitis
Arundellie, Alani de Galeweya constabularii Scocie,
Warini filii Geroldi, Petri filii Hereberti, Huberti de Burgo
*The division of Magna Carta into a preamble and sixty-three chapters is a
modern device for which there is no warrant in the Charter. Cf. supra^ 170.
No title or heading precedes the substance of the deed in any one of the four
known originals, but on the back of the Lincoln MS. (cf. supra^ 167) these
words are endorsed ; " Concordia inter Regent Johannem et Barones pro concessione
Hbertatum ecclesie et regni Anglie.'''' The form of the document is discussed
supra^ 104-9. The text is taken from that issued by the Trustees of the British
Museum founded on Cottonian version No. 2. Cf. supra, 166.
i86 MAGNA CARTA
senescalli Pictavie, Hugonis de Nevilla, Mathei filii Here-
berti, Thome Basset, Alani Basset, Philippi de Albiniaco,
Robert! de Roppeleia, Johannis Mariscalli, Johannis filii
Hugonis et aliorum fidelium nostrorum. «
John, by the grace of God, king of England, lord of Ireland,
duke of Normandy and Aquitaine, and count of Anjou, to the
archbishops, bishops, abbots, earls, barons, justiciars, foresters,
sheriffs, stewards, servants, and to all his bailiffs and liege
subjects, greeting. Know that, having regard to God and for
the salvation of our souls, and those of all our ancestors and
heirs, and unto the honour of God and the advancement of
holy Church, and for the reform of our realm, [we have
granted as underwritten] ^ by advice of our venerable fathers,
Stephen, archbishop of Canterbury, primate of all England and
cardinal of the holy Roman Church, Henry archbishop of
Dublin, William of London, Peter of Winchester, Jocelyn of
Bath and Glastonbury, Hugh of Lincoln, Walter of Worcester,
William of Coventry, Benedict of Rochester, bishops ; of master
Pandulf, subdeacon and member of the household of our lord
the Pope, of brother Aymeric (master of the Knights of the
Temple in England), and of the illustrious men^ William
Marshal, earl of Pembroke, William, earl of Salisbury, William,
earl Warenne, William, earl of Arundel, Alan of Galloway
(constable of Scotland), Waren Fitz Gerald, Peter Fitz Herbert,
Hubert de Burgh (seneschal of Poitou), Hugh de Neville,
Matthew Fitz Herbert, Thomas Basset, Alan Basset, Philip
d'Aubigny, Robert of Roppesley, John Marshal, John Fitz
Hugh, and others, our liegemen.
The Great Charter of John opens, in the form common
to royal charters of the period, with a greeting from the
sovereign to his magnates, officials, and faithful subjects,
and announces, in the pious legal formula used by impious
and pious kings alike, that he had made certain grants by
the advice of counsellors whom he names. Three features
call for comment.
I. The King's Title. Points of interest are suggested by
the form of royal style here adopted. John's assumption
of the royal plural " Sciatis Nos " reads, in the light of
^ The sentence is concluded in chapter one (see infra) — the usual division, here
followed, being a purely arbitrary one.
2 The phrase ^''nobiles virV was not used here in any technical sense; the
modern conception of a distinct class of "noblemen" did not take shape until
long after 1215. Cf. what is said of "peerage" under cc. 14 and 39.
PREAMBLE 187
subsequent history, as a tribute to his arrogance rather
than his greatness, when compared with the humbler first
person singular used by his father. In this particular,
however, Richard, not John, had been the innovator. ^
For a further alteration, John was alone responsible : to
the titles borne by his father and brother, he added
that of " lord of Ireland." When the wide territories of
Henry II., had been distributed among his elder sons, the
young John (hence known as " John Lackland ") was left
without a heritage, until his father bestowed on him the
island of Ireland, recently appropriated ; and this brought
with it the right to style himself " do minus Hiberniae," a
title retained after he had outlived his brothers and inherited
their wide lands and honours.
John began his reign in 1199 as ruler over the undivided
possessions of the House of Anjou from the Cheviots to
the Pyrenees. These lands were held, by him as by his
father, under a variety of titles and conditions. Anjou,
the original fief of the Plantagenet race, still carried with
it only the title of count. Henry II. had, at an early age,
become duke of Normandy in his mother's right, and there-
after duke of Aquitaine by marriage with Eleanor, its
heiress.^ These fiefs were held by Henry and his sons
under the King of France as Lord Paramount. Long
before 12 15, John had lost these wide dominions, except
the most distant of them all, his mother's dowry of Aqui-
taine. Anjou and Normandy were irretrievably lost, but
he still retained their empty titles; and in this Henry III.
followed him at first, until, by the Treaty of 1259, he
surrendered to Louis IX. all claim to Normandy and Anjou
with their dependencies, in return for a confirmation of his
claims on Aquitaine.^
Of Ireland, John was still, as formerly, " lord " not
" king." ^ The exact denotation of " dominus " has formed
^Coke [Second Institute, pp. 1-2) errs in attributing the change to John.
2 Aquitaine included Poitou and Gascony with the four dependent counties of
Angouleme, La Marche, Limoges and Perigord. See Norgate, Minority, 132.
' See Giry, Mamiel de diplomatique, 798.
* Henry VIII. was the first to call himself *' King of Ireland "—a singular proof
"of the success of Henry's policy." Gairdner, Lollardy, ii. 473.
l88 MAGNA CARTA
the subject of learned controversy. It is not, as has some-
times been suggested, an inferior title to that of rex,
appropriate only to a preliminary stage of the process
culminating in kingship. The two words imply distinct
relationships differing in kind. The one is national and
the other personal and feudal. Kingship is conferred by
" election " (or at least proclamation) followed by corona-
tion; lordship depends on the feudal contract made with
the individual vassal, by homage and fealty.^ England,
alone of John's possessions, was held by the style of
"Rex," implying sovereign rule, although John in 1213
had accepted Innocent as feudal overlord. In calling him-
self "Rex Angliae," in place of "Rex Anglorum" (as
Henry I. had done), he followed precedents of Stephen and
of Henry II. 2
No vindication of John's title is given. The simple
words, "Dei gratia rex Angliae,'' may be contrasted with
the laboured attempt of Stephen's second and more formal
charter of liberties (of April, 1136) to set forth a valid title
to the throne; where he describes himself as appointed
(" electus ") by consent of clergy and people ; consecrated
by William, Archbishop of Canterbury and Legate of Holy
Roman Church; and thereafter confirmed by Innocent,
Pontiff of the Holy See of Rome.^
Conscious of the claims of his cousin Matilda, Stephen
here ignores the element of hereditary succession in deter-
mining the title to the Crown, and emphasizes the element
of appointment or "election," both of which were blended
in the twelfth, as in earlier centuries, in proportions not
easy to define with accuracy. Professor Freeman pushed
to excess the supposed right of the Witenagemot to elect
the King, and transferred it to the Norman Curia. A
recent German writer, Dr. Oskar Rossler,* denies that the
^Cf. supra, p. 95. See Orpen, Ireland, I. 300 and II. 31, where it is pointed
out that William Marshal refused to support his King against his "lord." For
other theories, see Round's Mandeville, 70; Rossler's Mat tide, 291-4 and 424;
Ramsay's Fojindations, II. 403; Davis, England tinder Normans, 170.
^Stubbs, Early English History, p. 122, seems to be in error here.
^ See Charter in Appendix. * Matilde, passim.
PREAMBLE 189
Normans admitted the elective element at all. The theory
now usually held is a mean between these extremes, namely,
that the Norman Curia had a limited right of selecting
among the sons, brothers, or near relations of the last King,
the individual best suited to succeed him.^ Such a right,
never authoritatively enunciated, gradually sank to an
empty formality. Its place was taken, to some extent, by
the successful assertion by the spiritual power of a claim
to give or withhold the consecrating oil, without which no
one could be recognized as rex. John, secure in possession,
contents himself with the terse assertion of the fact of king-
ship : " John, by God's grace. King of England."
11. The Names of the consenting Nobles. It was
natural that the Charter should place on record the assent
of those magnates who remained in at least nominal allegi-
ance, and were therefore capable of acting as mediators.^
The leading men in England during this crisis may be
arranged in three groups: (i) the leaders of the host
opposed to John at Runnymede ; (2) the agents of John's
oppressions, extreme men, mostly aliens, many of whom
were in command of royal castles or of mercenary levies;
and (3) moderate men, churchmen or John's ministers or
relations, who, whatever their sympathies might be,
remained in allegiance to the King and helped to arrange
terms of peace — a comparatively small band, as the paucity
of names recited in Magna Carta testifies.^ The men, here
made consenters to John's grant, are again referred to,
^ See, however, Chadwick, Anglo-Saxon Institutions, p. 355 ff.
^Dr. Stubbs, Const. Hist., I. 582, gives the motive of thus naming them as
*' the hope of binding the persons whom it includes to the continued support of the
hard-won liberties." Those named were all moderate men. M. Paris (Chron.
Maj., II. 589) describes them as ** qttasi ex parte regis.'''' Cf. Annals of Dmistabky
III. 43. The neutrality of the prelates is proved by other evidence, (i) C. 62
gave them authority to certify by letters testimonial the correctness of copies of the
Charter. (2) The 25th of the Articles of the Barons left to their decision whether
John should enjoy a crusader's privileges ; while c. 55 gave Langton a special place
in determining what fines were unjust. (3) The Tower of London was placed in
the custody of the archbishop. (4) Copies are preserved of two protests by the
prelates in favour of the King. See Appendix.
^ Cf. supra, 36 ; for biographical information see authorities there cited.
190 MAGNA CARTA
though not by name, in chapter 63, in the character of
witnesses.
III. The Motives of the Grant, The preamble contains
a statement of John's reasons for conceding the Charter.
Tiiese are quaintly paraphrased by Coke : ^ " Here be four
notable causes of the making of this great charter rehearsed.
I. The honour of God. 2. For the health of the King's
soul. 3. For the exaltation of holy church, and fourthly,
for the amendment of the Kingdom." The real reason
must be sought in another direction, namely, in the army
of the rebels; and John in after days did not scruple to
plead consent given under threat of violence, as a reason for
voiding his grant. The technical legal " consideration,"
the quid pro quo which John received as the price of this
confirmation of their liberties was the renewal by his
opponents of the homage and fealty that they had solemnly
renounced. This " consideration " was not stated in the
charter, but the fact was known to all.^
CHAPTER ONE.
In primis concessisse Deo et hac presenti carta nostra con-
firmasse, pro nobis et heredibus nostris in perpetuum, quod
Anglicana ecclesia libera sit, et habeat jura sua Integra, et
libertates suas illesas; et ita volumus observari; quod
apparet ex eo quod libertatem electionum, que maxima et
magis necessaria reputatur ecclesie Anglicane, mera et
spontanea voluntate, ante discordiam inter nos et barones
nostros motam, concessimus et carta nostra confirmavimus,
et eam obtinuimus a domino papa Innocencio tercio con-
firmari ; quam et nos observabimus et ab heredibus nostris
in perpetuum bona fide volumus observari.^ Concessimus
eciam omnibus liberis hominibus regni nostri, pro nobis
et heredibus nostris in perpetuum, omnes libertates sub-
"^ Second Institute, I n. "Cf. sztpra, 40.
' Some editions place here the division between c. i and c.
CHAPTER ONE 191
scriptas, habendas et tenendas eis et heredibus suis, de nobis
et heredibus nostris.
In the first place we have granted to God, and by this our
present charter confirmed for us and our heirs for ever that the
EngHsh church shall be free, and shall have her rights entire,
and her liberties inviolate ; and we will that it be thus observed ;
which is apparent from this that the freedom of elections, which
is reckoned most important and very essential to the English
church, we, of our pure and unconstrained will, did grant, and
did by our charter confirm and did obtain the ratification of the
same from our lord. Pope Innocent III., before the quarrel
arose between us and our barons : and this we will observe, and
our will is that it be observed in good faith by our heirs for
ever. We have also granted to all freemen of our kingdom, for
us and our heirs forever, all the underwritten liberties, to be had
and held by them and their heirs, of us and our heirs forever.
This first of the sixty-three chapters of Magna Carta
places side by side, bracketed equal as it were, (a) a general
confirmation of the privileges of the English church, and
(b) a declaration that the rights to be afterwards specified
were granted " to all freemen " of the kingdom and to their
heirs for ever. The manner of this juxtaposition of the
church's rights with the lay rights of freemen, suggests an
intention to make it clear that neither group was to be
treated as of more importance than the other. If the civil
and political rights of the nation at large occupy the bulk
of the Charter, and are defined in their minutest details, the
church's rights receive a prior place. ^ A twofold division
thus suggests itself.
I. The Rights of the Church. A general promise that
the English church should be free was accompanied by
specific confirmation of the separate charter, guaranteeing
freedom of canonical election, granted on 21st November,
1 214. (i) Quod Anglicana ecclesia libera sit. This
emphatic declaration, which has no counterpart in the
Articles of the Barons, is repeated twice in Magna Carta, at
the beginning and the end respectively. If the original
scheme of the barons showed no special tenderness for
churchmen's privileges, Stephen Langton and his bishops
were careful to have that defect remedied. It is interesting
^ Cf. su/>ra, p. 39.
192 MAGNA CARTA
to note that, where the charters of Henry II. and earUer
Kings spoke of "holy church," Magna Carta speaks of
" ecclesia Anglicana." When EngHsh churchmen found
that the tyrant, against whom they made common cause with
English barons and townsmen, received sympathy and sup-
port from Rome, the conception of an English church that
was something more than a mere branch of the church
universal, began to take clearer shape. The use of the*
words ecclesia Anglicana may indicate, perhaps, that under
the influence of Stephen Langton, English churchmen were
beginning to regard themselves as members of a separate
community, that looked for guidance to Canterbury rather
than to Rome. John was now the feudal dependent of the
Holy See, and the " liberty of the English church " had to
be vindicated against the King and his lord paramount : the
phrase had thus an anti-papal as well as an anti-monarchical
bearing.
In promising that the English church should be free,
John used a phrase that was deplorably vague ; it scarcely
needed stretching, to cover the widest encroachments of
clerical arrogance. Yet the formula was by no means a
new one: both Henry I. and Stephen had confirmed the
claim of holy church to its freedom. ^
Henry II. had agreed in 1173 to give greater freedom of
elections, and in 11 76 that he would not keep sees vacant
for longer than one year,^ but avoided sweeping promises of
unlimited freedom. His whole reign, indeed, was an effort,
not unsuccessful, in spite of the disastrous consequences of
Becket's murder, to deprive the English church of what she
considered her freedom. John in 1215 receded from the
ground occupied by his father, confirming by the Great
Charter the promise given by the weakest of his Norman
predecessors, in a phrase repeated in all subsequent
confirmations.
It by no means follows that "freedom of the church,"
as promised by Stephen, meant exactly the same thing as
" freedom of the church " promised by John and his succes-
* See their Charters in Appendix.
"See Makower, ConsL Hist, of the Churchy 26, 315.
CHAPTER ONE 193
sors. The value to be attached to such assurances varied
in inverse ratio to the strength of the Kings who made
them, and this is well illustrated by a comparison of the
charters of Htenry I., Stephen, and John. Henry used
words, which may possibly be interpreted as defining and
restricting the grant of freedom, ^ until it meant little more
than freedom from the graver abuses of Rufus' reign.
Stephen's charter, on the contrary, supplements the same
phrase by definite declarations that the bishops should have
sole jurisdiction over churchmen and their goods, and that
all rights of wardship over church lands were renounced,
thus making it a " large and dangerous promise." ^
"Freedom of the church" had come in 1136 to include
" benefit of clergy " in a specially sweeping form, and much
besides.^ It is easy to understand why churchmen cher-
ished an elastic phrase which, wide as were the privileges it
already covered, might readily be stretched wider. Lay-
men, on the contrary, contended for a more restrictive
meaning; and the Constitutions of Clarendon must be
viewed as an attempt to settle disputed points of interpreta-
tion. Henry II. substantially held his ground, in spite of
his nominal surrender after Becket's murder. Thanks to
his firmness, " the church's freedom " shrank to more
reasonable proportions, so that the well-known formula,
when repeated by John, was emptied of much of the content
found in it by Stephen's bishops. Chapter 18 of Magna
Carta embodied, apparently with the approval of all classes,
the principle that questions of church patronage (assizes of
darrein presentment) * should be settled before the King's
Justices, a concession to the civil power inconsistent with
the more extreme interpretations formerly put by church-
men on the phrase.
In later reigns, the pretensions of the church to privileged
treatment were reduced to narrow bounds, and the process
of compression was facilitated by that very elasticity on
which the clergy had relied as being favourable to the
expansion of their claims. It was the civil government
1 Cf. supra, p. 97. 2 cf^ Pollock and Maitland, I. 74.
^Cf. supra, pp. 102-3. * For explanation see infra, c. 18.
N
194 MAGNA CARTA
which benefited in the end from the vagueness of the words
in which Magna Carta declared quod Anglicana ecclesia
libera sit.^
(2) Canonical Election. The charter, granted to the
church on 21st November, 12 14, had been reissued on 15th
January .2 Its tenor may be given in three words, " freedom
of election." In all cathedral and conventual churches and
monasteries, the appointment of prelates was to be free from
royal intervention for the future, provided always that
licence to fill the vacancy had first been asked of the King.
The bishops present at Runnymede succeeded in having
this concession inserted in the very forefront of Magna
Carta.
Henry III. in his reissues was made to repeat the phrase
quod Anglicana ecclesia libera sit, but omitted all reference
alike to canonical election and to John's charters to the
church. With the Pope's connivance or support, he
reduced the rights of cathedral chapj;ers to the sinecure they
had been before 12 15. It is true that Henry was prone to
lean on the papal arm, and that the Curia at Rome rather
than the Curia Regis often dominated appointments to
vacant sees : the canons elected the nominee of king or
pope, as each was, for the moment, in the ascendant.^ In
spite of Magna Carta, the independence of the English
church retrograded during the long alliance between Henry
III. and successive occupants of the papal throne.*
1 Mr. J. H. Round {Geoffrey de Mandeville, 3), speaking of Stephen's *' oath " to
restore the church her ** liberty," describes this as " a phrase the meaning of which
is well known." If "well" known, it was known chiefly as something which
baffled definition, because churchmen and laymen could never agree as to its
contents, while it tended also to vary from reign to reign. Mr. Round attempts
no definition. Sir James Ramsay {Angevin Empire, p. 475), writing of the phrase
as used in John's Charter, is less prudent. " It would relieve the clergy of all lay
control, and of all liability to contribute to the needs of the State beyond the
occasional scutages due from the higher clergy for their knights' fees." This
definition would not have satisfied John.
2 Cf. supra, p. 33. The text will be found in Statutes of the Realm, I. 5, and in
New Rymer, I. 126-7. It was confirmed by Innocent on 30th March, 12 15. See
Potthast, Regesta pontificum romanorum. No. 4963.
^Cf. supra, p. 141.
^Cf. Prothero, Simon de Montfort, p. 152. "The English church was indeed
CHAPTER ONE 195
II. Civil and Political Rights. After providing thus
briefly for the church, chapter one proceeds to give equal
prominence, but at greater length, to the grant or confirma-
tion of secular customs and liberties. A general enacting
clause leaves details to the remaining sixty-two chapters of
the Charter. Some of the more important points involved
have already been discussed in the Historical Introduction
— for example, the feudal form of the grant, better suited,
according to modern ideas, to the conveyance of a specific
piece of land, than to the securing of the liberties of a
mighty nation ; and the vexed question as to what classes
were intended, under the description of "freemen," to
participate in these rights.^
Another interesting point, though of minor importance,
calls. for separate treatment. John does not state that his
grants of civil and political rights had been made spon-
taneously. Whether deliberately or not, there is here a
marked distinction between the phraseology applied to
secular and to ecclesiastical rights respectively. While the
concessions to churchmen are said to have been granted
" mera et spontanea voluntate," no such statement is made
about the concessions to freemen. John may have favoured
this omission with an eye to the future repudiation of the
Great Charter on the ground that it had been sealed by him
under compulsion. Perhaps it was to anticipate the repeti-
tion of such arguments that the words spontanea et bona
voluntate nostra were inserted in the preamble of the reissue
of 1225, which had been purchased by a liberal grant.^
less independent of the king in 1258 than in 12 15, and far less independent of the
Pope than in the days of Becket."
^See supray pp. 104 and 114. For the meaning of "freeman" and Coke's
inclusion of villeins under that term for some purposes but not for others, see infra,
cc. 20 and 39.
^Cf. supra, 154, where the bearing of these words is discussed.
196 #.. MAGNA CARTA
CHAPTER TWO.
Si quis comitum vel baronum nostrorum, sive aliorum
tenendum de nobis in capite per servicium militare, mortuus
fuerit, et cum decesserit heres suus plene etatis fuerit et
relevium debeat, habeat hereditatem suam per antiquum
relevium ; scilicet heres vel heredes comitis de baronia
comitis Integra per centum libras ; heres vel heredes baronis
de baronia integra per centum libras; heres vel heredes
militis de feodo militis integro per centum solidos ad plus;
et qui minus debuerit minus det secundum antiquam con-
suetudinem feodorum.
If any of our earls or barons, or others holding of us in chief
by military service shall have died, and at the time of his death
his heir shall be full of age and owe "relief" he shall have his
inheritance on payment of the ancient relief, namely the heir or
heirs of an earl, ;^ioo for a whole earl's barony; the heir or
heirs of a baron, ^loo for a whole barony; the heir or heirs of
a knight, loos. at most for a whole knight's fee; and whoever
owes less let him give less, according to the ancient custom
of fiefs.
Preliminaries concluded, the Charter attacked what was,
in the barons' eyes, the chief of John's abuses, his arbitrary-
increase of feudal obligations. The Articles of the Barons,
indeed, had plunged at once into this most crucial question
without a word by way of pious phrases or legal formulas.
I. Assessment of Reliefs. Each " incident " had its own
possibilities of abuse, and the Great Charter deals with these
in turn. The present chapter defines the reliefs to be hence-
forth paid to John.i Vagueness as to the amount due was
a natural corollary of doubts as to whether the hereditary
principle was binding : the lord took as much as he could
grind from the inexperience or timidity of the youthful
vassal.
A process of definition, however, was early at work : some
conception of a " reasonable relief " was evolved. Yet the
criterion varied.^ Henry I., when bidding against duke
^Cf. supra, p. 59.
2 At an early date, in the midland counties, the thegn with more than six
CHAPTER TWO 197
Robert for the throne, was willing, in words if not in prac-
tice, to accept the limits set by contemporary opinion. His
Charter of Liberties promised that reliefs should be " just
and lawful " — an elastic phrase, liberally interpreted by
exchequer officials in their royal master's favour. When
Glanvill wrote the sums to be taken by mesne lords had
been fixed; but the Crown remained free to exact higher
rates. Baroniae capitales were charged relief at sums which
varied juxta voluntatem et misericordiam dommi regis.^
Every year, however, made for definition ; custom pointed
towards iocs, for a knight's fee, and ;^ioo for a barony.
Two entries on the Pipe Roll of 10 Richard I. amusingly
illustrate the unsettled practice: .;^ioo is described as a
" reasonable relief " for a barony, and yet a second entry
records an additional payment by way of " fine " to induce
the King to accept the sum his own roll had just declared
"manors" paid ^8 of relief to the King ; the thegn with six or fewer paid three
marks to the sheriff. See Domesday Book, I. 280, b (Derby and Notts). Contrast
Ibid., I. 56, where, however, relief seems to be confused with heriot.
^ Glanvill's words (IX. c. 4) are ambiguous. He distinguishes three cases :
(a) the normal knight's fee, from which loos. was due as relief (whether this
extends to fees of Crown-tenants does not appear) ; [b) socage lands, from which
one year's rent might be taken ; and {c) *^ capitales baroniae " were left subject to
the King's discretion. Now "barony" was a loose word: baronies, like barons,
might be small or great (cf. infra, c. 14) ; all Crown fiefs being *' baronies " in one
sense, but only certain larger "honours" being so reckoned in another. Glanvill
leaves this vital point undetermined, but Dialogus de Scaccario (II. x. E. p. 135
and II. xxiv. p. 155) supports the distinction between Crown-tenants and tenants
of mesne lords: only the latter had their reliefs fixed. Madox (I. 315-6) cites
from Pipe Rolls large sums exacted by the Crown : in one case ;^300 was paid for
six fees — or ten times what a mesne lord could have exacted. i^Pipe Roll, 24
Henry II.) There is further evidence to the same effect: where a barony had
escheated to the Crown, reliefs of the former under-tenants would in future be
payable directly to the Crown ; but it was the practice of Henry II. (confirmed by
c. 43 of Magna Carta, q.v.) to charge, in such cases, only the lower rates exigible
prior to the escheat. A similar rule applied to under-tenants of baronies in
wardship ; see the case of the knights of the see of Lincoln in the hands of a royal
warden in Pipe Roll, 14 Henry II. cited by Madox, ibid.). It would thus appear
that all holders of Crown fiefs (not merely barones majores) were in Glanvill's day
still liable to arbitrary extortions in name of reliefs. The editors of the Dialogus
(p. 223) are of this opinion. Pollock and Maitland (I. 289) maintain the opposite
— that the limitation to lOOs. was binding on the Crown as well as on mesne
lords.
198 MAGNA CARTA
"reasonable."^ John was more openly regardless of
reason. The Pipe Roll of 1202 shows how an unfortunate
heir failed to get his heritage until he paid 300 marks, with
the promise of an annual " acceptable present " to the
King.2
If John could ask so much, what prevented him asking
more ? He might name a prohibitive price, and so defeat
the hereditability of fiefs altogether. Such arbitrary exac-
tions must end, so the barons were determined in 1215 :
custom must be defined, so as to prevail henceforth against
royal discretion. The first demand of the Articles of the
Barons is, " that heirs of full age shall have their heritage
by the ancient relief to be set forth in the Charter," as
though the final bargain had not yet been made. Here it is,
then, duly set forth and defined as ;^ioo for an "earl's
barony," ;^ioo for " a baron's barony," iocs, for a knight's
fee, and a proportional part of loos. for every fraction of a
knight's fee. This clause produced the desired effect.
These rates were strictly observed by the exchequer of
Henry HI., as we know from the Pipe Rolls of his reign.
Thus, when a certain William Pantoll was charged with
;^ioo for his relief on the mistaken supposition that he had
a " barony," he protested that he held only five knights'
fees, and got off with the payment of £2^,,^ The relief of
a barony was subsequently reduced from ;^ 100 to 100 marks.
The date of this change, if we may rely on Madox,^ lies
between the twenty-first and thirty-fifth years of Edward I.^
Apparently all who paid reliefs to the King were m&cted
in a further payment (calculated at 9 per cent, of the relief)
in name of "Queen's Gold," to the private purse of the
Queen Consort, collected by an official representing her at
the exchequer.^
The charter here says nothing of socage or serjeanty.*^
1 Madox, I. 316. '^I5t(i.,l. 317. ^IhW.,1.3iS. Mbid.,l. :^i.
' ^ Apparently its first appearance is in the Inspeximus of loth October, 1297. '^See
Madox, 318; Pollock and Maitland, I. 289; Bemont, CharteSy p. 47.
® See note by editors of Dialcgus, p. 238 ; Poole, Exchequer, 16, 170. The
barons in 1258 {Sel. Charters ^ 382) protested against this, and the practice was
discontinued.
'Cf. supra^ pp. 54-6.
CHAPTER TWO 199
(a) Socage, The barons were not vitally interested in
socage, that being, in the normal case, the tenure of humbler
men.^ In later reigns the King, like an ordinary mesne
lord, contented himself with one year's rent of socage lands
in name of relief, (h) Serjeaniy, The barons cannot have
been indifferent to the fate of serjeanties, since many of
them held great estates by such tenures. Possibly they
assumed that the rules applied to knights' fees and baronies
would apply to serjeanties as well. The Crown acted
on quite a different view; large sums were frequently
extorted by Henry III. By the reign of Edward I., how-
ever, the exchequer limited itself to one year's rent^ for
petty serjeanties, which thus fell into line with socage.^
II. Units of Assessment. Some explanation is required
of the three groups into which Crown estates were thus
divided — knights' fees, barons' baronies, and earls'
baronies.
(i) Feodum Tuilitis integrum. There is little doubt, in
light of evidence accumulated by Mr. Round in his Feudal
England, that William I. stipulated verbally for the service
of a definite number of knights from every fief bestowed by
him on his Norman followers. A knjghtlsJee (or scutum)
became the measure of feudal assessment : servitium, unius
militis was a well-known legal unit. But difficult problems
emerge when it is asked what equation, if any, existed
between land and service. Unsuccessful attempts have
been made to identify the knights' fee with a fixed area of
five hides on the one hand, or with a fixed annual value of
;^20 upon the other. Prof. Vinogradoff^ has shown con-
clusively that no fixed ratio exists. Fees have been found
as small as one hide and as large as 48 ; and they vary in
^ It is possible to argue that the custom as to socage was already too well settled
to require confirmation : Glanvill (IX. c. 4) stated the relief for socage at one
year's value. It is not clear, however, whether this restriction applied to the
Crown. Further, no custom, however well established, was safe against John's
greed.
2 See Littleton, Tenures^ II. viii. s. 154, and Madox, I. 321, who cites the case
of a certain Henry, son of William le Moigne, who was fined in ;^i8 for the relief
of lands worth ;!^i8 a year held " by the serjeanty of the King's Lardinary."
^Cf. supra^ p. 57. See Round, King's Serjeanties^ p. 33.
* Engl. Soc. in Eleventh Centtiry, pp. 42-48.
1/
200 MAGNA CARTA
value from place to place, as well as from reign to reign.
William I. allowed himself a wide discretion in saddling
estates with service : favoured foundations like Gloucester
and Battle Abbey enjoyed complete exemption. Yet he did
not distribute burdens in pure wantonness ; and the majority
of holdings approximated to a normal standard of extent and
value. Under Henry II. two types appear, the larger of
1 6 marks and the smaller of lo. Under Edward I. a general
appreciation of values seems to have raised the former
standard to ;^20.^
The Crown tenant's holding consisted of a fixed number
of knights' fees — usually a multiple of five (a troop of ten
mounted soldiers forming the military unit of the Norman
Kings) ; and each fee, whatever might be its acreage or
rental, owed the service of one knight. Each fee, under the
Great Charter, paid relief at loos., unless the estate, of
which it formed part, was reckoned as a barony.
(2) Baronia integra. The word " barony " has under-
gone many changes.^ A " barony " at the Norman Con-
quest differed in almost every respect from a " barony " at
the present day. The word haro was originally synony-
mous with homo, meaning, in feudal usage, a vassal of any
lord. It soon became usual, however, to confine the word
to king's men ; " harones " were identical with " crown
tenants " — a considerable body at first ; but a new distinc-
tion arose (possibly as a consequence of the procedure for
summoning them to a Great Council as stipulated for in
chapter 14 of Magna Carta) between the great men and the
smaller men {harones major es and minor es). The latter
were called knights (milites), while " baron " was reserved
for the greater tenants.^ For determining what constituted
a "barony," however, it was impossible to lay down any
'^ Engl. Soc. in Eleventh Century, pp. 49-50.
2 See Pollock and Maitland, I. 262, and authorities there cited. "An honour
or barony is thus regarded as a mass of lands which from of old have been held by
a single title." See also Pike, House of Lords, pp. 88-9.
3 This change was not complete in 12 15, but Magna Carta, when it uses
'■'' harones"" alone, seems to refer to ^'- barones majores" (see cc. 2, 21, 61). Cf.
infra under c. 14.
CHAPTER TWO 201
absolute criterion. Mere size was not sufficient. Under
Henry II. baronies still paid relief at the King's good
pleasure.^ Richard and John were more rapacious than
their father. John, indeed, forced William de Braose, who
was heir to the barony of Limerick, to promise a relief
of 5000 marks — a sum he was quite unable to pay.^ Magna
Carta, here not merely declaratory, but making an addition
to existing custom, fixed ;^ 100 as the relief for a full barony
(a sum afterwards reduced to 100 marks) irrespective of size
or value. ^
(3) Baronia comitis Integra. Where a modern eye
expects to find " earldom," the text reads " earl's barony." ^
But " earldom " originally meant an office, the chief magis-
tracy of a county, not a title of dignity nor the ownership of
land : whereas " relief " was due for the land, not the office.
Therein lies also the explanation why the earl originally
paid no more for his barony than the baron paid for his.
The position of an earl under the Norman Kings had been
something far different from a modern " earldom " : it did
not pass, as matter of course, from father to son without
the King's confirmation ; it did not carry with it any right
to demand entry to the King's Council ; it was not one of
several "steps in the peerage," a conception that did not
then exist. ^
The policy of the Conqueror had been to bring each
county as far as possible under his own direct authority;
many districts had no earls, while in others the connection
'^ Dialogus, II. xxiv. ^ New Rymer^ I. loy.
^Madox, I. 216 fF. As the Exchequer, from the time of Edward I., exacted
100 marks from a barony and loos. from a knight's fee, the false equation of extent
•' I barony =13^ knights' fees" was deduced. Coke {On Littleton, IV. s. 1I2,
and Second Inst., 7) is sometimes credited with originating this error, but it
appears in Modus tenendi Parliamentum [Sel. Chart., 503). To suit the pro-
portion given in John's Charter the equation would need to be " i barony = 20
fees." There is, of course, no fixed equation ; baronies fnight be of any size ; we
read of land held *' in baronagio per sei-vitium feodi unitis militis " (Northumber-
land Eyre Roll, 7 Ed. I. ; Surtees Soc, 88, p. 327).
^In the Inspeximus of Edward I., however, comitatus (earldom) displaces the
baronia comitis of the text. See Statutes of Realm, I. 114.
^See Pike, House of Lords, 57.
202 MAGNA CARTA
of an earl with his titular shire was reduced to a shadow,
the only points of connection being the right to enjoy " the
third penny" (that is, the third part pro indiviso of the
profits of the county court) and the right to bear its name.
It is true that, in addition, the earl usually held valuable
estates in the shire, but he did this only as any other land-
owner might. For purposes of taxation the whole of his
lands were reckoned as one unit, here described as haronia
comitis Integra, the relief on which was taxed at ;£ioo.
Very gradually, in after ages, the conception of an earldom
suffered change. The official character made way for the
idea of tenure, and later on for the modern conception of a
hereditary dignity conferring rank and privileges. The
period of transition, when the tenurial idea prevailed, is
illustrated by the successful attempt of Ranulf, earl of
Chester and Lincoln, in the reign of Henry III. to dispose
of one of his two earldoms — described by him as the comi-
tatus of Lincoln.^ Earls are now, like barons, created by
letters patent, and need not be land-owners. Thus the
words " barony " and " earldom," so diverse in their origin
and early development, became closely united in their later
history.
III. Liability of Church Property to "Relief." The
Charter of John, unlike that of Henry I., makes no mention
of the lands of vacant sees in this connection, probably
because the main question had long been settled in favour
of the church. The position of a bishopric was, however,
a peculiar one : each prelate was a Crown tenant, and his
fief was reckoned a "barony," entitling its owner to all the
privileges, and saddling him with all the feudal obligations
of a baron. ^
It was not unnatural that, when a prelate died, the Crown
should -demand "relief" from his successor. Thus, in
^ See Pike, House of Lords, S^. This term comitatus was a word of many
meanings. Originally designating the "county "or " the county court," it came
to mean also the office of the earl who ruled the county, and later on it might
indicate either his titular connection with the shire, his estates, his share of the
profits of justice, or his rank in the peerage.
2 This was affirmed in 1164 by Article ii of the Constitutions of Clarendon,
which stipulated that each prelate should hold his lands sicut baroniam.
CHAPTER TWO 203
1092, Herbert Losinga paid ;^iooo of relief for the see of
Thetford, an act of simony for which his conscience pricked
him. Such demands met with bitter opposition. The
Crown, unwilling to forego its feudal dues, endeavoured to
shift their incidence from the revenues of the see to the
shoulders of the feudal under-tenants. After bishop Wulf-
stan's death on i8th January, 1095, a writ was issued in
William's name to the freeholders of the see of Worcester,
calling on each of them to pay, as a relief due on their
bishop's death, a specified sum, assessed by the barons of
exchequer.^
In revenge for such extortions, the historians of the day,
recruited from the clerical class, have heartily commended
Rufus and Flambard to the opprobrium of posterity.
Henry I., in his coronation Charter, promised to exact
nothing during vacancies from the demesne of the church
or from its tenants.^ No corresponding promise was
demanded from John, a proof that such exactions had
ceased. The Crown no longer extorted relief from church
lands, although wardship was, without protest, enforced
during vacancies.
CHAPTER THREE.
Si autem heres alicujus talium fuerit infra etatem et fuerit
in custodia, cum ad etatem pervenerit, habeat hereditatem
suam sine relevio et sine fine.
If, however, the heir of any one of the aforesaid has been
under age and in wardship, let him have his inheritance without
relief and without fine when he comes of age.
The Crown is here forbidden to exact relief where it had
already enjoyed wardship. It was hard on the youth,
escaping from leading-strings, to be met, when he " sued
^ Siait per barones meos disposui. The writ is given in Heming's Carhilaty, I.
79-80, and reprinted by Round, Feudal England^ 309. *
^ See supra, p. 98.
204 MAGNA CARTA
out his livery," with the demand for a large relief by the
exchequer which had appropriated all his revenues.^
Such double extortion had long been forbidden to mesne
lords ; Magna Carta was extending similar limitations to the
King. The grievance complained of had been intensified
by an unfair expedient which John sometimes adopted. In
cases of disputed succession he favoured the claims of a
minor, enjoyed the wardship, and thereafter repudiated his
title altogether, or confirmed it only in return for an exor-
bitant fine. The only safeguard was to provide that the
King should not enjoy wardship until he had allowed the
heir to perform homage, which pledged the King to
" warrant " the title against all rival claimants. This
expedient was actually adopted in the revised Charter of
1216.2
The alterations in that reissue were not altogether in the
vassal's favour. Another addition made a reasonable
stipulation in favour of the lord, which illustrates the theory
underlying wardship. Only a knight was capable of
bearing arms ; hence, the lord held the lands in ward until
the minor should reach man's estate. Ingenious attempts
had apparently been made to defeat these legitimate rights
of feudal lords by making the infant heir a "knight," thus
cutting away the basis on which wardship rested. The
reissue of 12 16 provided that the lands of a minor should
remain in wardship, although he was made a knight. ^
Incidentally, the same Charter declared twenty-one years to
^ Where there had already been a wardship, the relief was thus the price paid by
the heir in order to escape from the heavy hand of the King, and was therefore
known as ^^ ousterlemam." Taswell-Langmead {Engl. Const. Hist.^ 51 n.) states
the amount at half a year's profits. He cites no authorities, and is probably in
error. Dialogtis, 11. x. E. p. 135, forbids relief to be taken, when wardship had
been exercised per aliquot annos.
2 See chapter 3 of 12 16, which stipulates that no lord shall have wardship
''^ antequain homagium ejtis ceperity Seo-CokQ, Second Instittite, 10. Cf. Adams,
Origin, 204, on "homage as a recognition of title."
^ CoV&^ -Ibid. , p. 12, makes a subtle and unwarranted distinction depend on
whetherithe minor was made a knight before or after his ancestor's death. The
proviso, he argues, does not apply to the former case, because lands cannot
** remain " in wardship if they were not in it before.
f
^^^^R THREE 205
be the periou f?.^^v*^?5ii^a'^military tenant came of age, a point
on which John's Charter is silent.
In one case, exceptionally, wardship and relief might both
be exacted on account of the same death, though not by the
same lord. Where the dead man had formerly held two
estates, one of the Crown and one of a mesne lord, the
Crown might claim the wardship of both, and then the
disappointed mesne lord was allowed to exact relief as a
solatium for his loss.^
CHAPTER FOUR.
CusTOS terre hujusmodi heredis qui infra etatem fuerit,
non capiat de terra heredis nisi racionabiles exitus, et
racionabiles consuetudines, et racionabilia servicia, et hoc
sine destructione et vasto hominum vel rerum; et si nos
commiserimus custodiam alicujus talis terre vicecomiti vel
alicui alii qui de exitibus illius nobis respondere debeat,
et ille destructionem de custodia fecerit vel vastum, nos
ab illo capiemus emendam, et terra committatur duobus
legalibus et discretis hominibus de feodo illo, qui de exitibus
respondeant nobis vel ei cui eos assignaverimus; et si
dederimus vel vendiderimus alicui custodiam alicujus talis
terre, et ille destructionem inde fecerit vel vastum, amittat
ipsam custodiam, et tradatur duobus legalibus et discretis
hominibus de feodo illo qui similiter nobis respondeant
sicut predictum est.
The guardian of the land of an heir who is thus under age,
shall take from the land of the heir nothing but reasonable
produce, reasonable customs, and reasonable services, and that
without destruction or waste of men or goods ; and if we have
committed the wardship of the lands of any such minor to the
sheriff, or to any other who is responsible to us for its issues,
and he has made destruction or waste of what he holds in
wardship, we will take of him amends, and the land shall be
committed to two lawful and discreet men of that fee, who shall
be responsible for the issues to us or to him to whom we shall
^See Coke on Littleton, Book II. c. iv. s. 112 ; and cf. infra, cc. 37 and 43 for
the " prerogative wardship " of the Crown.
v
206 MAGNA (\rtA
assign them ; and if we have given or soJ^.-.r^ ,-araship of any
such land to anyone and he has therein made destruction or
waste, he shall lose that wardship, and it shall be transferred to
two lawful and discreet men of that fief, who shall be responsible
to us in hke manner as aforesaid.
This chapter and the next treat of wardship,^ a much-
hated feudal incident, which afforded opening for grave
abuses. It is a mistake, however, to regard its mere exist-
ence as an abuse : it seems to have been perfectly legal in
England from the date of the Norman Conquest, although
some writers ^ consider it an innovation devised by William
Rufus and Flambard. Their chief argument is that Henry
I., in promising redress of several inventions of Rufus,
promised also to reform wardship. This shows that ward-
ship was abused, but does not prove it an innovation.
The Charter of Henry committed him to drastic remedies,
which would have altered the character of wardship
altogether. Clause 4 of that document removed from the
lord's custody both the land and the person of the heir, and
gave them to the widow of the deceased tenant (or to one of
the kinsmen, if such kinsman had, by ancient custom,
rights prior to those of the widow) .^ This was one of the
\ many promises which the " lion of justice " never kept.
Wardship continued to be exercised as before, over lay fiefs,
throughout the reigns of Henry I. and Stephen. Article 4
of the Assize of Northampton (1176) merely confirmed the
existing practice when it allowed wardship to the lord of
the fee.* The barons in 12 15 made no attempt to revert to
the drastic remedies of the Charter of Henry I., although
the evils complained of had become worse under John's
misgovernment.
It must be remembered that " wardship " placed the
property and person of the heir at the mercy of the Crown.
Even if the popular belief as to the fate met by prince
^The nature of wardship is more fully explained supra, pp. 61-2.
"^ E.g. Taswell-Langmead, Eng. Const. Hist., p. 51 n.
^**This, it would seem, was the old English rule;" see Ramsay, Foundations
of England, II. 230.
* It is a common error to suppose that this Assize restores wardship to the lord.
CHAPTER FOUR 207
Arthur at his uncle's hands was unfounded, John was not
the guardian to inspire confidence in the widowed mother
of a Crown tenant whose estates the King might covet.
Further, the King might confer the office, with the dehcate
issues involved, upon whomsoever he would. When such
a trust was abused, it was difficult to obtain redress. In
^^33j ^ guardian, accused de puella quam dicitur violasse in
custodia sua, paid a fine to the Crown, if not as hush money,
at least in order to obtain protection from being sued else-
where than in the Curia Regis^
Guardians were of two kinds. The King might entrust
the lands to the sheriff of the county where they lay (or to
one of his bailiffs), such sheriff drawing the revenues on
the Crown's behalf, and accounting in due season at the
exchequer. Alternatively, the King might make an out-
and-out grant of the office, with all its profits, to a royal
favourite or the highest bidder. Commentators of a later
date 2 apply the word "committee" to the former type of
guardian, reserving " grantee " for the latter. This distinc-
tion, mentioned by Glanvill,^ obtains recognition in this
passage of the Charter. Neither type was likely to have
the interests of the minor at heart. They had always strong
inducements to exhaust the soil, stock, and timber, uproot-
ing and cutting down whatever would fetch a price, and
replacing nothing. The heir too often found impoverished
lands and empty barns.
William Marshal's experience affords apt illustration.
Early in Richard's reign, he married Isabel of Clare, but
John, Dominus Hiherniae, refused seisin of the bride's Irish
lands. When Richard was appealed to, John tried to make
conditions : " provided the grants of lands I have made to
my men hold good and be confirmed," to which the King
aptly replied : " That cannot be : for what would then
remain to him, seeing that you have given all to your
people? "4
The remedies proposed by Magna Carta were too timid
1 See Pipe Roll, 29 Henry II., cited Madox, I. 483.
"^ E.g. Coke, Second Institute, p. 13.
3 VII. c. 10. *Orpen, Ireland, II. 203.
2o8 MAGNA CARTA
and half-hearted; yet something was effected. It was
unnecessary to repeat the recognized rule that the minor
must receive, out of the revenues, maintenance and educa-
tion suited to his station ; but the Crown was restrained by
chapter 3 from exacting relief where wardship had already
been enjoyed; chapter 37 forbade John to exact wardship
in certain cases where it was not legally due; while here in
chapter 4 an attempt was made to protect the estate from
waste.
The promised reforms included a definition of " waste " ;
punishment of the wasteful guardian ; and protection
against repetition of the abuse. Each of these calls for
comment. (1) The definition of waste. The Charter uses
the words " vastum hominum vel rerum " (a phrase which
occurs also in Bracton).^ It is easy to understand waste of
goods ; but what is " waste of men " ? An answer may be
found in the " unknown Charter of Liberties," ^ which binds
guardians to hand over the land to the heir " sine venditione
nemorum et sine redemptione hominum." To enfranchise
villeins was one method of "wasting men." The young
heir, when he came to his estates, must not find his praedial
serfs enxapBfpated.3 In 1259, the Provisions of West-
minster (c. 20) forbade " farmers " to make waste, or sale, or
exile, of woods, or houses, or men. The statute of Marl-
borough placed such defaulters at the King's mercy.
(2) Punishment of wasteful guardians. The Charter
provides appropriate punishment for each of the two types
of guardian. John promises to take "amends," doubtless
of the nature of a fine, frorn the " committee " who had no
personal interest in the property ; while the " grantee " is to
^ II. folio 87. ^ See Appendix.
3 Another way of "wasting" villeins was by tallaging them excessively. (For
meaning of tallage cf. zn/ra, c. 12.) Thus Bracton's iVi?/<f ^^^y^ reveals how one
guardian destruxit villanos per tallagia {v. case 485) ; how another exiled or
destroyed villeins to the value of 300 marks (case 574) ; how a third destroyed two
rich villeins so that they became poor and beggars and exiles (case 632). Cf. also
case 691. Daines Barrington, writing towards the middle of the eighteenth
entury, went too far when he inferred from this passage "that the villeins who
'S6. by servile tenure were considered as so many negroes on a sugar plantation "
{Observations^ P- 7)-
CHAPTER FOUR 209
forfeit the guardianship, thus losing a valuable asset for
which he had probably paid a high price. While the
Statute of Westminster ^ merely repeated the words of
Magna Carta, the Statute of Gloucester ^ enacted that the
grantee who had committed waste should not only lose the
custody, but should, in addition, pay to the heir any balance
between the value of the wardship thus forfeited and the
total damage. More severe penalties were found necessary.
Statute 36 Edward III (c. 13) enacted that King's
escheators, guilty of waste, should " yield to the heir treble
damages." If the boy was still a minor, his friends might
bring a suit on his behalf ; or after he was of full age he
might bring it on his own account.^
(3) Provision against recurrence of the waste. It was
only fair that reasonable precautions should be taken to
prevent the heir who had already suffered hurt, from being
similarly abused a second time. John promised to super-
sede the keeper guilty of waste, by two trustworthy free-
holders on the heir's estate. These men, from their local
and personal ties to the young heir, might be expected to
deal tenderly with his property. The " unknown Charter '*
proposed a more drastic remedy : the lands were to be
entrusted at once to four knights of the fief, without waiting
until damage had been done. Even the milder provision
of Magna Carta was an innovation, and there is no evidence
that it was ever put in force.
CHAPTER FIVE.
CusTOS autem, quamdiu custodiam terre habuerit, sustentet
domos, parcos, vivaria, stagna, molendina, et cetera ad ter-
1 Edward I. c. 21. 2 Edward I. c. 5.
3 Coke, Second Institute^ p. 13, enunciates a doctrine at variance with this
statute, holding that the heir who suffered damage could not, on coming of age,
obtain triple damages, or indeed any damages at all, if the King had previously
taken amends himself. Coke further maintains that even after waste, the person
of the heir was left in the power of the unjust guardian, explaining that when the
Charter took away the office "this is understood of the land, and not of the
body."
O
210 MAGNA CARTA
ram illam pertinencia, de exitibus terre ejusdem ; et reddat
heredi, cum ad plenam etatem pervenerit, terram suam
totam instauratam de carrucis et waynagiis, secundum quod
tempus waynagii exiget et exitus terre racionabiliter
poterunt sustinere.
The guardian, moreover, so long as he has the wardship of
the land, shall keep up the houses, parks, ^ fishponds, stanks,^
mills, and other things pertaining to the land, out of the issues
of the same land ; and he shall restore to the heir, when he has
come to full age, all his land, stocked with ploughs and " way-
nage," ^ according as the season of husbandry shall require, and
the issues of the land can reasonably bear.
These stipulations form the complement, on the positive
side, of the negative provisions of chapter 4. It was not
sufiicient to prohibit acts of waste; the guardian must
keep the estates in good repair.
I. The Obligations of the Warden of a Lay-fief. It was
the duty of every custodian to preserve the lands from
neglect, together with all the usual equipment of a medieval
manor. Outlay thus required formed, in modern lan-
guage, a first charge on the revenues, before the balance was
appropriated by the " grantee, " or paid to the exchequer by
the "committee."
r^ This clause expands and improves the corresponding
Article of the Barons ; but the obligation to restore the land
and its appointments " in as good order as the revenues
would bear " came to be regarded as too stringent, obliging
the guardian to use up surplus revenue in repairing waste
committed in the time of the deceased. Henry's charters
modified this : the guardian need only hand over the land
and appointments in as good condition as he had received
them .4
New methods of abusing wardship were invented after
^ This term is explained, c. 47. infra.
2 It is difficult to distinguish between vivarium and stagnum. By Coke, in the
Statutis at large ^ vivarium is translated "warren"; but that word has its Latin
form in warrena. Westminster II. (c. 4) speaks of stagnum molendince (a mill-
pond). Statute of Merton (c. 11) refers to poachers taken mparcis et vivariis.
' Discussed infra^ under c. 20. * Cf. Blackstone, Great Charter, Ixxviii.
CHAPTER FIVE 211
Magna Carta. The Statute of Marlborough (c. 16) gave
to a ward, kept out of his heritage, an action of mort
d' ancestor against a mesne lord, but not against the Crown. ^
The Statute of Westminster I. (c. 48) narrates that heirs
were often carried off bodily to prevent them raising actions
against guardians. The whole subject was regulated in
1549 by Statute 32 Henry VIII. c. 46, which instituted
the Court of Wards and Liveries, the expensive and dilatory
procedure of which caused increasing discontent, until an
order of both Houses of Parliament, dated 24th February,
1646, abolished it along with " all wardships, liveries,
primer seisins, and ouster les mains." "^ This ordinance
was confirmed at the Restoration by Statute 12 Charles 11.
c. 24.3
II. Wardships over Vacant Sees. The church had its own
grievances. The Constitutions of Clarendon * had stipu-
lated that each prelate should hold his Crown land sicut
haroniam; and this view ultimately prevailed. It followed
that all appropriate feudal burdens affected church fiefs
equally with lay fiefs. The lands of a see were, however,
the property of an undying corporation (to use the language
of a later age) : a minority was impossible, and therefore, so
it might be argued, w^ardships could never arise. Rufus
objected to this reasoning, and devised a substitute for
ordinary wardships by keeping sees long vacant, and
meanwhile appropriating the revenues. Henry I., while
renouncing all pretensions to exact reliefs, retained his
right of wardship, promising merely that vacant sees should
neither be sold nor farmed out. Stephen went further,
renouncing expressly all wardships over church lands ; but
Henry II. ignored this concession, and reverted to the
practice of his grandfather. In his reign the wardship of
the rich properties of vacant sees formed a valuable asset of
the exchequer. During a vacancy the Crown drew not
only the rents and issues of the soil, but also the various
feudal payments which the under-tenants would otherwise
1 Cf. mfra, under c. i8. ^See R. S. Gardiner, Documents^ p. 207.
' See infra, under c. 37, for prerogative wardship.
* Article 11 : see Select Charters, 139. -
212 MAGNA CARTA
have paid to the bishop. The Pipe Roll of 14 Henry II. ^
records " reliefs " of £so and ;^20 paid by tenants of the
vacant see of Lincoln for six and four knights' fees
respectively .2
John reserved his wardships in his charter to the church ;
and Stephen Langton thought, perhaps, it was unnecessary
to press for their renunciation, since the promise not to delay
elections would render such wardship unprofitable.^
The omission was supplied in 12 16, when the provisions
applicable to lay fiefs were extended to vacant sees, with the
added proviso that church wardships should never be sold.
These provisions were supplemented by later acts. An
Act of 14 Edward III. (stat. 4, cc. 4 and 5) gave to the dean
and chapter of a vacant see a right to pre-emption of the
wardship at a fair price. If they failed to exercise this, the
King's right to appoint escheators or other keepers was
confirmed, but under strict rules as to waste.
CHAPTER SIX.
Heredes maritentur absque disparagacione, ita tamen
quod, antequam contrahatur matrimonium, ostendatur
propinquis de consanguinitate ipsius heredis.
Heirs shall be married without disparagement, yet so that
before the marriage takes place the nearest in blood to that heir
shall have notice.
The Crown's right to regulate the marriages of wards
had become an intolerable grievance. The origin of this
feudal incident and its extension to male as well as female
minors have been elsewhere explained."* John made a
regular traffic in the sale of wards — maids of fourteen and
widows alike. The Pipe Roll of John's first year ^ records
how the chattels of Alice Bertram were sold because she
refused " to come to marry herself " at the King's summons.
^ Cited by editors of the Dialogus^ p. 223. ^ Cf. under c. 43 infra.
, ' C. 46 of Magna Carta (see infra) confirmed barons^ who had founded abbeys,
in their rights of wardship over them during vacancies.
* See supra, 26-3. * Cited Madox, I. 565.
CHAPTER SIX 213
Only two expedients were open to those who objected to
mate with the men to whom John sold them. They might
take the veil, become dead in law, and forfeit their fiefs to
escape the burdens inherent in them ; or they might outbid
objectionable suitors. Brief entries in John's Exchequer
Rolls condense many a tragedy. In his first year, the widow
of Ralph of Cornhill offered 200 marks, with three palfreys
and two hawks, that she might not be espoused by Godfrey
of Louvain, but remain free to marry whom she chose, and
yet keep her lands. This was a case of desperate urgency,
since Godfrey, for love of the lady or of her lands, had
offered 400 marks, if she could show no reason to the con-
trary. It is satisfactory to learn that the lady escaped.^
Sometimes John varied his practice by selling, not the
woman herself, but the right to sell her. In 1203 Bar-
tholomew de Muleton bought for 400 marks the wardship
of the lands and heir of a certain Lambert, along with the
widow, to be married to whom he would, yet so that she
should not be disparaged.^
Great stress was placed on " disparagement " — that is,
forced marriage with one not an- equal. William of Scot-
land, by the treaty of 7th February, 121 2, conferred on
John the right to marry prince Alexander to whom he would,
"but always without disparagement."^ Such proviso was
understood where not expressed. It is not surprising, then,
to find it confirmed in Magna Carta. The Articles of the
Barons had, indeed, demanded that a royal ward should
only be married with consent of the next of kin. In our
text, this is softened down to the mere intimation of an
intended marriage : the opportunity was still afforded of
protesting against an unsuitable match. Insufficient as the
provision was, it was omitted from the reissues of Henry's
reign. The sale of heiresses went on unchecked.
Magna Carta made no attempt to define disparagement,,
but the Statute of Merton * gave two examples, — marriage
1 See /^o^u/i de Oblatis et Finibus, p. 37, and Pipe Roll, 2 John, cited by Madox,
I. 515.
Pipe Poll, 4 John, cited by Madox, I. 324. » See infra, c. 59.
* 20 Henry III. c. 6.
214 ^M^ MAGNA CARTA
to a villein or a burgess. This was not an exhaustive list:
Littleton ^ adds other illustrations : — " as if the heir that is
in ward be married to one who hath but one foot, or but one
hand, or who is deformed, decrepit, or having an horrible
disease, or else great and continual infirmity, and, if he be
an heir male, married to a woman past the age of child-
bearing." Plenty of roonPwas left tor torcmg on a~wafd"
an"oBjectionable spouse, who yet did not come within the
law's definition of " disparagement." The barons argued
in 1258 that an English heiress was disparaged if married
to anyone not English born.^
Was it in the power of the far-seeing father of a prospec-
tive heiress, by bestowing her in marriage during his own
life-time, to render nugatory the Crown's right to nominate
a husband ? Not entirely : the Charter of Henry I. reserved
the King's right to be consulted by the barons before they
bestowed the hand of female relations in marriage. Magna
Carta is silent on the point. Bracton ^ thus explains the
law : — No woman with an inheritance could marry without
the chief lord's consent, under pain of losing such inherit-
ance ; yet the lord when asked was bound to grant consent,
if he failed to show good reason to the contrary. He could
not, however, be compelled to accept homage from an
enemy or other unsuitable tenant. The Crown's rights in
such matters were apparently the same as those of a mesne
lord.^
^ Tenures, II. iv. s. 109.
2 See Petition of the Barons {Sel. Charters, 383). Gradually the conception of
disparagement expanded, partly from the natural development of legal principles
and partly from the increased power of the nobility. Coke commenting on
Littleton (Section 107) mentions four kinds of disparagements : (i) propter vitium
animi, e.g. lunatics ; (2) propter vitium sanguinis, villeins, burgesses, sons of
attainted persons, bastards, aliens ; (3) propter vitium corporis, as those who had
lost a limb or were diseased or impotent ; and (4) propter jacturam privilegii, or
such a marriage as would involve loss of *' benefit of clergy." The last clause had
no connection with the law as it stood in 12 15. Marriage with a widow or
widower was deemed by the Church in later days an act of bigamy, and involved
loss of benefit of clergy, until this was remedied by the Statute i Edward VI.
c. 12 (sect. 16).
3 II. folio 88.
* For further information on the age at which marriage could be tendered to a
ward, and the penalties for refusing, see Thomson Magna Carta, pp. 170- 171.
CHAPTER SEVEN 215
CHAPTER SEVEN.
Vidua post mortem mariti sui statim et sine difficultate
habeat maritagium et hereditatem suam, nee aliquid det pro
dote sua, vel pro maritagio suo, vel hereditate sua quam
hereditatem maritus suus et ipsa tenuerint die obitus ipsius
mariti, et maneat in domo mariti sui per quadraginta dies
post mortem ipsius, infra quos assignetur ei dos sua.
A widow, after the death of her husband, shall forthwith and
without difficulty have her marriage portion and inheritance;
nor shall she give anything for her dower, or for her marriage
portion, or for the inheritance which her husband and she held
on the day of the death of that husband ; and she may remain
in the house of her husband for forty days after his death, within
which time her dower shall be assigned to her.
No forethought of a Crown tenant, setting his house in
order, could rescue his widow from the unfortunate position
into which his death would plunge her. He must leave her
without adequate protection against the tyranny of the
King, who might inflict terrible hardships by harsh use of
rights vested in him for the safeguard of his feudal inci-
dents. She might, if deprived of her "estovers," find
herself in actual destitution, until she had made her bargain
with the Crown. She had a right, indeed, to one-third of
the lands of her husband (her dos rationalis) in addition to
any lands she might have brought as a marriage portion ;
but she could only enter into possession by permission of
the King, who had prior claims and could seize everything
by his prerogative of primer seisin.^ This chapter provides
a remedy. Vi^idows shall have their rights without delay,
without difficulty, and without payment.
I. The Widow's Share of Real Estate, Three words are
used : — dos, maritagium, and hereditas.
(i) Dower. A wife's dower is here the portion of her
husband's lands set aside to support her in her widowhood.
It was customary from an early date for a bridegroom to
^Cf. supra, 6$- S-
2i6 MAGNA CARTA
make provision for his bride on the day he married her.
The ceremony formed a picturesque feature of the marriage
rejoicings, taking place hterally at the church door, as man
and wife returned from the altar. The share thus set apart
for the young wife was known as her dos (or dowry), and
would support her if her husband died. In theory, the
transaction between the spouses partook of the nature of a
contract. The wife's role, however, was a passive one : her
concurrence was assumed. Yet, if no provision was made
at all, the law stepped in, on the presumption that the
omission had been unintentional, and fixed the dower at
one-third of all his lands. ^
John's Magna Carta contents itself with the brief enact-
ment "that a widow shall have her dower." The Charter
of 12 1 7 goes farther, containing an exact statement of the
law as it then stood : — ^" The widow shall have assigned to
her for her dower the third part of all her husband's land
which he had in his lifetime unless a smaller share had been
given her at the door of the church." Lawyers of a later
age have, by a strained construction of the words in vita
sua, made them an absolute protection to a wife against all
attempts to lessen her dower by alienations granted without
her consent during the marriage.^ Magna Carta contains
no warrant for such a proposition, although a later clause
(chapter ii) secures dower lands from attachment by the
husband's creditors, Jews or others.
(2) Maritagium, It was customary for a land-owner to
bestow marriage portions on his daughters. Land so
granted was usually relieved from burdens of service and
homage. It was hence known as " frank-marriage "^^
(liherum maritagium), which almost came to be recognized
as a separate form of feudal tenure. Such grants could be
made without the consent of the tenant's expectant heirs.
Maritagium was thus " a provision for a daughter — or per-
1 See Pollock and Maitland, II. 422-3. The ceremony at the church door,
when resorted to, was no longer an opportunity of giving material proof of affec-
tion to a bride, but a means of cheating her out of what the law considered her
legitimate provision, by substituting something of less value.
'Pollock and Maitland, II. 419.
CHAPTER SEVEN 217
haps some other near kinswoman — and her issue." ^ The
husband was, during the marriage, treated as virtual owner ;
but, on his death, the widow had an indisputable title.
The obvious meaning, however, has not always been
appreciated. Coke ^ reads the clause as allowing to widows
of under-tenants a right denied (by chapter 8) to widows of
Crown tenants — namely " freedom to marry where they will
without any licence or assent of their lords." This inter-
pretation is inherently improbable, since the barons at
Runnymede desired to place restrictions on the King, not
upon themselves ; and it is opposed to the law as expounded
by Bracton.2
Daines Harrington ^ invents an imaginary rule of law in
order to explain a supposed exception. An ordinary widow,
he declares, could not marry again within a year of her
husband's death, but widows of landowners were privileged
to cut short this period of mourning. " Maritagium " is
thus interpreted as a landowning widow's right of speedily
entering on second nuptials. This is a complete inversion of
the truth ; the possession of land really restricted freedom of
marriage. Yet several later authorities follow Barrington's
mistake.^ This is the more inexcusable in view of the clear
explanation given a century ago by John Reeves,^ who
distinguished between two kinds of marriage portion :
liherum maritagium, whence no service whatever was
exigible for three generations, and m^aritagium servitio
ohnoxium, liable to the usual services from the first,
^See Pollock and Maitland, II. 15, 16. Liberum maritagium, considered as a
tenure, has various peculiarities. The lady's husband became the feudal tenant of
her father. The issue of the marriage were heirs to the lands and would hold
them as tenants of the heir of the donor. For three generations, however, neither
service nor homage was due. After the third transmission, the land ceased to be
"free"; the peculiar tenure came to an end ; the new owner was subject to all
the usual burdens.
^ Second Institute, p. 16. ^See supra^ p. 214. * Observations^ pp. 8-10.
^ E.g. Thomson, Afagna Carta, p. 172. Dr. Stubbs has his own reading of
maritagium, namely, "the right of bestowing in marriage a feudal dependant."
See Glossary to Sel. Charters, p. 545. The word may sometimes bear this mean-
ing, but not in Magna Carta.
' See his History of English Law, I. 121 (3rd ed.).
2i8 MAGNA CARTA
although exempt from homage until after the death of the
third heir.^
(3) Hereditas. Is the third item here mentioned simply
another name for either dos or maritagium ? Or, is it some-
thing different? It is possible that " the inheritance which
her husband and she held on the day of the death of that
husband " denotes lands that had come to the lady as heiress
on the decease of relations, not as a gift at her marriage.
Such lands might be described as held by both spouses;
for a husband might even attend Parliament as a baron on
the strength of his wife's barony.
II. The Widow's Share of Personal Estate. The present
chapter says nothing of the widow's " peculiar " or share of
her deceased husband's money and chattels ; but chapter 26
secured to her the portion of one third allowed her by the
existing law.
III. Provision for the Widow's immediate Needs. Intri-
cate questions might arise before the land was divided into
aliquot portions. Meanwhile, temporary provision must be
made for her support. This was of two kinds : (i) Quaran-
tine. Magna Carta confirmed her right to the family home
for forty days, known to later lawyers as the widow's
quarantine. The charter of 1216 notes an exception, on
which John's Charter is silent : if the husband's place of
residence had been a castle, the widow could not stay there ;
feudal strongholds were not for women. In such cases
another residence must be substituted. In later days,
widows were provided with a writ, " de quarentina
hahenda," directing the sheriff to do her right.^
(2) Estovers of Common. The widow required more
than the protection of a roof ; until her dower lands had been
assigned to her, no portion of the produce of her husband's
manors could be strictly called her own. The estate was
held " in common " between her and her husband's heir.
It was only fair that, until her rights were ascertained, she
^ Cf. Idiif., I. 242, where Reeves rightly points out that Coke is mistaken,
although he fails to notice the distinction drawn, in the passage criticized, between
the Crown and mesne lords.
2 See Coke, Second histitnte^ p. i6.
CHAPTER SEVEN 219
should be allowed a reasonable share of the produce.
Neither John's Charter nor the first issue of Henry III. said
anything on this head. The reissue of 1217 supplied the
omission, expressly confirming her right to rationahile
estoverium suum interim de communi. Many explanations
of the word estovers might be cited : from Dr. Johnson, who
defines it broadly as " necessaries allowed by law," to Dr.
Stubbs, who narrows it to " firewood." ^ It was the right to
supply one's personal or domestic wants : this varied in
extent from full supply of all things necessary for the main-
tenance of life, down to a right to take one kind of produce
for one specific purpose only.-
In this passage the word bears its wider signification.
Such was Coke's view,^ who held that it implied the
widow's right to " sustenance " of every kind, including the
right to kill such oxen on the manor as she required for
food. Estovers " of common " should thus be read as
extending the widow's right of consumption for her own
and her household's use over every form of produce held
" in common " by her and the heir's guardian prior to a final
division.^ She could not, however, cut down trees.
^ See Glossary to Sekc^ Charters^ p. 539: "firewood; originally provision or
stuff generally."
2 Several instances of the wider use may be given. Bracton (III. folio 137)
explains that, pending the trial of a man accused of felony, his lands and chattels
were set aside by the sheriff; meanwhile the imprisoned man and his family
received ** reasonable estovers." (Cf. infra^ c. 32.) The Statute of Gloucestier
f6 Edward I. c. 4) mentions incidentally one method of stipulating for a return
from property alienated, viz., estovers of meat or clothes. Blackstone, again
{Co/nvientaries, I. 441), applies the name estovers to the alimony made to a
divorced woman "for her support out of the husband's estate." Sometimes the
word was more restricted. Coke [Second histitute, p. 17) says, "when estovers
are restrained to woods, it signifieth housebote, hedgebote, and ploughbote," —
that is, timber for repairing houses, hedges, and ploughs. Apparently it had an
even more restricted scope when used to describe the right of those who dwelt in
the King's forest, viz., to take dead timber as firewood. (Cf. infra^ c. 44.)
"^ Second Institute^ p. 17.
* There seems no reason to restrict her estovers to a right over "commons," in
the sense of pastures and woods held " in common " by her late husband and the
villeins of his manor. Some such meaning, indeed, attaches to the phrase "dower
of estovers" met with in later reigns, e.g. in Year Book of 2 Edward II. (Selden
Society), p. 58, where it was held that such a right (claimed as a permanent part
of dower) did not belong to a widow.
220 MAGNA CARTA
CHAPTER EIGHT.
Nulla vidua distringatur ad se maritandum dum voluerit
vivere sine marito ; ita tamen quod securitatem faciat quod
se non maritabit sine assensu nostro, si de nobis tenuerit,
vel sine assensu domini sui de quo tenuerit, si de alio
tenuerit.
No widow shall be compelled to marry, so long as she prefers
to live without a husband; provided always that she gives security
not to marry without our consent, if she holds of us, or without
the consent of the lord of whom she holds, if she holds of
another.
Wealthy widows were glad to escape from John's clutches
by agreeing to buy up the Crown's rights for a lump sum.
In the year of Magna Carta, Margaret, widow of Robert fitz
Roger, paid ;^iooo;^ and a few years earlier Petronilla,
countess of Leicester, had given 4000 marks.^ The Pipe
Rolls mention numerous smaller sums; in 1200, Juliana,
widow of John of Kilpec, accounts for 50 marks and a
palfrey.^ Horses, dogs, and falcons were frequently given
in addition to money fines, and testify eloquently to the
greed of the King, the anxiety of the victims, and the
extortionate nature of the system. In return, formal
charters were obtained, a good example of which is that
granted to Alice, countess of Warwick, dated 13th January,
1205,^ containing concessions that she should not be forced
to marry ; that she should be sole guardian of her sons ;
that she should have one-third part of her late husband's
lands as her reasonable dower ; and that she should be quit
from attendance at courts of shire and hundred, and from
payment of sheriff's aids during her widowhood. Another
charter, of 20th April, 1206, shows what a widow might
expect if she failed to make her bargain with the Crown :
John granted to Richard Fleming, an alien as his name
^ See Pipe Roll of i6 John, cited Madox, I, 491.
»See Pipe Roll of 6 John, cited Madox, L 488.
» See Pipe Roll of 6 John, cited Madox, I. 488.
^ New Rymer, I. 91.
CHAPTER EIGHT 221
implies, the wardship of the lands of the deceased Richard
Grenvill, with the rights of marriage of the widow and
children.^
Magna Carta, in substituting a rule of law for the pro-
visions of these private charters, repeated at greater length
the promises made (and never kept) by Henry I. in his
coronation charter : no widow was to be constrained to
marry against her will. This liberty must not be used,
however, to the prejudice of the Crown : the widow could
not marry without the King's consent. Magna Carta pro-
vided that she must find security to this effect, an annoying^
but not unfair stipulation. The Crown, in later days, com-
pelled the widow, when having her dower assigned to her
in Chancery, to swear not to marry without licence under
penalty of a fine of one year's value of her dower.^
CHAPTER NINE.
Nec nos nee ballivi nostri seisiemus terram aliquam nee
redditum pro debito aliquo, quamdiu catalla debitoris suffi-
ciunt ad debitum reddendum ; nec plegii ipsius debitoris
distringantur quamdiu ipse capitalis debitor sufficit ad
solucionem debiti; et si capitalis debitor defecerit in solu-
cione debiti, non habens unde solvat, plegii respondeant
de debito; et, si voluerint, habeant terras et redditus
debitoris, donee sit eis satisfactum de debito quod ante pro
eo solverint, nisi capitalis debitor monstraverit se esse
quietum inde versus eosdem plegios.
Neither we nor our bailiffs shall seize any land or rent for any
debt, so long as the chattels of the debtor are sufficient to repay
the debt; nor shall the sureties of the debtor be distrained
so long as the principal debtor is able to satisfy the debt ; and
if the principal debtor shall fail to pay the debt, having nothing
wherewith to pay it, then the sureties shall answer for the debt ;
and let them have the lands and rents of the debtor, if they
desire them, until they are indemnified for the debt which they
have paid for him, unless the principal debtor can show proof
that he is discharged thereof as against the said sureties.
^See Ne-iV Rymer, I. 92. 2See(;;oke, Second Institute, 18.
222 MAGNA CARTA
The Charter passes to another group of grievances.
Chapters 9 to ii treat of debts, usury, and the Jews, and
should be read in connection with chapter 26, which regu-
lates procedure for attaching personal estate of deceased
Crown tenants who were also Crown debtors. The present
chapter, although general in its terms, had special refer-
ence to cases where the Crown was creditor; while the
two following chapters treat more particularly of debts
contracted to money lenders.
The fact that John's subjects were indebted to his Exche-
quer did not imply that they had borrowed from the King.
What with feudal incidents and scutages, and indis-
criminate fines, a large proportion of Englishmen must
have been permanently indebted to the Crown. At
John's accession many northern barons still owed scutages
imposed by Richard. John remitted none of the arrears,
while imposing new burdens of his own : the attempts
made to collect these debts intensified the friction between
John and his barons.^
Three rules were laid down, (i) The personal estate of
a debtor must be exhausted before his real estate or its
revenues were attacked. To take away his land might
deprive him of his means of livelihood; for chattels could
not yield a permanent revenue. ^ This rule has not found
a place in modern systems of law, which usually leave the
option with the creditor. (2) The estate of the chief debtor
had to be exhausted before proceedings could be instituted
against his sureties. Magna Carta thus enunciated for
English law a rule that has found favour in most systems
of jurisprudence. The man who is only a surety for
another's debt is entitled to immunity, until the creditor has
taken all reasonable steps against the principal debtor.
Such a right is known to the civil law as beneficium ordinis,
and to Scots law as the "benefit of discussion." (3) If
^ See supra, pp. 73-6.
2 The Dialogus de Scaccario, II. xiv., half a century earlier, laid down rules
even more favourable to the debtor in two respects: (i) the order in which
moveables should be sold was prescribed ; and (2) certain chattels were absolutely-
reserved to the debtor, e.g. food prepared for use ; and, in the case of a knight,
his horse with its equipment.
CHAPTER NINE 223
these sureties had, after all, to pay the debt in whole or
part, they were allowed " a right of relief " against the
principal debtor, being put in possession of his lands and
rents. This rule has some analogy with the equitable
principle of modern law, which gives to the surety who
has paid his principal's debt, the right to whatever the
creditor held in security.
Even when the Crown's bailiffs obeyed Magna Carta,
they might still inflict terrible hardship upon debtors.
Sometimes they seized goods valuable out of all proportion
to the debt ; and an Act of 1266 ^ forbade this practice when
the disproportion was "outrageous." Sometimes they
attempted to extort prompt payment by selecting whatever
chattel was most indispensable : oxen were taken from the
plough and allowed to die of neglect. The practice of the
Exchequer, in the days of Henry II., had been more con-
siderate; oxen were to be spared as far as possible where
other personal effects were available.^ John's charter has
no such humane provision,^ and the abuse continued. The
Act of 1266, already cited, forbade officers to drive away
the owner who came to feed his impounded cattle at his
own expense. The Articuli super cartas'^ went further,
prohibiting seizure of beasts of the plough so long as other
effects might satisfy the debt.^
CHAPTER TEN.
Si quis mutuo ceperit aliquid a Judeis, plus vel minus,
et moriatur antequam illud solvatur, debitum non usuret
I51 Henry III. stat. 4 (among "statutes of uncertain date" in Statutes of
Realnty I. 197).
^See Dialogus de Scaccario, II. xiv.
'Cf., however, the rule as to amercements in c. 20.
*28 Edward I. c. 12. Cf. Statute of Marlborough, 52 Henry III. c. 15.
* Henry's reissues make two small additions explaining certain points of detail :
(l) the words *'(?/ ipse debitor parattis sit inde satis/acere" precede the clause
giving sureties exemption; and (2) the sureties are declared liable to distraint
when the chief debtor can pay, but will not.
224 MAGNA CARTA
quamdiu heres fuerit infra etatem, de quocumque teneat;
et si debitum illud inciderit in manus nostras, nos non
capiemus nisi catallum contentum in carta.
If one who has borrowed from the Jews any sum, great or
small, die before that loan be repaid, the debt shall not bear
interest while the heir is under age, of whomsoever he may
hold ; ^ and if the debt fall into our hands, we will not take
anything except the principal sum ^ contained in the bond.
Usury, denied by law to Christians, was carried on by
Jews under disadvantages and risks. The rates of interest
were proportionately high, ranging in normal cases from
two to four pence per pound per week ; that is, from 43J
to 86| per cent, per annum. ^ During his nonage a ward
had nothing wherewith to discharge either principal or
interest, since he who had the wardship drew the revenue.
At the end of a long minority, an heir would have found
the richest estates swallowed up by a debt which had
increased automatically ten or twenty-fold."*
Magna Carta prevented this injustice to the ward ; but,
in doing so, inflicted some injustice on the money-lenders.
During the minority no interest at all, it was provided,
should accrue to Jew or other usurer ; while, if the debt
passed to the Crown, the King must not use his preroga-
tive to extort more than a private debtor might; he must
confine himself to the principal sum specified in the docu-
ment of debt. The provision that no interest should run
^The words *'</<? quocnvique teneat^'' include Crown-tenants and under-tenants,
and sugi^est that only freeholders were protected by this clause.
- Catallum and lucrum were the technical words for " principal " and *' interest."
See Round, Ancient Charters (Pipe Roll Society, Vol. X.), No. 51, and John's
Charter to the Jews, Rot. Chart., p. 93.
3 See Pollock and Maitland, I. 452, and Round's Ancient Charters, notes to
Charter No. 51.
* The Crown was sometimes called in to enable a debtor, overwhelmed by the
accumulation of interest, to come to a settlement with his creditors. In 1199
Geoffrey de Neville gave a palfrey to the King to have his aid "in making a
moderate fine with those Jews to whom he was indebted." See RotuK de Finibus,
p. 40. Ought we to view John's intervention as an attempt to arrange a reason-
able composition with unreasonable usurers, or was it simply a conspiracy to cheat
Geoffrey's creditors ?
CHAPTER TEN 225
during minorities was confirmed by the Statute of Merton,^
which made it clear, however, that its provisions should
not operate to discharge the principal sum or interest
accrued before the ancestor's death. The Statute of Jewry,
of uncertain date,^ made interest irrecoverable by legal pro-
cess. All previous acts against usury were repealed by
the statute 37 Henry VIII. c. 9, which, however, forbade
the exaction of interest at a higher rate than 10 per cent.,
and this remained the legal rate until reduced to 8 per cent,
by 21 James I. c. 17.
I. The History of the Jews in England, In the policy
of the Crown towards aliens of the Hebrew race, three
periods may be distinguished. From the Norman Con-
quest to the coronation of Richard I., the Jews were fleeced
and tolerated; during the reigns of Richard and John
and the minority of Henry III., they were fleeced and
protected ; and finally they were fleeced and persecuted,
this last stage ending with the ordinance of 1290, which
banished Jews from England. The details of this long
story of hardship and oppression, tempered fitfully by
royal clemency, can be only glanced at here. There were
Jews in England before the Norman Conquest; but the
first great influx came in the reign of Rufus, whose financial
genius recognized in them an instrument for his gain, and
who would the more gladly protect them, as likely to prove
a thorn in the side of his enemy the Church. A new
immigration led to the disarming of Jews in 1 181, a measure
which left them at the mercy of the Christian rabble.
When a disturbance occurred at the coronation of
Richard I., on 3rd September, 1189, a general massacre
took place in London. York and other towns were not
slow to follow this example. The King was moved to
anger, not so much by the sufferings of the Jews, as by
the destruction of their bonds ; for the more the Jews had,
the more could be extorted from them. Richard, returning
from captivity a few years later, in urgent need of money,
determined to prevent a repetition of such interference with
a valuable source of revenue. His motive was selfish, but
^20 Henry III. c. 5. "^Statutes of Realm, I. 221.
226 MAGNA CARTA
that was no reason why the Israelites should not pay for
a measure designed for their own protection. Assembled
at Nottingham, they granted a liberal aid, in return for a
new expedient devised to secure their bonds. This scheme, ^
for the details of which Richard was probably indebted to
the genius of his great justiciar, Archbishop Hubert
Walter, was of a comprehensive and practical character.
In London, York, and other important cities, offices or
bureaus were established under the Crown's protection,
containing treasure chests, called archae, fitted with triple
locks, to be opened in presence of custodians, known as
chirographers, who kept the keys. These were four in
number, two Christians and two Jews, chosen by juries
summoned for that purpose by the sheriff; and they were
obliged to find sureties. Only in their presence could loans
be validly contracted between Jews and Christians; and
it was their duty to see such bargains reduced to writing
in duplicate copies. No contract was binding unless a
written copy or chirograph had been preserved in one or
other of those repositories or arks, which thus served every
purpose of a modern register, and other purposes as well.
If the money-lender suffered violence and was robbed of
his copy of the bond, the debtor was still held to his obliga-
tions by the duplicate which remained. If the Jew and all
his relatives were slain, even then the debtor did not escape,
but was confronted by a new and more powerful creditor,
the King himself, armed with the chirograph. Lists of
transactions were preserved, and all acquittances and
assignments of debts, known from their Hebrew name as
" Starrs," had to be carefully enrolled.^ Stringent rules,
codified by Hubert Walter, were issued to the judges when
starting on their circuit in September, 1194.^
If this cunningly-devised system prevented the Christian
debtor from evading his obligations, it also placed the
Jewish creditor completely at the mercy of the Crown ;
for the exact wealth of every Jew could be accurately
1 Cf. C(rp. dejudaeis (Sel. Ch. 262).
*Cf. J. M. Rigg, Sel. Pleas ofthejexvish Exchequer, p. xix.
^Sel. Charters, 262.
CHAPTER TEN 227
gauged from a scrutiny of the contents of the archae. The
King's officials knew, to a penny, how much it was possible
to wring from the coffers of the Jews, whose bonds, more-
over, could be conveniently attached until they paid the
tallage demanded. The custom of fixing on royal castles
as the places for keeping these arks, probably explains
the origin of the special jurisdiction exercised over Jews
by King's constables (" qui turres nostras custodierunt").'^
In their dungeons, horrible engines were at hand for
enforcing obedience. Such jurisdiction, however, extended
legitimately over trivial debts only.^ Important pleas were
reserved for the officials of the exchequer of the Jews, a
special government department, which controlled and regu-
lated the whole procedure. Evidences of the existence of
this separate exchequer have been traced back to 1198,
although no record has been found of a date prior to 1218.^
John, while despising the Jews, protected their wealth as a
reservoir from which he might draw in time of need. Thus,
by a charter dated loth April, 1201, he took 4000 marks
for confirming their privileges; and he obtained a similar
amount after his rupture with Rome. The charter of 1201
was only a confirmation of rights already enjoyed by
English Jews in virtue of the liberal interpretation put upon
the terms of an earlier charter, granted by Henry I. to a
particular father in Israel and his household, but subse-
quently extended, with the tacit concurrence of the Crown,
to the whole Hebrew race. Under John's charter they
enjoyed valuable and definite privileges, which exempted
them from all jurisdictions except those of the King's
justices and castellans.*
When a repetition of the massacres that had disgraced
his brother's coronation was threatened in 1203, John
promptly ordered the mayor of London to suppress all
^See John's Charter to the Jews of loth April, 1201, in Rotuli Chartarum,
P- 93-
2 See Pollock and Maitland, I. 453 n. ^ Rigg» Op. cit.^ xx.
* See Rot. Chart. , I. 93. Complaints brought by Christians against Jews were
to be judged ^' per pares yudez," a phrase which Harcourt, Steward^ 228, interprets
as equivalent to ** the justices or custodes of the Jews," but see infra under c. 39.
228 MAGNA CARTA
such attempts : his promise of protection, " even though
granted to a dog," must be held inviolate. ^ Protection was
accorded, however, only that they might furnish a richer
booty when the occasion came : suddenly John issued
orders for a wholesale arrest of Jews throughout England.
The most wealthy members of their community were
brought together at Bristol, and, on ist November, 1210,
compelled to give reluctant consent to a tallage of the
enormous sum of 66,000 marks. This amount had been
fixed as the result of an exaggerated estimate of the
contents of the archae, and was more than they could
pay. The methods adopted by John's castellans to extort
arrears are well known, especially the case of the unfor-
tunate Jew of Bristol, from whom seven teeth were extracted,
one each day, until he consented to pay the sum demanded.^
It was doubly hard that the race thus plundered and
tortured by the King should be subjected to harsh treat-
ment by the King's enemies on the ground that they were
pampered proteges of the Crown. Yet such was the case :
on Sunday, 17th May, 12 15, when the insurgents on their
way to Runny mede entered London, they robbed and
murdered Jews, using the stones of their houses to fortify
the city walls.^ It is not to be wondered, then, that the
same insurgents, in forcing on King John the demands
that formed the basis of Magna Carta, included provisions
against usury.
The advisers of the young Henry in 12 16 omitted these
clauses, but not from love of the Jews. They were unwil-
ling to impair so useful a source of revenue, which has been
compared to a sponge which slowly absorbed the wealth of
the nation, to be quickly squeezed dry again by the King.
The Jews were always willing to disgorge a portion of their
gains in return for protection in the rest; but their lot
became hard indeed when Henry III., urged by popular
clamour and the wishes of the Pope, began a course of
active persecution. In 1253, ^ severe ordinance inflicted
vexatious regulations on the Hebrews, almost converting
^ J^oL Pai., I. p. 33, and New Rymer^ I. 89. The date is 29th July, 1203.
2 See Rigg, ibid,, xxiv. ^gee Miss Norgate, yi?>^« Lackland, p. 231.
CHAPTER TEN 229
their quarters in each great city into ghettos, like those of
the Continent of Europe.
This was merely the commencement of oppres'sive
measures, the outcome of the growing hatred with which
Christians regarded Jews — a result partly of the heated
imagination of the rabble, ready to believe unauthenticated
stories of the crucifixion of Christian children, and partly
of the fact that rich Jews, in spite of all persecution, had
possessed themselves of the landed estates of freeholders
and barons and claimed to act as lords of Christian tenants,
enjoying wardships, escheats, and advowsons, as any
Christian might have done. The scope of this enquiry
excludes any detailed account of the stages through which
repressive legislation passed. The Statute of Jewry, how-
ever,^ was of exceptional importance; taking from usurers
the right to recover interest by legal process, and limiting
execution for the principal to one half of the debtor's lands
and chattels. In return, some temporary concessions were
granted. One by one, however, these privileges were again
withdrawn, until the end came in 1290 with the issue of a
decree of perpetual banishment by Edward I., who was
compelled to sacrifice his royal preserve of Jews, in defer-
ence to national prejudice.
II. Legal Position of the Jews. All through these
vicissitudes of fortune, the legal status of the Jews had
remained unchanged in essentials. Their position was
doubly hard ; they were plundered by the Crown and perse-
cuted by the populace. If John saved them from being
robbed by his Christian subjects, it was that they might
be better worth the robbing by a Christian king. Yet, for
this protection, at once fitful and interested, the Jews had
to pay a heavy price; not only were they liable to be
tallaged arbitrarily at the King's will, without limit and
without appeal, but they were hated by rich and poor as
the King's allies. Such feelings would of themselves
account for the unsympathetic treatment accorded to money-
lenders by Magna Carta; two other reasons contributed.
Usury was looked on in the Middle Ages as immoral
^ Statutes of Realm ^ I. 221.
230 MAGNA CARTA
(although illegal only for Christians) ; while excessive
interest was habitually exacted.
The feudal scheme of society had no place for Jews.
They shared the disabilities common to aliens, in a form
unmitigated by the protection extended to other foreigners
by their Sovereigns and by the Church. As exiles in a
foreign land, exposed to attacks of a hostile mob, they were
forced to rely absolutely on the arm of the King. The
Jews became the mere perquisites or chattels of the Crown,
in much the same way as the villeins became the serfs or
chattels of their lords. Rights they might have against
others by royal sufferance, but they had no legal remedy
against their master. In the words of Bracton,^ " the Jew
could have nothing of his own, for whatever he acquired,
he acquired not for himself but for the king." His pro-
perty was his merely by royal courtesy, not under protec-
tion of the law. When he died, his relations had no legal
title to succeed to his mortgages, goods, or money ; the
exchequer, fortified by an intimate knowledge of the extent
of his wealth (for that consisted chiefly in registered bonds),
stepped into possession and could do what it pleased. The
King usually, indeed, in practice contented himself with
one-third of the whole ; but if the relations of the deceased
Jew received less than the balance of two-thirds, they would
be well advised to offer no remonstrance. The Crown did
not admit a legal obligation ; and there was no one either
powerful enough, or interested enough, to compel fulfil-
ment of the tacit understanding that restricted the royal
claims. Whatever the Jew had amassed belonged legally
and potentially not to him but to the Crown. Magna
Carta, in striking at money-lenders, was striking at the
King.
CHAPTER eleven:
Et si quis moriatur, et debitum debeat Judeis, uxor ejus
habeat dotem suam, et nichil reddat de debito illo; et
1 Folio 386b.
CHAPTER ELEVEN 231
si liberi ipsius defuncti qui fuerint infra etatem remanserint,
provideantur eis necessaria secundum tenementum quod
fuerit defuncti, et de residuo solvatur debitum, salvo servicio
dominorum ; simili modo fiat de debitis que debentur aliis
quam Judeis.
And if anyone die indebted to the Jews, his wife shall have
her dower and pay nothing of that debt ; and if any children of
the deceased are left under age, necessaries shall be provided
for them in keeping with the holding of the deceased ; and out
of the residue the debt shall be paid, reserving, however, service
due to feudal lords; in like manner let it be done touching
debts due to others than Jews.
If the preceding chapter deprived Jews of part of their
interest, the present one deprived them of part of the
security on which they had lent the principal. The widow's
dower lands were discharged from her husband's debts,
only two-thirds of the original security thus remaining
under the mortgage. Even this must submit to a prior
claim, namely the right of the debtor's minor children to
such " necessaries " as befitted their station in life. Magna
Carta, at the same time, with characteristic care for feudal
rights, provided that the full service due to lords of fiefs
must not be prejudiced, whoever suffered loss. Finally,
these rudiments of a law of bankruptcy were made appli-
cable to Gentile creditors equally as to Jews. These pro-
visions, with others injuriously affecting the royal revenue,
were omitted in 12 16, not to be restored in future charters :
but they were re-enacted in their essential principle, though
not in detail, by the Statute of Jewry, which limited a
creditor's rights of execution to one moiety of his debtor's
lands and chattels.
CHAPTER TWELVE.
Nullum scutagium vel auxilium ponatur in regno nostro,
nisi per commune consilium regni nostri, nisi ad corpus
nostrum redimendum, et primogenitum filium nostrum
militem faciendum, et ad filiam nostram primogenitam
232 MAGNA CARTA
semel maritandam, et ad hec non fiat nisi racionabile auxi-
lium : simili modo fiat de auxiliis de civitate Londonie.
No scutage nor aid shall be imposed on our kingdom, unless
by common counsel of our kingdom, except for ransoming our
person, for making our eldest son a knight, and for once
marrying our eldest daughter ; and for these there shall not be
levied more than a reasonable aid. In like manner it shall
be done concerning aids from the city of London.
This is a famous clause, greatly valued at the time it was
framed because of its precise terms and narrow scope (which
made evasion difficult), and even more highly valued in
after days for different reasons. It came indeed to be inter-
preted in a broad general sense by enthusiasts who, with
the fully-developed British Constitution before them, found
in it the modern doctrine that the Crown can impose no
financial burden on the people without consent of Parlia-
ment. Before discussing how far such an estimate is
justified, it will be necessary to examine the historical
context, with special reference to two classes, feudal tenants
and the citizens of London respectively.
I. Protection of Crown Tenants from arbitrary Exactions,
Apart from payments such as reliefs and amercements, the
occasions of which were independent of the royal will,
feudal exactions were of two types : scutages and aids.
By these two expedients the King could arbitrarily increase
the burdens of his feudal tenants beyond the letter of the
original feudal contract. Recognized usage, however,
required the consent of the vassals before they were sub-
jected to extraordinary exactions. The barons were within
their rights in seeking to embody this general principle in
Magna Carta, although it would appear (from comparison
of the versions of 12 15, 12 16 and 12 17) that they had
difficulty in devising a proper formula to give effect to it.
The present chapter attempts a rough compromise of the
question at issue, by requiring consent of the Crown tenants
to all scutages and also to aids other than the recognized
three. ^
(i) Feudal aids. The three recognized aids are liere
*See supra, p. 65.
CHAPTER TWELVE 233
specified, but no reform is attempted with regard to them,
and in particular (in marked contrast to the care taken in
chapter two to define the exact rate of " relief "), nothing
is said of the amount payable in name of " aid." It is only
the extraordinary aids^ that are regulated by this chapter :
these are not to be taken without " common counsel " or
the " Common Council " — for the Latin will bear either of
these two meanings, which indeed in 12 15 were probably
not yet differentiated from each other. If the Crown
tenants by " common counsel " could refuse a grant, they
could a fortiori make one upon conditions; fixing, for
example, the amount of an extraordinary aid as well as the
occasions of its payment. So far as aids were concerned,
there was here no innovation upon existing practice.
(2) Scutage. With regard to scutage, the requirement of
consent was something very different. Scutage, in lieu of
military service, was of the essence of the feudal relation :
to make it impossible for the Crown ever to levy a scutage
without consent of those who had to pay, was to go much
beyond redress of the grievance suffered under John : it
was to impose on him restrictions that his father had never
acknowledged.^
The total omission of this chapter in 12 16 may have been
partly occasioned by the consciousness that it contained an
innovation unwarranted by custom: the reissue of 12 17
said nothing of aids, and contented itself, in regard to the
vexed question of scutages, with the vague declaration that
for the future these should be taken as had been the custom
under Henrv 11.^
In spite, however, of the omission of chapter 12 from all
reissues of the Great Charter, it was customary for Henry's
advisers to consult " the Common Council " before exacting
a scutage or aid. This was done, for example, in 1222,
1 " Extraordinary aids " here mean all aids other than the three normal ones.
2 Miss Norgate, Minority, 15, thinks the innovation so undoubted as to justify
Innocent's Bull annulling the Great Charter. Cf. Adams, Origin, 276 n. : "a
demand in regard to scutage which custom did not warrant." Cf. ibid., 221-2,
and supra, 71.
2 See supra, p. 148.
234 MAGNA CARTA
when a Council granted an " aid for the Holy Land " of
three marks for an earl, one mark for a baron, and twelve
pence for a knight.^ The consent of a Council, indeed,
was usually taken even for one of the three recognized
feudal aids.
II. Protection of London from arbitrary Exactions.
Some attempt was made to protect the men of London from
arbitrary demands : the insurgent leaders in this way dis-
charged part of their debt to an ally with claims upon their
gratitude.2 The Articles of the Barons contained important
provisions affecting London ; and these were embodied in
the Charter in slightly altered terms. ^ The present clause,
for example, uses only one word, "aic^s," where the 32nd
Article of the Barons referred to "tallages and aids."
There is no evidence to show whether the omission had
been deliberately planned, or was the result of inadvertence ;
and the ambiguity inherent in both words makes it danger-
ous to hazard a dogmatic opinion on the practical effect of
the alteration. Yet a clearly-marked line can be traced
between the respective meanings of the two terms when
they are technically used."*
(i) " Aidy" a vague word, is applicable to any payment
that can be regarded as, in any sense, a freewill offering. It
embraced gifts to the Crown, whether from prelate or
burgess or feudal baron. London was stimulated towards
acts of generosity by Kings of England both before and
after John. There were times when "voluntary" aids
(like the "benevolences " of Tudor days) could not safely
be withheld.^
^Miss Norgate, Minority^ p. 194. ^See supra^ p. 35.
3 See Article 23 (which became c. 33), Article 31 (c. 41), and Article 32 (cc. 12
and 13), and cf. supra, p. 117. Whether Article 12 (c. 35) was more a benefit
to, than a restraint upon, traders seems doubtful.
* See, however, Ballard {British Borough Charters, Ixxx. ff. ) who seems to
make the two things shade into each other.
^Bracton, I. 2S8, holds that aids of this sort are personal not predial, for they
look to persons not fiefs. Auxilium burgorum was sometimes a technical term,
meaning sums paid by boroughs in lieu of 'Danegeld. See Round, Eng. Hist.
Rev., XVIII. 309. In our text, however, "aids" must be more broadly inter-
preted.
CHAPTER TWELVE 235
^^ (2) Tallage would appear to mean a toll or exaction
imposed on individuals who had no option of refusal.
Villeins were talliable at their lord's caprice, without appeal.
Liability to tallage, however, did not necessarily imply
servile status; for the King could tallage all inhabitants
of towns on royal demesne. London itself, for all its
wealth, political importance, and chartered privileges, still
shared this unwelcome liability.^
(3) Comparison of Aid and Tallage, The " aid," being '
a voluntary offering, differed fundamentally from tallage,
which was a forced payment. In theory, the citizens were
free to name the sum they proposed to pay. If the King
was satisfied, the city collectively became responsible for
assessing, collecting and paying over the money : the
King's representatives had no need nor right to interfere
with individual citizens. The amount of a tallage, on the
contrary, was fixed by the King's Justices, assessed by
them per capita on individual citizens, who were subject to
direct distraint by the agents of the Crown. It was to the
advantage of a borough to forestall, by a liberal aid, the
Crown's anticipated demand for a tallage, for the hated
tax-gatherer was thus kept outside the city gates. An aid
was more to the King's advantage also than a tallage :
not only was he saved the trouble, expense, and delay of
collection, but he obviated risk of loss through the insolv-
ency of some of the individuals fixed upon.
A story told by Madox^ brings out the contrast. A
dispute had arisen between the King and the Londoners
in 1255. To Henry's demand for 3000 marks of " tallage,"
they at first replied by offering 2000 marks of " aid," which
the King refused. The citizens then denied outright their
liability to tallage, but were confronted with entries in
Exchequer and Chancery Rolls which contradicted their
contention. On the morrow, the mayor and citizens
^This statement, for which evidence is given infra, is not always admitted.
Taswell-Langmead, En^. Const. Hist., p. 107, says: "The city of London can
never have been regarded as a demesne of the Crown." For lists of prelates and
barons paying tallage see Ludwig Riess, Historische Zeitschrift, Vol. 14, N.S.
pp. 21 ff. (1904).
2 T. 712, citing Mem. Roll 39 Henry III.
236 MAGNA CARTA
acknowledged that they were talliable, and paid the sum
demanded.
(4) London's attempts to escape tallage. There is ample
evidence that London in John's reign was galled by the
liability to tallage, and was ready to seize any loophole of
escape. John's letter to the city in 1206 ^ refers to the
serious damage done to his capital by the manner in which
tallages had been assessed and collected. A document
compiled about 12 10, in the interests of London, partly
from authentic sources, purporting to be a Charter by
William L, declares that all freemen shall hold their lands
and possessions " free from every unjust exaction and from
every tallage." ^ Finally, Miss Bateson in 1902 ^ called
attention to a document of nine articles, which seem to be
the heads of a petition prepared by the Londoners, probably
in 12 15, in which they ask inter alia the abolition of all
tallages except per comrnunem assensum regni et civitatis.
(5) Effects of omission of " tallage " from Magna Carta.
Why, if not through pure inadvertence, was the word
" tallage," occurring in Articuli Baronum, omitted from the
Charter? Widely different answers have been given.
Prof. G. B. Adams ^ ingeniously argues that the omission
was deliberately made in the interests of London. That
city, now a full-blown commune, enjoyed the status of a
feudal vassal : though liable to aids, its burghers resented
any allusion to the servile " tallage " in connection with
themselves. If Prof. Adams here interprets their attitude
aright, the Londoners were ill-advised to refuse, on any
such punctilio, to secure in the Charter incorporation of a
definite protection from arbitrary tallage by the Crown —
a grievance from which they were destined to suffer for
more than a century thereafter.
The true explanation, however, is more likely to lie in
an opposite direction. The omission was, perhaps, made
deliberately to the detriment of London, in deference to
John's strong feeling on a point that did not affect the
^jRoL C/aus., I. 64.
2 Willehii Articuli Londoniis Retractati, in Liebermann, Gesetze, I. 490, c. 5.
"^Eng. Hist. Rev., XVII. 726. ^ Ibid., XIX. 702 ; Origin, 358 fif.
CHAPTER TWELVE 237
barons personally. John, for his part, would be readily
persuaded to renounce the right to take " aids " from the
wealthy traders of the capital, if he preserved the more
drastic privilege of tallaging them at will. The word
" tallage " was dropt from the Charter, not to gratify
London's pride, but to enable the Crown to have access
to the city's treasure chests.
(6) Nature of the protection accorded to London. The
arrangement of this chapter is noteworthy : after securing
redress of abuses pressing on the barons, a few compara-
tively careless words are added : " in like manner it shall
be done concerning aids from the city of London." The
words " in like manner " are difficult to interpret, for the two
cases are far from parallel. Do they mean that no aid can
be taken from London without the same " common counsel
of the realm " previously stipulated for the taking of
scutages from the tenants in chief ? Probably not, for the
method provided in chapter 14 for obtaining " the common
counsel " would have been peculiarly ill-adapted to protect
the Londoners, whose interests were not represented in the
baronial assembly. The Petition of nine heads ^ had asked
more than this, namely, that no tallage should be taken
without common assent " of the kingdom " (that is, of the
baronial assembly) and " of the city " — a double consent
being thus required, as though " the common counsel " was
not enough.
High authorities suggest a different explanation for the
clause in chapter 12, which is read simply as an assertion
that only " reasonable " aids should be taken from London. ^
If that be so, no criterion of reasonableness is suggested,
and such might be difficult to find.^ Subsequent history
sheds no clear light on the intention of this clause. As
the chapter was omitted from all reissues, no occasion ever
arose of testing its meaning by actual practice.
In deciding between the two suggested explanations,
^ See stipra, p. 236. 2 Lords' Report on the Dignity of a Peer, I. 65.
^In 1 168, when Henry II. took an aid for the marriage of his daughter, London
contributed £(>\'J i6s. 8d., which might afford a precedent for a "reasonable " aid.
See Pipe Roll, 14 Henry II., cited Madox, I. 585.
238 MAGNA CARTA
however, it should be noted that, though " councils " framed
on the model of 12 15 continued for half a century to meet,
they made no claim to interfere with the Crown's right to
tallage London. Neither Henry nor Edward waited for the
" common counsel of the realm " before enforcing their
demands.
Whatever may have been the intention of the framers of
this clause with regard to London, it is notable that they
allowed that city to stand alone. Magna Carta completely
ignored that provision of the Articles of the Barons which
extended the same protection " to citizens of other places
who thence have their liberties," meaning the boroughs
whose chartered privileges had been modelled upon those
of the metropolis.^ Here, again, the alteration was pro-
bably a concession to John made by the barons at their
allies' expense.^
(7) Later history of the Crown's right to tallage the towns.
The Crown continued at intervals to take tallages from
London until 1340. It has sometimes been maintained,
indeed, that the Confirmatio Cartarum of 1297 was intended
to abolish this prerogative, and a document once con-
sidered an authoritative version of the Confirmatio bore the
suggestive title of De tallagio non concedendo. It is now
well known that the latter document is unauthentic ; while,
if the Confirmatio itself was intended to relieve the towns
from tallages, it signally failed. Edward III. exacted
tallages from London and other towns. Parliament,
however, succeeded, in 1340, in passing a statute which
abolished unparliamentary taxation of every kind. This
act, sometimes styled by modern writers " the real statutum
de tallagio non concedendo," finally settled the law,^ but
did not prevent the King from trying to break that law.
^ Cf. however, Davis, England under Normans^ 380.
^ It might be argued that the last clause of chapter 13, extending to all towns a
confirmation of liberties and customs, was intended to embrace this provision as to
aids. If so, the draftsman has expressed himself clumsily,
^ See Stubbs, Const. Hist., II. 548. *' Of the scope of this enactment there can
be no doubt ; it must have been intended to cover every species of tax not author-
ised by parliament, and ... it seems to have had the effect of abolishing the royal
prerogative of tallaging demesne. "
CHAPTER TWELVE 239
Edward frequently disregarded the restrictions placed upon
his financial resources, and with varying success. He
rarely did so, however, without meeting protests; and the
rule of law laid down in the act of 1340 was never repealed.
III. Magna Carta and the Theory of Parliamentary
Taxation. It is a commonplace of our text-books that
chapters 12 and 14, taken together, amount to the Crown's
absolute surrender of all powers of arbitrary taxation, and
even that they enunciate a doctrine of the nation's right
to tax itself.^ Yet the very idea of " taxation " in its
abstract form, as opposed to specific tallages and exactions,
levied on definite things or individuals, is essentially
modern. The doctrine of the day was that the King in
normal times ought "to live of his own," like any other
land-owning gentleman. A regular scheme of " taxation"
to meet t^e ordinary expenses of government was un-
dreamt of. It is too much to suppose, then, that our
ancestors in 1215 sought to abolish something which,
strictly speaking, did not exist. The famous clause treats,
not of " taxation " in the abstract, but of the scutages and
aids already discussed. It does not concern itself with the
rights of Englishmen as such, but chiefly with the interests
of barons who held freeholds of the Crown, and incidentally
and inadequately with those of the citizens of London.
Several considerations place this beyond reasonable
doubt.
(i) The terms of the restriction are by no means wide
or sweeping; but precise, accurate, and narrow. The
" common counsel of the realm " was required for three
exactions at the most : for scutages and for extraordinary
aids from feudal tenants, and possibly also for aids from
the city of London : that is all. Not a word is said of
other forms of taxation or other groups of taxpayers. (2)
If under-tenants received, by chapter 15, protection against
mesne lords, they received none against the King. The
Charter affected, not national "taxation," but feudal dues.
^E.^. Taswell-Langmead, £ng/. Const. Hist., io6. Dr. Stubbs, Const. Hist.,
I. 573, considers that these words ** admit the right of the nation to ordain
taxation."
240 MAGNA CARTA
(3) The scant measure of protection did not extend even to
all Crown tenants. The King's villeins were, of course,
excluded; and so were even freeholders whose tenure was
other than that of chivalry. Socage tenants were left liable
to carucage, while the-GfOtvh's right to raise the " farms "
of its own demesnes was reserved. ^ (4) The Crown's initia-
tive in " taxation " (here restricted in regard to " aids " and
" scutages ") was, under many other names and forms, left
intact. The King required no consent before taking prises
and custom dues from merchandise reaching or leaving
England, or before taking tolls and fines at inland markets
under the plea of regulating trade. Tallages also were
exigible at discretion from aliens and Jews, from tenants
of demesne, from London and other chartered towns. (5)
The assembly to be convened for taking " common counsel "
was a narrow body, representative neither of the ranks and
classes of the community, nor of the separate national
interests, nor yet of the various districts of England. Its
composition was homogeneous, an aristocratic council of
the military tenants of the Crown, convened in such a way
that only the greater among them were likely to attend.^
These facts serve as a warning not to read into Magna
Carta modern conceptions which its own words will not
warrant. This famous clause was far from formulating any
doctrine of self-taxation ; it primarily affected impositions
levied by John, not qua sovereign but qua feudal lord.
Such as it was, it was omitted, along with its corollary
(chapter 14), in 12 16 and subsequent reissues.
CHAPTER THIRTEEN.
Et civitas Londonie habeat omnes antiquas libertates et
liberas consuetudines suas, tam per terras, quam per aquas.
Preterea volumus et concedimus quod omnes alie civitates,
^ See infra, under c. 25.
2 Even when an honour escheated, its tenants "were not suitors of the Curia
Regis.^^ See Report on Dignity of a Peer, I. 60.
CHAPTER THIRTEEN 241
et burgi, et ville, et portus, habeant omnes libertates et
liberas consuetudines suas.
And the city of London shall have all its ancient liberties and
free customs, as well by land as by water; furthermore, we
decree and grant that all other cities, boroughs, towns, and
ports shall have all their liberties and free customs.
A full list of London's liberties and customs would be
a long one ; and to relate how each of these grew up and
was confirmed by the Crown need not be here attempted.
The most cherished of the privileges enjoyed in John's day
by the citizens were the right to appoint a civic chief, who
bore the name of mayor, and the right to choose the sheriffs
who should collect the city's firma ^ (or annual rent payable
to the exchequer), so as to obviate the intrusion of royal
bailiffs. Qnly a brief account of the way in which the
metropolis obtained these privileges is here required.
The chief feature of London before the Norman Conquest
seems to have been lack of proper municipal organization.
Dr. Stubbs describes the capital during the eleventh century
as " a bundle of communities, townships, parishes, and lord-
ships, of which each has its own constitution." 2 It was
thus a collection of small administrative units, rather than
one large unit. Some semblance of legal unity was, it
is true, afforded by the folkmoot, in which the citizens
regularly assembled; by its smaller council known as
" husteng " ; and perhaps also by its " cnihtengild " (if,
indeed, this third body be not entirely mythical) ; while
the existence of a " portreeve " shows that for some financial
purposes the city was treated as one whole. London, how-
ever, prior to the reign of Henry I. was far from possessing
the machinery of an efficient municipal government.
The first step towards a constitution is generally supposed
to have been taken by the citizens when they obtained a
charter from Henry I. in the last years of his reign (i 130-35).
This is not strictly accurate. London, indeed, by that
1 Firma is explained infra^ c. 25.
2 Stubbs, Const. Hist.^ I. 439. Round, Commune^ 220, is in substantial agree-
ment. Miss Bateson, however, thinks "there has been a tendency unduly to
minimise the measure of administrative unity in the twelfth-century shire of
London." See evidence produced by her, Engl. Hist. Rev., XVII. 480-510.
Q
242 MAGNA CARTA
grant gained valuable privileges; but it did not obtain a
constitution. The chief rights actually conferred by Henry
were as follows : — (i) The firma was fixed at the reduced
rate of £300 per annum, the citizens obtaining a lease in
perpetuity of their own city with the surrounding county
of Middlesex — the grant being made to the citizens and
their heirs; (2) they acquired the right to appoint the
sheriffs of London and Middlesex, implying the exclusion
of the King's tax-collectors by men of their own choosing ;
(3) a similar right of appointing their r)wn nominee as
justiciar was also conferred on them, to the exclusion
apparently of the royal justices of eyre. Many minor
privileges were confirmed which need not here be specified.
Mr. J. H. Round ^ argues with convincing force that these
concessions, important as they were, did not confer a civic
constitution upon London. Henry's charter, in his
opinion, confirmed the separate jurisdictions and franchises,
perpetuating the old state of disunion, rather than creating
a new principle of cohesion. Mr. Round proves, further,
that the new concessions were cancelled by Stephen in 1 141,
when Geoffrey de Mandeville compelled Stephen to appoint
him as sheriff and justiciar of London. Earlier in the same
year, the citizens had risen against Matilda and tried to
establish a sworn Commune, presumably of the continental
type.2 When London was placed in Earl Geoffrey's hands,
all vestige of this would be swept away, along with any of
the privileges granted by Henry L that had endured till
then.
Henry H., indeed, granted a charter in 1155, which is
usually interpreted as a full confirmation of the concessions
of the earlier Henry .^ Mr. Round has proved the error of
this opinion.^ The charter of 1155 restricted, rather than
enlarged, the privileges of London, being couched in
cautious and somewhat grudging terms. The main conces-
sions of the earlier charter were omitted : the citizens no
longer elected their sheriffs or justiciar ; the reduction of the
^ Geoffrey de Mandeville^ 356. 2 William of Malmesbury, 11. 576.
3 See e.g. Miss Norgate, Angevin Kings, II. 471.
* Geoffrey, 367.
CHAPTER THIRTEEN 243
firma to £300 was not confirmed; and subsequent pipe
rolls show that Henry doubled that amount.
The next crisis came early in Richard's reign. Then it
was, perhaps, that London obtained its municipal constitu-
tion. Then also it may have regained the privileges
precariously held under Henry I. and Stephen. The form
in which the constitution came at last was, Mr. Round
argues, borrowed from France, and was neither more nor
less than the Commune, so well known on the Continent in
the twelfth and thirteenth centuries. Mr. Round ^ has
shown that these concessions were not, as has sometimes
been supposed, voluntarily granted in 1189 by Richard I.,
but were extorted from his brother John, when that
ambitious^prince was bidding for powerful allies to support
his claim to act as Regent. London, Mr. Round main-
tains, got its constitution on 8th October, 1191, under
picturesque and memorable circumstances. While Richard
tarried in the Holy Land, a scramble took place at home for
the right to represent him. The Chancellor Longchamp
had been appointed Regent; but John, wily and unscrupu-
lous, ousted him, with the help of the men of London. At
the critical moment, the metropolis had offered support on
conditions, which included restoration of the short-lived
privileges conferred by Henry L, and, in addition, a
municipal constitution of the continental type.
Mr. Round, in a notable passage, describes the scene.
"When, in the crisis of October, 1191, the administration
found itself paralysed by the conflict between John, as the
King's brother, and Longchamp, as the King's represen-
tative, London, finding that she held the scales, promptly
named the ' Commune ' as the price of her support. The
chronicles of the day enable us to picture to ourselves the
scene, as the excited citizens, who had poured forth over-
night, with lanterns and torches to welcome John to the
capital, streamed together on the morning of the eventful
8th October at the well-known sound of the great bell,
swinging out from its campanile in St. Paul's Churchyard.
There they heard John take the oath to the 'Commune,'
^ Commune of London, 222.
244 MAGNA CARTA
like a French king or lord; and then London, for the first
time, had a municipality of her own." ^
For any accurate definition of a Commune we look in
vain to contemporary writers. Richard of Devizes ^ quotes
with approval, " Communia est tumor flehis, timor regni,
tepor sacerdotii." Some insight has been gained in recent
years, however, into its exact nature. A Commune was a
town that had obtained recognition as a corporate entity,
as a link in the feudal chain, becoming the free vassal of
the King or other lord, and itself capable of having sub-
vassals of its own. 3 Its chief institutions were a mayor and
elective council, generally composed of twenty-four mem-
bers, some or all of whom were known as echevins or
skivini. Perhaps, the chief peculiarity of the Commune
was the method of its formation, namely, by popular
association or conspiracy, involving the taking of an oath of
a more or less revolutionary nature by the citizens, and its
subsequent ratification by those in authority. It is gener-
ally admitted that these communes, though revolutionary
in origin, were not necessarily democratic in their
sympathies.
From 1 191 onwards, London was governed by its own
mayor, an official chosen by the citizens, but holding office
for life, until the citizens obtained a further concession in
1 2 15. It has sometimes been argued that as a mayor was
the natural head of a Commune, the continued existence of
the one implied the existence of the other. It seems more
likely, however, that if a Commune was actually set up in
1 191, it did not long survive Richard's return from cap-
tivity. Mayors were to be found in the twelfth century
ruling over boroughs that were not technically " Com-
munes " ; and Richard may have been willing to accept a
mayor of London's choosing, while he repudiated the city's
claim to independence as a Commune.
When John became King, he granted three charters to
the capital for a gersuma (or slump payment) of 3000
1 Commune of London, 224. ^Select Charters, p. 252.
^Luchaire, Communes Fran^aises, p. 97, defines it as '' seigneurie collective
populaire.''^
CHAPTER THIRTEEN 245
marks.i All franchises specified in the charter of Henry I.
were confirmed, with one exception : the liberty to appoint
a justiciar of their own, now seen to be inconsistent with
the Crown's centralizing policy, was abandoned. None of
these charters made mention of mayor or commune, but
they confirmed some minor privileges gained in Richard's
reign .2
A fourth charter, dated 20th March, 1201, was of tem-
porary interest. The fifth and last of the series came in the
crisis of 12 15, and some light is possibly shed on it by
comparison with the petition of nine articles already men-
tioned,^ which seems to represent the demands made by
the Londoners at that date. Besides exemption from
arbitrary tallage and several minor concessions, they
demanded the control of Thames, the annual election of
their mayor in the folkmoot, freedom of access for foreign
traders, and the right to distrain for debt against the persons
and property of debtors.
Some of these demands were granted by John's fifth
charter, dated gth May, 12 15, some five weeks previous to
Magna Charter, and representing the bait thrown by John
to gain their support in this new crisis as he had gained
it in the earlier crisis of 1191. The men of London
obtained the right to appoint a mayor annually, and, if they
chose, to depose him at the year's end and appoint another
in his place, a right which Miss Norgate aptly calls " the
crowning privilege of a fully constituted municipality." ^
The charter at the same time confirms all liberties already
enjoyed, " as well within London as without, as well on
water as on land, salva nobis chamherlengia nostra." The
control of Thames and Medway, mentioned with more
particularity in Magna Carta, seems to be here granted;
while the freedom of access of foreign merchants is qualified
1 Miss Bateson, En^/. Hist. Rev., XVII. 508.
2 E.g. removal of obstacles in Thames and Medway. Cf. infra, c. 33.
^ Supra, p. 236.
'^John Lackland, 228. From this date the list of mayors shows frequent, some-
times annual, changes. Serlo, the mercer, was mayor in May, 121 5, when London
opened its gates to the insurgents, while William Hardell had succeeded him
before 2nd June, 1216.
246 MAGNA CARTA
by John's reservation of the right to take toll from them by-
appropriating such of their choicest wares as his chamber-
lain might select for the royal household.^
If the nine articles contain London's demands in 12 15,
the Charter of 9th May gives what John was willing to
promise in return for the city's support; and the Articuli
Baronum what the barons compelled him to grant to the
city after it had preferred their alliance to his ; while Magna
Carta shows some slight modifications in the King's favour.
Such was the London whose privileges were confirmed by
chapter 13 of Magna Carta in words that avoided details and
confined themselves to a general confirmation of ancient
"liberties and free customs." ^ Neither mayor nor Com-
mune is mentioned; but the question has been raised
whether by implication the Great Charter does not recognize
the existence of one or both of these.
As the charter of 9th May had granted to London the
right to elect a mayor, and as the mayor was appointed one
of the 25 executors under chapter 61, it is clear that Magna
Carta accepted that magistrate as head of the city's govern-
ment ; and the recognition of a mayor has sometimes been
held to suggest also the recognition of a Commune.
Professor Adams, on the other hand, has based an argu-
ment for the existence of a Commune after June, 12 15,
mainly upon the omission of the word tallage from chapter
12, which thus makes it possible to infer that an auxilium
is the only imposition to be lawfully levied on London .^ He
seeks to show, further, that London lost this status of a
Commune in 12 16, when the charter was reissued without
the chapter associating London with the payment of
auxilium : " this clause was omitted, and with it London's
legal right to a Commune fell to the ground." ^
^ See text of Charter in Se/. Chart. ^ 315.
"^ The meaning of both words is discussed infra, c. 39.
3 See supra, p. 236. M. Petit-Dutaillis {Studies Supplementary, 102) doubts
whether the citizens in 1215 had any wish to become a Commune, and holds that
their desire was to escape burdensome exactions, no matter what these might be
called. Prof. Adams {Origin, 367) maintains, in reply, that the only practicable
method of effecting this exemption was to obtain recognition as a Commune.
^Ibid.,z(,i.
CHAPTER THIRTEEN 247
It is pertinent to note, however, that the Patent Rolls
for 1221 ^ refer to " the mayor and Commune of London."
If this implies the existence of a real Commune of the conti-
nental type, the date of its final abolition may possibly have
been the year following, when London quarrelled with the
young King's ministers and had difficulty in making peace.^
On the whole, it must be left an open question whether or
not the privileges granted to London in 12 15 included the
establishment of a Commune, and, if so, when that form of
municipal government came to an end.
In this chapter of John's Magna Carta (in contrast with
the last clause of chapter 12), London did not stand alone.
** All other cities, boroughs, towns and ports " were con-
firmed in their liberties and free customs. A specification
of these was, of course, impossible ; each borough was left
to prove its privileges as best it might. In the reissues of
Henry, London shared the distinction of being mentioned
by name with " the barons of the Cinque ports," who from
their wealth, their situation, and their fleet, were allies worth
conciliating. They played, indeed, a prominent part in the
decisive naval victory gained by Hubert de Burgh on 24th
August, 12 17. 3
Among the most cherished privileges claimed by the
chartered boroughs were the rights to exact tolls and to
place oppressive restrictions upon rival traders not mernbers
of their guilds, foreigners and denizens alike. The general
confirmation of privileges in this chapter has been held
to contradict chapter 41, which grants protection and
immunities to foreign merchants.^ The inconsistency,
however, is perhaps greater in appearance than reality,
since the later chapter aimed at abolition of " evil customs "
inflicted by the King, not of those inflicted by the boroughs.
At the same time, any favour shown to aliens would be
bitterly resented by English traders. If the charter had
been put in force in its integrity, the more specific privileges
in favour of foreign merchants wo'uld have prevailed in
''Rot. Pat., 303-4.
2 See Norgate, Minority, 186, and authorities there cited.
2 See supra^ p. 145. ' * Cf. Pollock and Maitland, I. 447-8.
MAGNA CARTA
opposition to the vague confirmation of borough " liberties,"
wherever the two conflicted.^
Other portions of John's Great Charter that specially
affected Londoners were the last clause of chapter 12, and
chapters S3 ^i^d 4^ ; while many of the privileges granted
or confirmed in other chapters were shared by them.
CHAPTER FOURTEEN.
Et ad habendum commune consilium regni, de auxilio
assidendo aliter quam in tribus casibus predictis, vel de
scutagio assidendo, summoneri faciemus archiepiscopos,
episcopos, abbates, comites, et majores barones, sigillatim
per litteras nostras; et preterea faciemus summoneri in
generali, per vicecomites et ballivos nostros, omnes illos qui
de nobis tenent in capite; ad certum diem, scilicet ad ter-
minum quadraginta dierum ad minus, et ad certum locum ;
et in omnibus litteris illius summonicionis causam sum-
monicionis exprimemus; et sic facta summonicione nego-
cium ad diem assignatum procedat secundum consilium
illorum qui presentes fuerint, quamvis non omnes sum-
moniti venerint.
And for obtaining the common counsel of the kingdom anent
the assessing of an aid (except m the three cases aforesaid) or
of a scutage, we will cause to be summoned the archbishops,
bishops, abbots, earls, and greater barons, severally by our
letters ; and we will moreover cause to be summoned generally,
through our sheriffs and bailiffs, all others who hold of us in
chief, for a fixed date, namely, after the expiry of at least forty
days, and at a fixed place ; and in all letters of such summons
we will specify the reason of the summons. And when the
summons has thus been made, the business shall proceed on
the day appointed, according to the counsel of such as are
present, although not all who were summoned have come.
This chapter, which has no equivalent among the Articles
of the Barons, appears here incidentally : it would never
have found a place in Magna Carta but for the need of
machinery to give effect to chapter 12.2
^Cf. in/ra, c. 41.
*0n the whole subject of the comimme concilhan, cf. supra, 129-131 and 149.
CHAPTER FOURTEEN 249
As chapter 12 is frequently supposed to enunciate a
general doctrine of taxation, so this one is cited as enunci-
ating a doctrine of parliamentary representation ; while the
close connection between the chapters is taken as evidence
that the framers of Magna Carta had grasped the essentially
modern principle that taxation and representation ought
always to go together. ^ In this view, the barons at Runny-
mede are given credit for anticipating the best features
of modern parliamentary government. The text, however,
will scarcely bear so liberal an interpretation. ^ Vital points
of difference between the principles of Magna Carta and the
modern doctrine of representation are revealed by analysis.
Under chapter 12, scutages and extraordinary aids could
only be levied "with common counsel of our kingdom,"
and now chapter 14 fixes authoritatively the composition of
an assembly charged with this function. The same Latin
words which signify joint " consent " or counsel came to
signify also the "Common Council," afterwards of vital
constitutional importance, continuing under a new name the
old curia regis, and passing in turn into the modern Parlia-
ment. The duties and constitutional importance of this
commune concilium may be considered under six heads.
I. Nature of the Summons. Formal writs had to be
issued, specifying the time, place, and reason of assembling,
at least forty days in advance. Each of the really powerful
men of the realm — archbishops, bishops, abbots, earls, and
" other greater barons " — received a separate writ addressed
to him individually, while the " smaller barons " were sum-
moned collectively and indirectly through the sheriffs and
bailiffs of each district.
II. Composition of the Council. It is clear that the
meetings contemplated were purely baronial assemblies,
since none but Crown tenants were invited to attend. " The
common consent of my kingdom," in John's mouth, was
^ E.g: Anson, Lmu and Custom of the Constitution (ist ed.), I. 14, declares that
one of the two cardinal principles of the Charter is "that representation is a con-
dition precedent to taxation." This has been altered in later editions.
2 Prof. Adams {Origin, 276 n.) perhaps goes too far towards the opposite extreme
in holding this chapter " an unnecessary addition to the Articles of the Barons and
quite without importance." Contrast Round as cited infra, p. 251.
250 MAGNA CARTA
synonymous with " the consent of my barons." ^ The
King's Council had by this time freed itself from any com-
plicated theories as to its own composition, which may ever
have hampered it. It was now entirely homogeneous, a
feudal muster of Crown-vassals.^
It is unnecessary here to examine the rival theories pro-
fessing to explain the composition of the Anglo-Saxon
Witenagemot, or to discuss the exact connection between
that institution and the Curia Regis of the Norman Kings.
As matter of fact, the early constitution of the court of the
Conqueror or of Ruf us seems to have been monarchic rather
than aristocratic or democratic ; that is to say, it depended
to a great extent on the personal will of the King. No
evidence exists, of date anterior to the Great Charter, of
any magnate thrusting himself unbidden into a royal
council or forcing the King to issue a formal invitation.
On one occasion, indeed, the action of Henry II. in omit-
ting to issue a writ laid him open to criticism. This was in
October, 1164, when a special council was summoned to
Northampton to pass judgment upon questions at issue
between the King and Thomas a Becket. The primate was
ordered to appear for judgment; but the formal writ of
summons, which every holder of a barony was wont to
receive, was withheld. Apparently, contemporary opinion
condemned this omission.^ It is safer to infer, then, that
as early as 1164, the method of issuing these writs had
^This is illustrated by comparison with the phrases in which Henry and his
sons expressed •* the common consent " : e.^. (i) the Assize of Clarendon in 1166
{Select Charters, 143) bears to have been ordained by Henry II. **^<j consilio
07nniuin baj-omiin stio?-um^^ ; (2) John's Charter to Innocent in 1213 declares that
he acted ^^ coitinmni consilio baronum nostrorttin^'' {Select Charters, 285); (3)
Matthew Paris makes Earl Richard complain to Henry III. in 1255 that the
Apulian business had been entered on '•'■sine consilio suo et assensu barnagii"
{Chron. Maj\, V. 520).
2Cf. Round {Peerage and Pedigree, 349 ff.), who speaks of this as creating "a
harsh and artificial division of society." Its composition was stereotyped, and
Mr. Round rejects alike the theory of Stubbs {Const. Hist., I. 566) that the
Council was being gradually extended, and that of Freeman {Norman Conquest,
V. 419) that it was suffering contraction. Cf. also Adams, Origin, 226 n., and
the authorities there collected.
^ See Ramsay, Angevin Empire, p. 54, and authorities there cited.
CHAPTER FOURTEEN 251
become uniform, but this constitutional understanding was
not reduced to writing until embodied in Magna Carta.
It was in 12 15 that the magnates of England formulated a
distinct claim to be present at the King's councils; -^nd
even then the demand only referred to assemblies sum-
moned for one specific purpose. Previously, attendance
was reckoned not as a privilege, but rather as a burden
incident to the possession of land.^
Mr. Round ^ maintains that under John " the writ of
summons suddenly assumed a very real importance," and
argues, with much plausibility, that the present chapter
proves " that the Crown had been endeavouring to iise
the writ as a means of excluding its opponents from the
assembly." The barons, on their part, unable to assert a
right to attend uninvited, " insisted that they all must be
summoned."
III. Position of " Minor Barons." Crown-tenants varied
in power and position from the great earl, who owned the
larger share of one or more counties, to the small free-
holder with a few hides or acres of his own. A rough
division was drawn somewhere in the midst; but the
boundary was vague, and this vagueness was probably
encouraged by the Crown, whose requirements might vary
from time to time.^ The Crown-tenants on one side of this
fluctuating line were harones majores ; those on the other,
harones minores. The distinction had been recognized as
early as the days of Henry 11.;^ but Magna Carta helped
to stereotype it, and contributed to the growing tendency
to confine the word " baron " to the greater men.^ The
smaller barons grudged the long journeys and the expense
of attending Councils whose decisions they were powerless
1 See L. O. Pike, House of Lords, 92, "There is no trace of any desire on the
part of the barons to be summoned to the King's great Council as a privilege and
an honour before the reign of John." Cf. also Report on the Dignity of a Peer, I.
389.
2 Peerage and Pedigree, 355-6. ^ See Prof. Medley, Eng. Const, Hist., 123.
^Dialogus de Scaccario, II. x.D., '' baronias scilicet majores seu minores.^^
^Cf. supra, c. 2. Prof. Vinogradoff, Law Quart. Rev. XXI. 255, shows that
" baronia " long remained a technical term for the body of freemen holding from
the king, both great and small.
252 MAGNA CARTA
to influence ; and they found a more fitting sphere for their
energies in the meetings of the shire. For these reasons,
they were prepared to ignore any summonses they might
receive. In this respect, in Mr. Round's ^ opinion, the
feudal theory " broke down in England."
Three distinct theories have been advanced as to the
position occupied by the "minor barons " in the Common
Council, (i) The duty of attendance was burdensome on
the poorer Crown-tenants. It has been suggested that the
device of inviting them by general summons was intended
as an intimation that they need not come. This is the view
taken by Prof. Medley .^ (2) Dr. Hannis Taylor holds an
opposite opinion, reading this chapter as an attempt " to
rouse the lesser baronage to the exercise of rights which
had practically passed into desuetude." ^ If such an
attempt had really been made, and had succeeded, the
result would have been to leave no room for the future
introduction of the representative principle into the national
council. (3) A third theory holds that the smaller Crown-
tenants were called in a representative capacity. A few
knights (probably elected for this purpose by their fellows)
were expected to attend to represent the others. Dr. Stubbs
seems predisposed towards this opinion, although he
expresses himself with his usual caution.^
It may be suggested, even at the risk of seeming to invent
a fourth theory in a series already too numerous, that to
the great men who framed the clause it was a matter of
supreme indifference whether their humbler fellow-tenants
attended or stayed away. The general summons expressed
neither an urgent desire for their presence, nor yet an
intimation that they were not wanted; but merely con-
formed with established usage, and left with each " minor
baron " the decision whether he should come or stay away.
1 op. cit.^ 353. Cf. also his A'm^j Serjeanties^ 36; Commune of London \ 252-3.
'^ Eng. Const. Hist., 123. "The smaller tenants-in-chief would thankfully
regard the general summons as an intimation to stay away."
^ Eng. Const., I. 466.
* See Const. Hist., I. 666. " Whether or no the fourteenth Article of the Great
Charter intended to provide for a representation of the minor tenants-in-chief by a
body of knights elected in the county court," etc.
CHAPTER FOURTEEN 253
His presence would make little difference upon the delibera-
tions of the magnates.
IV. Representation. It is well to hesitate before apply-
ing to ancient institutions a word so essentially modern as
" representation." In a sense, the reeve and four best men
of every village " represented " their fellows in the county
court from an early age ; and in a somewhat different sense
the feudal lord " represented " his free tenants and villeins
in the King's court; but* in neither instance was there any-
thing approaching the definite relation which exists at
present between the member of Parliament and his con-
stituents. Magna Carta shows no tendency whatever to
adapt this expedient of representation, even in its crudest
form, to the composition o(_ the Common Council. The
councillors whose summons was enjoined were all of one
type, military tenants of the Crown, each of whom was to
attend in his own interests not in those of his class, still
less of his district or of the community as a whole. The
barons, great and small, might be present, each man for
himself; but the other contributors to the King's exche-
quer were ignored.^
V. Powers of the Council. It was not until long after
the days of Magna Carta that Parliament secured the most
important of those functions now deemed essential to its
existence. No claim was made on behalf of the commune
conciliutn to be consulted in the making of laws or in the
performance of administrative duties by the Crown : no
effort was made towards formulating any doctrine of
ministerial responsibility. This assembly, narrow and
aristocratic in composition, had only one right secured
to it, a limited control over taxation. Even here, as we
^The writs of 7th November, 1213, are commonly regarded as introducing the
representative principle into the national assembly, and in this view the barons'
scheme embodied in Magna Carta has been considered as reactionary by com-
parison. Cf. Anson, Law and Czistom^ !• 44 : '* The provisions of 1 215 described
an assembly which was already passing away." There are difficulties, however,
connected with the interpretation of those writs ; and recent authorities are
inclined to point to 1264, rather than to 1213, as the beginning of the systematic
appHcation of representation to Parliament. See Adams, Origin^ 317, 340. Cf.
also sjipra, 29-30.
254 ..^> MAGNA CARTA
have seen, no general claim was put forward. It had no
right to control the national purse : the barons merely
protected their own individual pockets against an increase
of feudal burdens. A modern Magna Carta would have
contained a careful list of the powers and privileges of
" the common council of the realm." ^
It would, indeed, have been an evil thing for England,
if this narrow baronial assembly had established a claim
to tax the important classes of the community, townsmen
and vassals of mesne lords, who were totally unrepresented
in it. Doubtless, it would have been ready enough to
substitute, if it could, a scheme of taxation that relieved
Crown-tenants of the burden of scutages and aids, at the
expense of their humbler neighbours.
VI. Rights of Majorities and Minorities. The medieval
conception of solidarity was defective ; the King's council
j acted too much like a fortuitous gathering of unrelated
[individuals, and too Tittle like a recognized organ of the
i body politic. " No new exactions without consent of the
individual taxed" was nearer the ideals of 12 15 than "no
taxation without consent of Parliament." Each " baron "
was summoned on his own behalf; and it is doubtful how
far a dissenting minority could be bound by a decision of
the rest. Accordingly, the framers of Magna Carta deemed
it necessary to assert what would be too obvious to modern
politicians to require assertion — namely, that when the
commune concilium had been properly convened, its power
to transact business should not be lost because a section
of those summoned chose to stay away. " The business
shall proceed on the day appointed, according to the advice
of such as shall be present, although all that were sum-
moned do not come." Not all business was competent,
however, for the cause of summons had to be mentioned in
the writs. If these writs were in order, the Council, so we
may presume, had power to impose aids or scutages on
those who were absent. ^
* Cf. Report on Dignity of a Peer, I. ^t^.
^Cf. Stubbs, Const. Hist., I. 607: ''Absence, like silence, on such occasions
implies consent."
I
CHAPTER FOURTEEN 255
Nothing is said, however, as to the validity of a protest
made by those who came and expressed disapproval. As
the substance of this chapter was observed in practice
(though omitted from subsequent confirmations), a prece-
dent of the year 1221 may illustrate the interpretation put
upon it by contemporary practice. A Council summoned
by William Marshal had consented to a scutage, and the
Bishop of Winchester was assessed at 159 marks for his
knight's fees. He refused to pay, on the ground, quite
untenable by modern standards, that he had dissented from
the grant. The plea was accepted by the Regent, and the
exchequer adjudged bishop Peter quit of the payment. ^
The incident shows how far the statesmen of the day were
from realizing the principles of modern political theory.
They had not yet grasped the conception of a Council
endowed with constitutional authority to impose its will on
a dissenting minority. Here it was apparently a minority
of one.2
From this time forward the Common Council was almost
invariably consulted before the Crown attempted to levy
such contributions; and sometimes was bold enough to
make conditions or to decline payment altogether, the first
instance on record of an outright refusal taking place in
a Parliament held at London in January, 1242.^ The
barons, in October, 1255, if Matthew Paris has not fallen
into error, considered that the provisions of chapters 12
and 14 of John's Magna Carta were still in force, although
they had been omitted in the reissues of Henry HI. When
the King asked a liberal aid in furtherance of his scheme
for securing the Crown of Sicily for his son Edmund, those
present at the Council deliberately refused, on the ground
that some of their peers had not been summoned " accord-
ing to the tenor of Magna Carta." ^
^See Pipe Roll Qi^ Henry III., cited Madox, I. 675.
2 For the beginnings of the modern doctrine of the rights of majorities see infra
under c. 61.
3 See Prothero, Simon de Montfort, 67, and authorities there mentioned.
*See M. Paris, Chron. MaJ., V. 520. Note, however, that the version of the
Charter given in his own history contains no such requirement. The barons in
1255 may have had access to the version of 12 15.
256 MAGNA CARTA
CHAPTER FIFTEEN.
Nos non concedemus de cetero alicui quod capiat auxi-
lium de liberis hominibus suis, nisi ad corpus suum
redimendum, et ad faciendum primogenitum filium suum
militem, et ad primogenitam filiam suam semel mari-
tandam, et ad hec non fiat nisi racionabile auxilium.
We will not for the future grant to any one licence to take an
aid from his own free tenants, except to ransom his body, to
make his eldest son a knight, and once to marry his eldest
daughter ; and on each of these occasions there shall be levied
only a reasonable aid.
This chapter confers on the tenants of mesne lords pro-
tection similar to that already conferred on Crown-tenants :
money is no longer to be extorted arbitrarily by their lords. ^
Different machinery, however, had here to be adopted, since
the expedient of chapter 12 (" the common counsel of the
realm ") was inapplicable.
I. Points of difference between tenants-in-chief and
under-tenants. Tenants of mesne lords were in some
respects better off than tenants of the King,^ but in others
their position was worse. Not only had they to satisfy
demands of their own lord for "aids," but part of every
burden laid by the King upon that lord's shoulders was
transferred to theirs. In seeking to protect under-tenants,
Magna Carta looked, not to the common council, but to
the King. No mesne lord could compel his tenants to
contribute to his necessities without written licence from
the Crown ; and the Crown was now forbidden to issue
such licences except upon the usual three occasions.^
^ The chapter is, therefore, on the one hand, a supplement of cc. 12 and 14 ; on
the other, a particular application of the principle enunciated in c. 60, which
extended to sub-tenants benefits secured to Crown-tenants by previous chapters.
2 The exemptions enjoyed by them are explained under c. 43.
^ By strict feudal theory the King had no right to interfere between the barons
and their sub-tenants, (i) The need for royal writs was thus a usurpation.
(2) Those writs were "only letters of request," not binding on sub-tenants. See
Adams, Origin, 230-2.
CHAPTER FIFTEEN 257
Contrast this procedure with that which affected Crown-
tenants : —
(i) While chapter 12 had spoken of "aids and scutages,"
this one speaks of " aids " alone. The omission can be
readily explained : a mesne lord in England had no
admitted right of private war,- and was debarred from
demanding scutage upon his own initiative. He might,
indeed, allocate upon his freeholders part of any scutage
which the Crown had taken from him ; but the barons who
framed the Charter had no intention to renounce so just a
right. The restriction of this clause to " aids " was thus
intentional.
(2) It would have been absurd to require " the common
counsel of the realm " for every aid paid by the freeholders
of a manor. The embryo Parliament had no time for petty
local affairs; and the present chapter makes no such
suggestion. Some substitute had, however, to be found.
A natural expedient would have been to compel the rtiesne
lord, who wished an aid, to take " the common consent "
of the freeholders of his manor, assembled in court baron,
as in a local Parliament. This course was sometimes
followed. Henry Tracey, for example, in 1235 (although
armed with a royal writ), convened his Devonshire knights
and obtained their consent to an aid of 20s. per fee on his
daughter's marriage. ^ No such obligation, however, had
been placed on mesne lords by Magna Carta, which had
sought a practical substitute for " the common counsel of
the realm " in a different direction.
(3) A check upon such exactions was sought, not in the
court baron, but in the need for a royal licence. The
necessity for this may at first have been a practical, rather
than a legal, one ; for executive power lay with the officers
of the Crown alone, and the sheriff gave his services only
at the King's command. ^ The Crown thus exercised what
iBracton's Note-hook^ No. 1146, cited Pollock and Maitland, I. 331.
2 In theory, in Henry II. 's reign at least, a royal writ was not required in the
normal case. See Dialogus, II. viii., and the editors' comment (p. 191) :
'* Normally the levying of money under any pretext from a landowner gave him a
right to make a similar levy on his under-tenants." As regards scutage^ a distinc-
R
258 MAGNA CARTA
was virtually a power of veto over all aids taken by mesne
lords. Such a right, conscientiously used, would have
placed an effectual restraint on their rapacity. John,
however, sold writs to every needy lord who proposed to
enrich himself at his tenants' expense. Magna Carta for-
bade the two tyrants thus to combine against sub-tenants,
enunciating a hard-and-fast rule which, if duly observed,
would have struck at the root of the grievance : no writ
could be lawfully issued except on the three well-known
occasions.
II. The Influence of Magna Carta upon later Practice.
This chapter, along with chapters 12 and 14, was discarded
by Henry III.; and little difference, if any, can be traced
between the practices that prevailed before and after 12 15.
Mesne lords invariably asked the Crown's help to collect
their aids. They could not legally distrain their free-
holders, except through the sheriff, and this was, in part
at least, a result of Magna Carta. ^
Henry III., however, disregarded the rule which forbade
the licensing of extraordinary aids. Like his ancestors,
he was prepared to grant writs on almost any plausible
pretext. From the Patent and Close Rolls, as well as from
other sources, illustrations of the Crown's earlier and later
practice can readily be collected: (i) Scutages. In 1217,
for example, Henry granted permission to all Crown
tenants who had served in person to collect scutage from
their knights.^ /
(2) Ordinary Aids, (a) John in 1204 authorized the
collection of " an effectual aid " from the knights and free-
tion was recognized. The lord who actually paid scutage might collect it from his
sub-tenants without a licence ; but, if he served in person, he could recover none
of his expenses except by royal writ. See idid.j and cf. Madox, I. 675. It is
necessary, however, to avoid confusion between two types of writ, (a) that which
merely authorized contributions, e.£., de scutagio habendo', {b) that which com-
manded the sheriff to give his active help. In later practice, the sheriff often
collected scutage from the sub-tenants and paid it directly to the Crown. Pollock
and Maitland, I. 249-253.
^Cf. Pollock and Maitland, I. 331 : "The clause expunged from the Charter
seems practically to have fixed the law. "
* Close Rolls, I. 306, cited Pollock and Maitland, I. 331.
CHAPTER FIFTEEN 259
holders of the Constable of Chester for the ransom of their
lord.^ (b) A royal writ in 1235 allowed Henry Tracey, as
already mentioned, to take an aid for his daughter's
marriage.
(3) Special Aids, (a) When a fine of sixty marks was
incurred in 1206 by the Abbot of Peterborough, John
allowed him to distrain his under-tenants.^ (b) An heir,
paying relief, might likewise take reasonable contributions
from freeholders.^ (c) The lord's debts were frequently
paid by his tenants. The returns to the Inquest of 11 70
contain particulars of " sums given individually by some
forty burgesses of Castle Rising towards paying off the
mortgages of their lord, the Earl of Arundel, who was
clearly in the hands of the Jews " ^ while in 1234 the Earl
of Oxford and the Prior of Lewes each obtained a letter
patent distraining tenants to contribute to discharge their
debts. ^ Evidence is thus preserved that Henry III. took
full advantage of the omission from his own charters of
this part of his father's promises. He did not question
the justice of such writs, if good fees were paid. His
letters authorized the taking of a " reasonable " aid, without
hinting at any mode of determining what that was. This
is illustrated by the procedure adopted by Henry Tracey in
1235, when he debated with his assembled knights of
Devonshire the amount to be paid as "reasonable," and
finally accepted 20s. per fee.^ This same mesne lord, how-
ever, twelve years later, obtained a writ bidding the sheriff
of Somerset assist him to collect " the scutage of Gascony "
at 40s. per fee.'^
The first Statute of Westminster virtually reverted to the
rule laid down in 12 15, for its terms imply that aids could
only be taken on the three well-known occasions. Only
'^Patent Rolls, 5 John, cited Madox, I. 615.
2 Close Rolls, 7 John, cited Madox, I. 616.
'See Glanvill, IX. 8. *See Round, Commune of London, 130.
5 See Madox, I. 617, citing Patent Rolls, 18 Henry III. Various other
examples are given by Pollock and Maitland, I. 331, e.^. ''the earl of Salisbury,
to enable him to stock his land."
^ Supra, p. 257, and cf. Pollock and Maitland, I. 331.
'See Madox, I. 677.
26o MAGNA CARTA
20S. could be taken from a knight's fee and an equal sum
from land held in socage of the annual value of ;^20. No
aid for a knighthood could be taken before a son was 15
years of age, or for a marriage until a daughter was 7.
CHAPTER SIXTEEN.
NuLLUS distringatur ad faciendum majus servicium de
feodo militis, nee de alio libero tenemento, quam inde
debetur.
No one shall be distrained for performance of greater service for
a knight's fee, or for any other free tenement, than is due therefrom.
For military tenants, the transition from scutage to
service was a natural one. John declared that no free-
holder should be constrained to do more service for his
lands than he was legally bound to do. Disputes might
arise, however, as to what extent of service actually was
due in each particular case, and Magna Carta did nothing
to remove such ambiguities. The difficulties of definition,
indeed, were enormous, since the duration and conditions
of service might vary widely, in consequence of special
exemptions or special burdens which appeared in title
deeds or rested upon immemorial usage. The barons could
not enter on so intricate and laborious a task^
One grievance may have been specially in their minds.
They had frequently objected to serve abroad, particularly
during John's campaigns in Poitou.^ To force them to
serve in the south of France, or to fine them for staying at
home, was, they may well have argued, to distrain them
ad faciendum majus sermcium. de feodo militis quam. inde
dehetur. When they inserted these words in the Charter,
they doubtless regarded them as a prohibition of com-
pulsory service in Poitou, at all events.^ The clause was
^ See the authorities cited supra, p. 68, n. 3, and 69, n. i.
2 In the so-called "unknown Charter of Liberties " (see Appendix) John con-
cedes to his men ** ne cant in exercitu extra Angliam nisi in Normanniam et in
Brittaniam" a not unfair compromise, which may possibly represent the sense in
which the present chapter was interpreted by the barons. See, however, Adams,
Origitiy 232, who takes a different view.
CHAPTER SIXTEEN 261
wide enough, however, to include minor grievances. The
barons did not confine its provisions to miHtary service,
but extended it to other forms of freehold tenure (" nee de
alio libera tenemento "). No freeholder, whether in socage,
serjeanty, or frankalmoin, could in future be compelled to
render services not legally due.
If the barons thought they had thus settled the vexed
questions connected with foreign service, they deceived
themselves. Although this chapter (unlike those dealing
with scutage) remained in all subsequent confirmations,
it was far from preventing disputes. Yet the disputants in
future reigns occupied somewhat different ground. From
the days of William I. to those of Charles II., when the
feudal system was abolished, quarrels frequently arose, the
most famous of which, in 1297, led to Edward's unseemly
wrangle with his hereditary Constable and Marshal, who
refused to embark for Gascony except in attendance on the
King's person.^
It has been shown in the Introduction ^ how the obliga-
tions of a military tenant fell naturally into three groups
(services, incidents, and aids), while a fourth group
(scutages) was added when the Crown commuted military
service for its equivalent in money. Feudal grievances
may be arranged in four corresponding groups, each
redressed by special clauses of Magna Carta : abuse of aids
by chapters 12, 14, and 15 ; of feudal incidents y by chapters
2 to 8; of scutage, by chapters 12 and 14; and of service,
by the present chapter.
CHAPTER SEVENTEEN.
CoMMUNiA placita non sequantur curiam nostram sed
teneantur in aliquo loco certo.
Common pleas shall not follow our court, but shall be held in
some fixed place.
^ Walter of Hemingburgh, II. 121. Cf., on the whole subject of foreign service,
supra, 67-76.
^ Supra, 59-69.
262 MAGNA CARTA
An attempt was here made to render royal justice cheaper
and more accessible. Law-suits in which the Crown had
no special interest, common pleas, were to be held in some
pre-appointed spot, and no longer to follow the King from
place to place. The full extent of this boon will be better
appreciated after a short consideration of the method of
dispensing justice adopted by Henry II. and his sons.
I. The Curia Regis as a Court of Law. The evil com-
plained of was a characteristically medieval one, and arose
from the fact that all departments of government were
centred in the King's household. This Curia Regis,
indeed, united in itself the functions of the modern Cabinet,
of the administrative departments (such as the Home Office,
the Foreign Office, and the Admiralty), and of the various
legal tribunals. It was the parent inter alia of the Court
at St. James's and the courts at Westminster. Nothing
could be done outside of the royal household, and that
household never tarried long in any one spot. Everything
was focussed to one point, but to a point constantly in
motion. Wherever the King went, there the Curia Regis,
with all its departments, went also. The entire machinery
of royal justice followed Henry II., as he passed, some-
times on the impulse of the moment, from one of his
favourite hunting seats to another. Crowds thronged after
him in hot pursuit, since it was difficult to transact business
of moment elsewhere.
This meant intolerable delay, annoyance, and expense.
The case of Richard of Anesty is often cited in illustration.
His own account is a graphic record of his journeyings in
search of justice, throughout a period of five years, during
which he visited in the King's wake most parts of England,
Normandy, Aquitaine, and Anjou. The plaintiff, although
ultimately successful, paid dearly for his legal triumph.
Reduced to borrow from the Jews to meet enormous out-
lays, mostly travelling expenses, he had to discharge his
debts with accumulations of interest at the ruinous rate of
86f per cent.^
II. Common Pleas and Royal Pleas. Long before 12 15,
1 Cf. J. F. Stephen, I/isL of Crim, Law, I. 88-9.
CHAPTER SEVENTEEN 263
litigations conducted before the King's courts had come
to be divided roughly into two classes, according as the
royal interests were or were not specially affected by the
issue. Those on one side of this fluctuating line were
known as royal pleas, or "pleas of the Crown," provisions
for holding which are contained in chapter 24, those on
the other side as ordinary or "common pleas," to which
alone the present chapter refers. As these ordinary suits
did not require to be determined in the royal presence, it
was possible to appoint a bench of judges to sit permanently
in some fixed spot, selected to suit the convenience of
litigants. No town was named in Magna Carta; but
Westminster, even then the natural home of law, was
probably intended from the first. It is Westminster that
Sir Frederick Pollock has in mind when he writes : " We
may also say that Magna Carta gave England a capital." ^
The barons in 12 15, in asking this reform, were not insist-
ing on any startling innovation, but demanding merely
the observance of a rule long recognized. During most of
John's reign, a court did sit at Westminster dispensing
justice, with more or less regularity; and there most
" common pleas " were tried, unless John ordered other-
wise.2 Magna Carta confirmed the understanding that
'' common pleas " should not dance attendance on the King,
though it did not name any one fixed place where they
should be tried.^
III. Influences of Magna Carta on genesis of Courts of
Common Law, The ultimate consequences of this reform
reached further than was foreseen. Intended to remove a
practical grievance, it had important effects on the develop-
ment of the English Constitution. By securing for
common pleas a permanent home, it gave an impetus to
'^Jurisprudence and Ethics, 209. Sometimes, however, another " fixed place "
was substituted. The Court of Common Pleas sat once at York under Edward III.
and once at Hertford under Elizabeth. See Maitland, Select Pleas of the Crown^ xiii.
The statute 2 Edward III. c. ii enacted that it should. not be removed to any
new place without due notice.
■See Prof. Maitland, Select Pleas of the Crown, xiii.-xvi.
3 See Pollock, Expansion of Common Law, 63 n. Cf. Holdsworth, I. 75.
264 MAGNA CARTA
the disintegrating tendencies already at work within the
many-sided household of the King. It helped forward
the cleavage destined to divide completely the future Courts
of Westminster from the Court of St. James's and from
Downing Street. Nor was this all : the special treatment
accorded to " common pleas " emphasized the distinction
between them and royal pleas, and so contributed to the
splitting up of the same Curia Regis, on its judicial side,
into two distinct tribunals. One little group of judges
were set apart for hearing common pleas, and known as
" the King's Judges of the Bench," or more briefly as " the
Bench," and at a later date as the Court of Common Pleas.
A second group, reserved for royal pleas, became the court
Coram Rege, known subsequently as the Court of King's
Bench. There were thus two benches : a common bench
for common pleas and a royal bench for pleas of the Crown. ^
The double process, by which these two small courts
separated slowly from the parent court and from each other,
began long prior to Magna Carta, and was not completed
before the close of the thirteenth century. These benches
were also closely linked with a third bench, known for
centuries as the Court of Exchequer, which was in its
origin merely one department of that government bureau,
the King's financial Exchequer in which money was
weighed and tested and the royal accounts drawn up.
Many disputes or pleas affecting Crown debts had to be
there decided, and a group of officials were set aside to try
these. These men, called "barons of the exchequer,"
formed what was in fact, though not in name, a third bench
or court of justice.
All three of the Courts of Common Law were thus off-
shoots of the King's household. In theory, each of these
ought to have confined itself to a special class of suits —
royal pleas, common pleas, and exchequer pleas respec-
tively; but, by a process known to law-courts in all ages,
each encroached on the jurisdictions and fees appropriate
to the others, until they became, for most purposes, three
sister courts of co-ordinate authority. They were bound
^Cf. supra, 90.
CHAPTER SEVENTEEN 265
to decide all suits according to the technical and inflexible
rules of common law; and their jurisdiction required a
supplement, which was supplied by the genesis of the
Court of Chancery, dispensing, not common law, but
equity, which professed to give (and, for a short time,
actually did give) redress on the merits of each case as it
arose, unrestrained by precedents and legal subtleties.
IV. The Evolution of the Court of Common Pleas, The
comment usually made upon the present chapter is that
we have here the origin of the Court of Common Pleas.
Now, legal institutions do not spring, full-fledged, into
being : the Common Pleas, like its sister Courts of King's
Bench and Exchequer, was the result of a long process of
bifurcation from a common stem.
Three stages may be emphasized, (i) The earliest trace
of a definite bench, set apart for common pleas, is to be
found in 1178. Henry II., returning from Normandy,
found that there had been irregularities. To prevent their
recurrence, he effected changes, the exact nature of which is
matter of dispute. A contemporary writer ^ relates how
Henry chose two clerks and three laymen from the officials
of his own household, and gave to these five men authority
to hear all complaints and to do right " and not to recede
from his court." It was long thought that this marked the
origin of the King's Bench, ^ but Mr. Pike ^ has conclu-
sively proved that the bench thus established was the
predecessor, not of the royal bench, but rather of the bench
for common pleas.
In 1 178, then, these five judges were set apart to hear
ordinary suits ; but they were specially directed not to leave
Henry's court; so that common pleas still " followed the
King," even ordinary litigants in non-royal pleas having
to pursue the King in quest of justice as he passed from
place to place in quest of sport or business.
It must not be supposed that the arrangement thus made
^ Author of Ges^a Regis Henrici^ I. 207.
2 Eigelow, Procedure, 89 ; Stubbs, Gesta Regis Henrici, I. Ixxi.
^ House of Lo7'ds, 32. See also Poole, Exchequer, 180, and Adams, Origin,
136 ff.
/
266 MAGNA CARTA
settled the practice for the whole period of thirty-seven
years preceding the grant of Magna Carta. On the con-
trary, it was merely one of many experiments tried by that
restless reformer, Henry of Anjou ; and the separate bench
then instituted may have been pulled down and set up again
many times. It had probably, at best, a fitful and inter-
mittent existence. There is evidence, however, that some
such court did exist and did try common pleas in the reigns
of Richard and John.^ On the other hand, this tribunal
had in John's reign ceased to follow the King's movements
habitually, and established itself at Westminster.^ It was
in 1 2 15 considered an abuse for John to try a common plea
elsewhere.
(2) Magna Carta, in 12 15, gave authoritative sanction to
this understanding, and thus marks a stage in the evolu-
tion of the Court of Common Pleas. ^ Ordinary pleas were
no longer to follow the King.^ Young Henry renewed
this promise, and his minority favoured its strict observ-
ance : a mere boy could not make progresses through the
land dispensing justice as he went. Accordingly, all pleas
continued for some twenty years to be heard at West-
minster. The same circumstance may have temporarily
arrested the process of cleavage between the two benches.
(3) About 1234, Henry began to follow the precedent,
set by his ancestors, of moving through his realm with
judges in his train. While one group went with him,
another remained at Westminster : some method of allocat-
ing business had therefore to be found. Common pleas,
in accordance with Magna Carta, remained stationary;
while pleas of the Crown went on their travels. The split
between the two benches now became absolute : from the
^See Prof. Maitland, Sel. PL Crown, xiii.-xvi. ; see also in FiJ)e Roll, 7 John
(cited Madox, I. 791) how money was paid that a plea pending before the
lusticiarii de banco might be heard coram rege. This entry proves the existence
in 1205 of the de banco as distinct from the coram rege.
2 See Maitland, ibid.
3 Cf. Poole, Exchequer, 183, who insists, however, that " it said nothing about
a distinct court."
* For attempts to evade this prohibition on the ground of the special character of
particular pleas, see Bracton's Note-book, Nos. 12 13 and 1220.
CHAPTER SEVENTEEN 267
year 1234, two continuous series of distinct rolls can be
traced, known respectively as rotuli placitorum coram rege
and rotuli placitorum de banco. If any date in the history
of one law court, which is in process of becoming two, can
be reckoned as marking the point of separation, it should
be that at which separate rolls appear. The court's memory
lies in its records, which are thus closely associated with
its identity. The common bench and the royal bench had
become distinct.^ While Henry and his justices sat in
judgment at Worcester, in 1238, a litigant protested against
his suit being tried before them. It was a " common plea "
and therefore, he argued, ought not to follow the King, in
violation of Magna Carta. At Westminster only, not at
Worcester or elsewhere, could his case be heard. ^
With royal pleas it was different : for long they con-
tinued to follow the King's person without any protest
being raised; and the Court of King's Bench did not
finally settle at Westminster for nearly a century after the
Court of Common Pleas had been established there. It is
doubtful whether, even in 1258, a separate royal bench had
been constituted.^ So late as 1300, Edward I. ordained,
by the Articuli super cartas, that " the Justices of his
Bench " (as well as his Chancellor) should follow him, so
that he might have at all times near him " some sages of
the law, which be able duly to order all such matters as
shall come into the Court at all times when need shall
require." * The matters here referred to were royal pleas :
common pleas were tried at Westminster.
V. Common Pleas and the Exchequer. Records speak
of the curia regis meeting for legal business ad scaccarium
(that is, in the room where the business of the Exchequer
of Accounts was normally transacted) long before the
genesis of a separate Court of Exchequer.^
Formal sessions of the Exchequer for auditing the
^ See Maitland, Se/. PI, Crown, xviii.
2 See Placitorum Abbreviatio (p. 105), 21 Henry III., cited Pike, House of
Lords, p. 41 Cf. also Bracton's Note-book, pleas Nos. 1213 and 1220.
2 Poole, Exchequer, 183. ^ 28 Edward I. c. 5.
5 For stages in this genesis in 1234, 1236, and 13 17, see Poole, Exchequer, 183.
268 MAGNA CARTA
Sheriffs' accounts could only be held at Westminster, where
the necessary apparatus was kept; but "the Exchequer,"
using that elastic word in a somewhat different sense, with
much of its impedimenta of writs and tallies, would
accompany the King on his progresses through the realm.
In 1 2 10, for example, the Exchequer was at Northampton;
in 1266, at St. Paul's; in 1277, at Shrewsbury; and in
1299, at York.^
Now, the Exchequer, when it sat as a Court of law,
was ever willing — for a consideration — to place its potent
procedure, devised for the King's use, at the disposal of
private creditors, treating " common pleas " as " exchequer
pleas." Ordinary debtors, summoned to answer for their
debts before the harones scaccarii were subjected to more
rapid pressure than they would have experienced elsewhere.
Debtors were thus as anxious to escape the jurisdiction of
the Exchequer, as creditors were to invite it.
Both before and after Magna Carta, it would appear that
common pleas were sometimes tried at sessions of the
Exchequer, held not only at Westminster but also during
its wanderings in the King's train. It was natural enough
that defendants who fo^nd themselves hustled by the
stringent Exchequer protess should seek shelter under the
present chapter of the Great Charter. That they did so
is proved by the words of the Articuli super Cartas of 1300,
which declared that no common pleas should henceforth be
held in the Exchequer " contrary to the form of the Great
Charter." 2
The implication of this clause of the statute of 1300 has
sometimes been accepted literally .^ Magna Carta, how-
iStubbs, Cons^. Hist., 11. 281 n.
2 See 28 Edward I. c. 4. Many previous attempts had been made to keep
common pleas out of the Exchequer, e.g. the writs of 56 Henry III. and 5
Edward I. (cited Madox, II. 73-4), and the so-called statute of Rhuddlan (12
Edward I.), see Statutes of Realm, I, 70.
3 Thus Madox (II. 73-4) holds that c. 17 relates to the Exchequer ; so does Mr.
Bigelow {^History of Procedurcy 130-1), who explains the grievance as a difficulty
of getting speedy justice at the Exchequer, because the barons refused to sit after
their fiscal business had been finished. This seems to be an error : the Barons of
Exchequer made no difficulty about hearing pleas : quite the contrary. Plaintiffs
CHAPTER SEVENTEEN 269
ever, in set terms at least, contains no such prohibition.
If the present chapter excludes common pleas from the
jurisdiction of a travelling Exchequer equally as from that
of a travelling King's bench, its words cannot be so
stretched as to apply to normal sessions of the Exchequer
held at Westminster. The Articuli super Cartas, how-
ever, attempted what the Charter of 12 15 did not. After
1300 it was clearly illegal to hold any pleas in the Exche-
quer, unless such as affected the Crown and its ministers.
Subsequent statutes confirmed this; but their plain inten-
tion was always defeated by the ingenious use of legal
fictions and the connivance of the barons of Exchequer,
who welcomed the increase of fees that kept pace with the
increase of business.^
CHAPTER EIGHTEEN.
Recogniciones de nova dissaisina, de morte antecessoris,
et de ultima presentacione, non capiantur nisi in suis
comitatibus et hoc modo ; nos, vel si extra regnum fuerimus,
capitalis justiciarius noster, mittemus duos justiciarios per
unumquemque comitatum per quatuor vices in anno, qui,
cum quatuor militibus cujuslibet comitatus electis per
comitatum, capiant in comitatu et in die et loco comitatus
assisas predictas.
Inquests of noi'el disseisin, of mor^ d''a7icestor, and of darrein
presentmeftt, shall not be held elsewhere than in their own
were equally eager to purchase the writs which they were keen to sell : it was
only defendants (debtors) who objected to the rapid and stringent procedure for
enforcing payment adopted by this efficient court. The sheriffs and others waiting
to render accounts before the Exchequer also protested against the congestion of
business produced at the Exchequer by the eagerness of litigants who pressed
there for justice. See Madox, II. 73. Plaintiffs had no reason to complain.
^The fiction of "Crown debtors" is well known : plaintiffs obtamed a hearing
in the exchequer for their common pleas by alleging that they wished to recover
debts due to them " in order to enable them to answer the debts they owed to the
king." See Madox, II. 192.
270 MAGNA CARTA
county-courts,^ and that in manner following, — We, or, if we
should be out of the realm, our chief justiciar, will send two
justiciars through every county four times a year, who shall,
along with four knights of the county chosen by the county,
hold the said assizes in the county court, on the day and in the
place of meeting of that court.
Provision is here made for holding before the King's
travelling justices, frequently and in a convenient manner,
three species of judicial inquests known as " petty assizes."
These are of exceptional interest from their connection with
the genesis of trial by jury and the Justices of Assize.
I. The Curia Regis and the travelling Justices. From
an early date, certainly from the accession of Henry I.,
it was the Crown's practice to supplement the labours of
its officials at the royal exchequer by the occasional despatch
of chosen individuals to inspect the provinces, collecting
information and revenue, and, incidentally, hearing law-
suits. Justice was thus dispensed in the King's name by
his delegates in every shire of England, and a distinction
arose between two types of royal courts : (i) the King's
Council and its offshoots (including the three courts of
common law and the court of chancery), which at first
followed the King's person, but gradually, as already
shown, 2 found a settled home at Westminster ; and (2) the
courts of the itinerant justices which exercised such dele-
gated authority as the Crown chose from time to time to
entrust to them. The sphere of labour of these commis-
sioners, as they passed from district to district, was the
court of each shire, convened to meet them. They formed,
in this way, a link between the old local popular courts and
the system of royal justice. These travelling justices were
of two types, Justices in Eyre and Justices of Assize respec-
tively.
(a) The Justices in Eyre were the earliest form of travel-
ling judges, though their original duties were rather
financial and administrative, than strictly judicial. Their
^ " Coinitatus^^ indicates both the county where the lands lay and the court of
that county. It was originally the sphere of influence of a comes or earl. Cf.
supruy c. 2.
2 See supra, c. 17.
CHAPTER EIGHTEEN 271
history extends from Henry I. to the end of the fourteenth
century. 1 Their outstanding characteristics were the sweep-
ing nature of their commissions (ad omnia placita)^ the
harsh and drastic way in which they used their authority,
and their intense unpopularity. Their advent was dreaded
like a pestilence : each district visited was left impoverished
by fines and penalties. On one occasion, the men of Corn-
wall " from fear of their coming, fled to the woods." ^ An
eyre was only resorted to at long intervals — seven years
came to be the recognized term — and was a method of
punishing delinquencies and miscarriages of justice and of
collecting royal dues. It was not a visit from these hated
Justices of Eyre that the barons in 12 15 desired to have four
times a year.
(h) The Justices of Assize also were travelling judges, but
in their original form at least, possessed hardly another
feature in common with the Justices in Eyre. Their
history extends from a period not earlier than the reign of
Henry II. down to the present day.^ They seem to have
been popular from the first, as they used a speedy and
rational procedure; while the scope of their jurisdiction,
although extended as their popularity increased, was
limited by the terms of their commissions. They were
regarded, not as royal tax-gatherers armed with harsh
powers of coercion, but as welcome bearers of justice to
the doors of those who needed it.
At first their duties were confined to enquiries of the kind
mentioned in the text, known as " assizes " ; and the new
species of travelling judges were hence called " Justices of
Assize," a name that has clung to them for centuries,
although their jurisdiction has been gradually increased
till it now includes both civil and criminal pleas of every
description, and although meanwhile the invention of new
forms of process has superseded the old " assizes, " and at
^ See W. S. Holdsworth {History of English Law, p. 115), who cites 1397 as
the date of the final abolition of Eyres.
2 This was in 1233 : see Pollock and Maitland, I. 181.
^Blackstone, Commentaries, III. 58, assigns 1176 (the assize of Northampton),
as the date of their institution.
272 MAGNA CARTA
last necessitated their total abolition. ^ They are still
" justices of assize " in an age which knows nothing of the
old assizes.
II. Nature and Origin of the Petty Assizes. The institu-
tion of the " assizes " — particular forms of the sworn inquest
— occupied a prominent place among the expedients by
which Henry II. hoped to substitute a more rational pro-
cedure for the form of proof known as trial by combat.^
The duellum, introduced at the Norman Conquest,
remained for a century the chief method in use among the
upper classes for determining serious litigations. Gradu-
ally, however, it was confined to two groups of pleas, one
civil and the other criminal : appeals of treason and felony
on the one hand, and suits to determine title to land on the
other. ^ The process of restriction was carried further by
Henry II., who provided for the defendant or accused party,
wherever possible, an option to trial by battle. Under
chapter 36 will be explained the expedient adopted for
evading combat in criminal cases. The present chapter
relates to certain important groups of civil pleas,* namely,
the three Petty Assizes, the frequent use of which was now
insisted on, although the Grand Assize was still viewed
askance, for reasons to be explained in connection with
chapter 34.
^See statute 3 and 4 William IV. c. 27, §§36-7. The last actual case of a
Grand Assize occurred in Davies v. Lotmdes, in 1835 ^^^ ^^3^ (i Bing. N.C. 597,
and 5 Bing. N.C. 161).
2 The name "Assize" is sometimes a source of confusion, because of its various
meanings, (i) Originally, it denoted a session or meeting of any sort. (2) It
came to be reserved for sessions of the King's Council. (3) It was applied to any
Ordinance enacted in such a session, e.g. Assize of Clarendon. (4) It was extended
to every institution or procedure established by royal ordinance, but (5) more par-
ticularly applied to the procedures known as Grand Assize and Petty Assizes.
(6) Finally, it denotes at the present day a "session" of these Justices of Assize,
thus combining something of its earliest meaning with something of its latest. In
certain contexts, it has other meanings still, e.g. (7) an assessment or financial
burden imposed at a *' session."
^ See Neilson, Trial by Combat, 33-6, and authorities there cited.
^ Cf, supra, p. 85, for the place of " combat " in legal procedure ; and
p. 89, for Henry's policy in discouraging it. For the later history of trial by
battle, see infra^ under c. 36.
CHAPTER EIGHTEEN 273
(i) The Grand Assise is not mentioned in Magna Carta;
but some acquaintance with it is necessary to an apprecia-
tion of the Petty Assizes. In the troubled reign of Stephen,
lands changed hands frequently : there was hardly an
important estate in England to which, at Henry's accession,
two or more rival magnates did not lay claim. Constant
litigations resulted, and the only legal method of deciding
the issue was the duellum,
Henry II. introduced a startling innovation. The actual
holder of a property de facto, when challenged to combat by
a rival claimant, was allowed an option : he might force
the claimant (if the latter persisted) to refer the matter to
the oath of twelve knights of the neighbourhood. Henry's
ordinance provided for the appointment of these recognitors.
Four leading knights of the county were first to be chosen,
on whom was placed the duty of selecting twelve knights
of the particular district where the lands lay, and these, with
all due solemnity and in presence of the King's justiciars,
declared upon oath to which suitor the lands belonged.
/Their decision was final, and determined the question of
( ownership for all time.^ The name Grand Assize was
j applied alike to the procedure and to the knights who gave
I the verdict.
The procedure was slow; many formalities and possi-
bilities of delay intervened, involving expensive journeys to
the central Curia, first by the four appointing knights and
afterwards by the twelve appointed. Months and even
years might elapse before the final verdict was obtained.
To lighten these hardships in comparatively unimportant
cases, the Ca^ihfZa of .1194 authorized Justices of Eyre to
hold Grand Assizes where the lands did not exceed ;^5 in
annual value.^
Normally, however, this procedure was for the King's
central Curia, neither for county court nor yet for baronial
jurisdictions. For one thing, only magnates with wide
1 See Glanvill, II. 7.
'^Se/. Chart. ^ 259. The Assize of Northampton in 11 76 {ibid. 152) had given
them jurisdiction over estates of half a knight's fee or less, but nothing was there
said of the mode of proof.
S
274 MAGNA CARTA
demesnes were likely to command the attendance of twelve
knights (or even of tv^elve freeholders) from their own terri-
tories. In combination with the rule given by Glanvill,^
that no plea concerning title to land could be commenced in
any court without royal writ, and with the use made by the
King of the writ praecipe,^ the Grand Assize, while super-
seding trial by battle, became also an expedient for cur-
tailing the jurisdiction of mesne lords. It is easy to under-
stand why (unlike the petty assizes) it never became popular
with the magnates.
Valuable boon as was the option to substitute the verdict
of twelve knights for the duellum in questions of title to
land, the reform had one obvious weak point : the option
conferred might sometimes be usurped by the wrong man,
if a turbulent claimant took the law into his own hands,
evicted the holder by the rude method of self-help, and
thereafter claimed the protection of Henry's ordinance. In
such a case the man of violence — the holder mala fide —
would enjoy the option intended for his innocent victim.
(2) The petty assises may, perhaps, have been the out-
come of Henry's determination to prevent misuse of his new
engine of justice.^ If a demandant alleged that the present
possessor had usurped his place by violence, the King
allowed the preliminary plea thus raised to be summarily
decided by the oath of twelve local landowners, according
to a procedure known as a petty assize. These petty assizesj
of which three are here mentioned, related to questions of
" possession," as opposed to " ownership."
(a) Novel disseisin. The word "seisin," originally*
synonymous with " possession " in general, was gradually
restricted by medieval lawyers to the possession of real
estate. " Disseisin " thus meant the interruption of seisin
1 Glanvill, XII. 25. ^See infra, under c. 34.
3 In the matter of actual date, the received opinion is that the " novel disseisin "
procedure dates from 1166, and the Grand Assize came later. Round {A^/tenaeum
for 28th Jany., 1899) suggests 1179. The evolution of the various writs was,
however, a slow process, and steps in the chain are wanting. Under Geoffrey
Plantagenet in Normandy various writs shade off into one another. See Haskins,
Amer. Hist. Rev., VIII. 613 ff. In any view, the logical sequence seems to be
that given in the text.
CHAPTER EIGHTEEN 275
(or possession) of land ; and was the technical term applied
to violent acts of eviction. " Novel " implies that such ejec-
tion was of recent date; for a summary remedy could be
given only where there had not been undue delay in apply-
ing for it.^
The first of the petty assizes, then, was a rapid and
peaceable method of ascertaining, by reference to sworn local
testimony, whether an alleged recent eviction had really
taken place or not. Without any of the law's delays, with-
out any expensive journeys to the King's Court or to West-
minster, but quickly and in the district where the lands lay,
twelve local gentlemen determined upon oath all allegations
of this nature, li the recognitors of the petty assize
answered " Yes," then the evicted man would have " seisin "
immediately restored to him, and along with " seisin " went
the valued option of determining what proof should decide
the " ownership, " — whether it should be battle or the Grand
Assize. An ordinance instituting this most famous of the
three petty assizes was issued probably in 1 166, a year fertile
in legal expedients.
(b) Mort d'ancestor. The protection afforded to the
victim of " disseisin " did not remove all possibility of justice
miscarrying ; interested parties, other than the man ejected,
were unprotected. An heir might be deprived of his tene-
ment by his lord or by some rival claimant before he had an
opportunity to take possession ; never having been " in
seisin," he could not plead that he had been disseised. For
the benefit of such an heir, a second petty assize, known as
" mort d'ancestor," was invented.^ This is mentioned in
article 4 of the Assize of Northampton, issued in 1 176, where
procedure, essentially similar to, though not quite so speedy
as that already described, was put at the heir's disposal. If
successful, he took the lands temporarily, subject to all
defects in his ancestor's title, leaving as before the question
^In Normandy the corresponding period was "since the last harvest." See
Maitland, Eqtnty, 323.
2 At so late a date as 1267 it was found necessary to recognize by statute the
right of the heir, who had come of age, to oust his guardian from his lands by an
assize of mort d'ancestor. See Statute of Marlborough, c. 16.
276 MAGNA CARTA
of absolute ownership to be determined (if challenged) by
the more cumbrous machinery of the Grand Assize.
(c) Darrein presentment, Advowson (or the right of
appointing to a vacant church) was then, as now, a species
of real estate. Such patronage was highly prized, affording
a living for a younger son or needy relative ; or it might be
converted into cash. Disputes often arose as to possession
and ownership of advowsons. Any one who claimed the
absolute property, as against the holder, must offer battle, as
in the case of any other form of real estate; and the Charter
says nothing on this head.^
The less vital question of possession was more rapidly
determined : if a benefice fell vacant, and two proprietors
claimed the patronage, the Church could not remain without
a shepherd until the question was decided.^ No ; the man
in possession was allowed to make the appointment. But
who was the man in possession ? Clearly he who had (or
whose father had) presented a nominee to the living when
the last vacancy occurred. Here, however, there might be
a dispute as to facts. Twelve local men decided which
claimant had made the last appointment (the " darrein pre-
sentment ") ; and the claimant thus preferred filled up
vacancies, until ousted by battle or the Grand Assize.
All three forms of petty assize were merely new applica-
tions of the royal procedure known in England, since the
Norman Conquest, as inquisitio or recognition
III. Aims of Magna Carta. If the petty assizes were
objects of suspicion when first invented by Henry II., public
opinion, half a century later, had vindicated their wisdom.
The insurgent barons in 12 15 were far from demanding
their abolition ; their new grievance was rather that sessions
iSuch was the law as late as 1285. Westminster II. c. 5 explains that, when
any one had wrongfully presented to a vacant church, the real patron could not
recover his advowson except by writ of right ** gttod habet terminari per duellurn
vel per magna}?t assisam.''''
2 A Lateran Council in 1179 authorized the diocesan bishop to appoint after
three months' vacancy. Hence there was additional need of haste.
3 The relations of the assizes to the ancient inquisitio and to the modern jury are
discussed supra, pp. 134-8.
CHAPTER EIGHTEEN 277
of assize were not held often enough. In prescribing the
way in which these assizes must be held, several points were
emphasized : — (i) No inquiry of the kind was to be held
elsewhere than in the county where the property was situ-
ated. This was intended to meet the convenience of
litigants, of those who served on assizes, and of all con-
cerned.^ Within two years it was seen that this provision
went too far. It was more convenient to hold certain
inquiries before the Bench at Westminster, and the reissue
of 1 2 17 made two modifications : (a) Assizes of darrein pre-
sentment were thereafter to be taken before " the Justices of
the Bench " ; (b) any novel disseisin or mort d'ancestor,
revealing points of special difficulty, might also be reserved
for the decision of the Bench. An element of uncertainty
was thus introduced, of which the Crown took advantage.
In a reported case of the year 1221, it was decided that an
assize of mort d'ancestor should be held in its own county,
not at Westminster.^
(2) John's Charter further insists on quarterly circuits
of Justices of Assize; so that litigants in every county
of England might have four opportunities each year of
having their disputes thus settled. Such frequency
involved expense and labour out of proportion to the good
effected. The Charter of 12 17, accordingly, provided that
circuits should be made only once a year. In 1285, how-
ever, it was enacted that they might be held three times a
year, but not oftener.^
(3) The Charter regulates the composition of the tribunal.
Two justices appointed by the King (or by his chief jus-
ticiar) are directed to hold the assizes, along with four
knights of the shire. The bench of six thus combines
^Thus two successive chapters of Magna Carta emphasize two divergent
tendencies: c. 17 had demanded that "common pleas" should all be held at
Westminster, while c. 18 demands that "assizes" should no^ be taken there. In
both cases, the object was to consult the convenience of litigants.
2 See Bracton's Note-book, No. 1478 ; cited Coke {^Second Institute, proem). If
this assize had presented points of special difficulty it might have been held at
Westminster without violating Magna Carta, as amended in 12 17.
' 13 Edward I. c. 30. Stephen, History of Criminal Law, 105-7, gives further
details.
278 MAGNA CARTA
representatives of the Curia with local landowners. No
mention is made of the twelve recognitors : nor was this
necessary, as their functions and status were well known in
1 2 15, and their verdict formed the essential feature of the
procedure.^ Chapter 19 provides that the classes, from
whom recognitors had to be selected, should attend in
sufficient numbers " for the efficient making of judgments."
(4) The four knights were to be " elected " by the county
court (quatuor Tniliiihus . . . electis per comitatum),^ and
emphasis has been laid on this provision by historians
searching for ancient prototypes of modern institutions.
These knights have been incautiously welcomed as county
magistrates elected on a more or less extended suffrage.^
As the provisions of the reissue of 12 17 are more care-
lessly expressed, and as in particular they contain no word
implying " election," it is sometimes assumed that a change
was intended; that a step tentatively taken towards repre-
sentative local government in 12 15 was deliberately retraced
two years later."* " Electus, " however, in medieval Latin
was a vague word, differing widely from the ideas usually
associated with a modern " election," and applied indis-
criminately to all methods of appointment or selection, even
to the proceedings of officers engaged by Edward I. to com-
pel the impressment of soldiers. The twelve knights were
to be " appointed," not " elected," in the county court; and
it remains doubtful whether the sheriff, the magnates, or
the body of the suitors, would have the chief share in the
appointment. No evidence is forthcoming that any
importance was attached in 121 7 to the word " electus," and
its omission may have been due to inadvertence.
^ Cf. Assize of Northampton, c. 4.
2 Cf. infruy c. 48, where twelve sworn knights are to be chosen per probes
homines ejusdem comitatus. Cf. also Forma Procedendi of 1194 i^Sel. Charters y
255)-
^See, (?.,^. '&\.\x\i\i%, preface to R. Hoveden, IV. xcviii.; Blackstone, Great Charter,
XXX vi.; Medley, Eng. Const. Hist.^ 130.
* Blackstone, ibid.^ points out these changes in the charter of 1217 : "the
leaving indefinite the number of the knights and the justices of assize, the
abolishing of the election of the former, and the reducing the times of taking
assizes to once in every year."
CHAPTER EIGHTEEN 279
IV. Effects of Magna Carta, The stipulations of the
Great Charter were not strictly followed in practice. It was
not the custom under Henry III. for the Crown to grant
general commissions to hold petty assizes. On the con-
trary, each litigant was left to make separate application to
the King, who would then assign a justice by letters patent
to preside over that one particular plea. Hundreds of such
commissions might be issued in one year, and recognitors
were separately summoned for each one of these. In 1258
the Petition of the Barons (c. 19) complained of this, and
an attempt was made at organization. The Statute of
Westminster II. (c. 30) ordained that two sworn justices
should be assigned, before whom and none others assizes
of Novel Disseisin and Mort d'ancestor (along with attaints)
should be taken. They were to go on circuit three times a
year, and to associate wMth themselves one or more of the
discreetest knights of each county — instructions which fall
short of the stipulations of Magna Carta, ^
V. An Erroneous View. Hallam, commenting on this
chapter, seems to have misapprehended the issues at stake.
" This clause stood opposed on the one hand to the encroach-
ments of the King's court, which might otherwise, by
drawing pleas of land to itself, have defeated the suitor's
right to a jury from the vicinage : and, on the other, to those
of the feudal aristocracy, who hated any interference of the
Crown to chastise their violations of law, or control their
own jurisdiction." ^ Hallam thus interprets the chapter as
denoting a triumph of the old local popular courts over both
the King's courts and the courts of the feudal magnates.
It denoted no such thing, but marked in reality a triumph
(so far as it went) of the King's courts over the tribunals of
the feudal magnates — courts baron, as they were afterwards
called. The assizes, it is true, were to be taken in the
county court, but they were to be taken there by the
King's justices. The county courts by this time had fallen
completely under the King's domination, and were to all
^ On the whole subject, see an admirable article by G. J. Turner, EncycL Lazvs of
Engl., III. 76 ff.
2 See Middle Ages, II. 464.
28o MAGNA CARTA
intents and purposes royal courts. The present chapter is
thus conclusive evidence of the triumph of the King's jus-
tice, which was the best article in the market, and, in spite
of all defects, deserved the popularity it had won.
VI. Later History of the Justices of Assize. Whatever
may have been the exact date when there first went on tour
throughout England travelling judges entitled to the
description of "Justices of Assize," such circuits, once
instituted, continued to be held at more or less regular
intervals from the beginning of the thirteenth century to the
present day. Their jurisdiction steadily widened under
successive kings, from Henry II. to Edward III. ; and they
gradually superseded the older Justices of Eyre, taking over
such functions as were not inconsistent with the change
from the medieval to the modern system of justice.^
For centuries it was customary for the Crown to issue to
the justices of each new circuit several commissions, each
conferring jurisdiction over a different class of pleas.
Founding on the authority of Sir Francis Bacon, historians
have been wont to enumerate five distinct commissions.^
(i) The commission of assize, already discussed, allowed
them to hold petty assizes, but not (in the normal case) the
grand assize.^
(2) Commission of Nisi Prius. Under Statute West-
minster II. c. 30, the sheriff was directed to summon jurors
to Westminster " unless at an earlier date " (nisi prius) the
justices of assize should happen to arrive in the county in
question. This was interpreted as creating a jurisdiction in
the justices of assize to try all non-criminal pleas of the
county — a jurisdiction afterwards known as "nisi prius."^
Thereafter, any such plea, whether begun in King's Bench
or Common Pleas, might be determined locally in its appro-
* Cf. Coke, Ftrsf Institute, 293b : " As the power of justices of assizes by many
Acts of Parliament and other commissions increased, so these justices itinerant by
little and little vanished away."
2 On whole subject see Stubbs, Sel. Chart., 141 -3 ; Stephen, Hist. Crim.
Law, I. 79-111 ; Holdsworth, I. 116-123. Contrast, however. Turner, op. cit., III.
76flF.
' For the exception where lands were under ;^5 in annual value, see supra,.
P- 273-
CHAPTER EIGHTEEN 281
priate shire as well as at Westminster. According to the
opinion generally received, a separate commission of " 7iisi
prius " was issued to each group of justices of assize, but
it has recently been urged that no separate commission was
required, the one jurisdiction being merely incidental to the
other. ^
(3) The commission of gaol delivery was, subsequently to
1299, conferred on the justices of assize, in accordance with
a statute of that year, 2 authorizing them to inspect all gaols
and enquire into all charges against prisoners, and to set
free those unjustly detained. Previously, similar powers had
been spasmodically conferred on separate commissioners,
who had too often abused their authority.
(4) Commissions of Oyer and Terminer, issued spasmodi-
cally from as early a date as 1285,^ to more or less respon-
sible individuals, were from 1329 onwards conferred
exclusively on the justices of assize, who thus obtained
authority^ "to hear and determine" all criminal pleas
pending in the counties they visited. This, combined with
the commission of gaol delivery, amounted to a full juris-
diction over crimes and criminals of every kind and degree ;
just as the commission of assize (with or without an added
commission of nisi prius) conferred full jurisdiction over
civil pleas. ^
(5) In the generally received opinion, a fifth commission
was invariably issued to the justices, in the form of a special
commission of the peace, from the reign of Edward III.
onwards.^
^G. J. Turner, ibid., p. 79.
8 27 Edward I. c. 3. For early history of gaol delivery, see Pollock and Mait-
land, II. 642.
'13 Edward I. c. 39; see Stephen, Hist. Criminal Law, p. 106.
* Edward III. c. 2. Ibid., no.
5 It is unnecessary to do more than notice the exceptional "commissions of
trailbaston," supposed to date from the Statute of Rageman (1276), conferring
special powers for the suppression of powerful wrongdoers. These were soon
superseded by the commissions of oyer and terminer.
6 Mr. Turner {ibid., p. 79) suggests, however, that a separate commission was
not needed, as "all justices of assize and gaol delivery were in the commission of
the peace within the precincts of the court." In his view the justices received
three distinct commissions, not five.
282 MAGNA CARTA
The justices of assize, from the small beginnings referred
to in John's Great Charter, thus gathered to themselves the
powers exercised originally by various sets of commis-
sioners. They have continued for many centuries to per-
form the functions conferred by these various commissions,
and form a characteristic part of the judicial system of
England.
CHAPTER NINETEEN.
Et si in die comitatus assise predicte capi non possint, tot
milites et libere tenentes remaneant de illis qui interfuerint
comitatui die illo, per quos possint judicia sufficienter fieri,
secundum quod negocium fuerit majus vel minus.
And if any of the said assizes cannot be taken on the
day of the county court, let there remain of the knights and
freeholders, who were present at the county court on that day, as
many as may be required for the efficient making of judgments,
according as the business be more or less.
This supplement to the preceding chapter prescribed the
course to be followed when press of business prevented some
of the assizes on the agenda from being disposed of on the
court day. The shiremoot lasted for one day only, and to
hold an adjourned session of all the suitors would inflict
hardship on those whose presence was required elsewhere.
The framers of the charter here sought to provide for the
presence of a sufficient supply of recognitors, without
insisting on the continued attendance of the whole body of
suitors. They were doing their best to give effect to two
requirements of the Articuli Baronum not readily recon-
cilable, namely, that only those actually required as recog-
nitors should be summoned (article 8) ; and that assizes
should be " shortened " (article 13), implying the presence
of sufficient recognitors for a rapid despatch of business.
The terms of Magna Carta made it clear that assizes in
the normal case should be held in the county court — a point
upon which the Articles had been silent. This was a
salutary provision, since a healthy publicity accompanied
CHAPTER NINETEEN 283
the proceedings of the shiremoot. If there was more busi-
ness than could be got through in one day, a compromise
must be made between the claims of litigants wishing their
pleas hastened and the desire of other people to be dis-
charged from further attendance. The justices were
directed to complete their labours on the morrow, but were
forbidden to retain anyone in attendance except the actual
parties to suits and a sufficient number of jurors. Those
whom Magna Carta thus compelled to wait a second day
were exactly those whose presence the Articles had required
upon the first day. The discrepancy between the two
documents might be explained on the supposition that the
device of synchronizing the visit of the justices with the
date of holding the monthly shiremoot was only thought of
after the Articles of the Barons had been sealed.^
The Charter of 1217 made a different provision for the
same contingency. Unfinished assizes need no longer be
taken in their own county on the day following the county
court, nor, indeed, on any other day. The judges received
full authority to bring them to a conclusion elsewhere on
their circuit according as it might suit their convenience.
This concession to the justices, taken in connection with the
further provisions of 12 17, reserving all darrein present-
ments, together with other assizes of any difficulty, for the
decision of the bench, shows a comparative disregard of the
convenience of jurors, who might, in the option of
the justices, find themselves compelled either to follow the
assizes from shire to shire, or else to undertake the irksome
journey to Westminster, from which the Charter of 1215
had relieved them.^
■ Subsequent practice did not conform to this rule. One novel disseisin, or one
mort d' ancestor, might be held by itself; and complaint was made in 1258 that
the sheriffs proclaimed in the market places that all knights and freeholders must
assemble for such an inquest, and when they came not, amerced them at will {pro
voluniate sua). See Petition of Barons, c. 19 {^Sel. Charters, 385).
2 Subsequent legislation vacillated between two policies, actuated at times by a
desire to restrain the discretionary powers of the justices ; and at others by experi-
ence of the hardships inflicted upon litigants by inflexible rules. The Statute of
Westminster II. (13 Edward I. c. 30) confirmed the power of the justices to
reserve cases of mort d'anceator for decision by the bench, and per contra allowed
284 MAGNA CARTA
CHAPTER TWENTY.
Liber homo non amercietur pro parvo delicto, nisi secun-
dum modum delicti ; et pro magno delicto amercietur secun-
dum magnitudinem delicti, salvo contenemento suo; et
mercator eodem modo, salva mercandisa sua; et villanus
eodem modo amercietur salvo waynagio suo, si inciderint
in misericordiam nostram ; et nulla predictarum miseri-
cordiarum ponatur, nisi per sacramentum proborum homi-
num de visneto.
A freeman shall not be amerced for a slight offence, except in
accordance with the degree of the offence ; and for a grave
offence he shall be amerced in accordance with the gravity of the
offence, yet saving always his " contenement " ; and a merchant
in the same way, saving his " merchandise " ; and a villein shall
be amerced in the same way, saving his "wainage" — if they
have fallen into our mercy : and none of the aforesaid amerce-
ments shall be imposed except by the oath of honest men of the
neighbourhood.
This is the first of three chapters that seek to remedy
abuses connected with royal amercements. To understand
what these were requires some knowledge, not only of the
system of legal procedure of which they formed part, but
also of previous systems.
I. Three stages of criminal law. The efforts made in
medieval England to devise machinery for suppressing
crime took various forms. Three periods may be dis-
tinguished.
(i) The hloodfeud. The earliest method of redressing
wrongs was retaliation, or the bloodfeud. The injured
man, or his heir, took the law into his own hands and
exacted satisfaction by the aid of battle-axe or spear.
assizes of darrein presentment to be taken "in their own counties." 6 Richard II.
c. 5 curtailed the discretionary powers, directing that justices assigned to take
assizes and deliver gaols should hold sessions in the county towns in which the
shire courts were wont to be held. 1 1 Richard II. c. 1 1 once more relaxed this
rule, alleging that it had resulted in the inconvenience of suitors. Authority was
given to the chancellor, with the advice of the justices, to determine in what places
assizes might be held.
CHAPTER TWENTY 285
(2) Fixed money 'payments. At some early, but uncer-
tain, date it became customary to accept money in lieu of
vengeance. The new practice, at first exceptional, was
gradually extended. It was made compulsory to offer
solatium in money, and, finally, to accept it when offered.
The right of private revenge was lawful only after the
aggrieved individual had demanded, and been refused,
compensation at the recognized rate. Various codes formu-
lated rules for determining the amounts thus payable.
Each man had his money value or wer (from the simple
freeman, reckoned at 200 shillings, up to prelates and lay
nobles, estimated at much higher figures). Slighter
wrongs could be compensated by smaller sums, known as
bots : so much for a foot, or an eye, or a tooth. The King
or other lord exacted further payments from the wrong-doer,
under the name of wites, which are sometimes explained
as the price charged by the magistrate for enforcing pay-
ment of the wer or hot; sometimes as sums due to the
community, on the ground that every evil deed inflicts a
wrong on society in general, as well as upon its victim.
(3) Amercements, A third system succeeded. This is
found in working order soon after the Norman Conquest,
but was still regarded as an innovation at the accession of
Henry I. It is known as the system of amercements.
None of our authorities contains an entirely satisfactory
account of how the change took place; but the following
suggestions may be hazarded. The sums demanded from
a wrong-doer, who wished to buy himself back under pro-
tection of the law, became increasingly burdensome. He
had to satisfy claims of the victim's family, of the victim's
lord, of the lord within whose territory the crime had been
committed, of the church, mayhap, whose sanctuary had
been invaded, of other lords who could show an interest
of any sort, and finally of the King as lord paramount. It
became practically impossible to buy back the peace once
it had been broken. The Crown, however, stepped in, and
offered protection on certain conditions : the culprit sur-
rendered himself and all that he had to the King, placing
himself " in misericordiam regis," and delivering a tangible
286 MAGNA CARTA
pledge (vadium) as evidence and security of the surrender.^
Strictly speaking, the man's life and limbs and all that he
had were at the King's mercy .^ The Crown, however,
found that it might defeat its own interests by excessive
greed; and generally contented itself with moderate for-
feits. Rules of procedure were formulated : the amounts
taken were regulated partly by the wealth of the offender,
and partly by the gravity of the offence. Further, it became
a recognized rule that the amount should be assessed by
what was .practically a jury of the culprit's neighbours ; and
attempts were also made to fix a maximum.^
Thus a sort of tariff grew up, which the Crown usually
respected in practice, without abandoning the right to
demand more. Such payments were known as " amerce-
ments." For petty offences, men were constantly placed
" in mercy " : for failure to attend meetings of hundred or
county ; for false or mistaken verdicts ; for infringements
of forest rights. The Charter of Henry I. (chapter 8) had
promised a remedy, drastic indeed but of a reactionary and
impossible nature. His promise, to abolish altogether the
system of amercements (then of recent introduction) and to
revert to the earlier Anglo-Saxon system of bots and wites,
was made only to be broken.^
^ See Charter of Henry I. c. 8, which, however, condemns the whole practice
among the other innovations of the Conqueror and Rufus.
^ See Dialogiis de Scaccario, II. xvi.
^Cf. Pollock and Maitland, II. 51 1-4. There were, however, exceptions, e.g.
Henry II. would not accept money payments for certain forest offences : mutila-
tion was inflicted. See Assize of Woodstock, c. i, and contrast Forest Charter
of 1217, c. 10.
*Cf. Pollock and Maitland (II. 512), who describe Henry's promise as "a
return to the old Anglo-Saxon system of pre-appointed wites." In order to avoid
confusion, no mention has been made in the account given above of a classifica-
tion of amercements into three degrees, which increases the obscurity surrounding
their origin. ThQ Dialogus de Scaccario, II. xvi., tells how (i) for grave crimes,
the culprit's life and limbs were at the King's mercy, as well as his property ; (2) for
less important offences, his lands were forfeited, but his person was safe ; while (3)
for minor faults, his personal effects only were at the King's disposal. In the last
case, the offender was "z« misericordia regis de pecnnia sua." Thus to be "in
mercy " did not always mean the same thing. Further, a villein or dependent
freeman on a manor might fall into the *' mercy " of his lord, as well as of the King.
The records of manorial courts are full of amercements for petty transgressions of
customs of the manor.
CHAPTER TWENTY 287
II. Magna Carta and Amercements. No one could
expect to pass through Hfe (perhaps hardly through a single
year) without being subjected to amercements.^ Three
chapters of Magna Carta accordingly are occupied with
remedies. Chapter 20 seeks to protect the ordinary layman ;
chapter 21, the barons; and chapter 22, the clergy — thus
anticipating the conception of three estates of the realm ; —
commons, nobles, clergy. The " third estate " is analysed
for purposes of this clause, into three subdivisions — the free-
man, the villein, and the merchant.^
(i) Amercement of freeholders. The great object of the
reforms here promised was to eliminate the arbitrary
element; the Crown must conform to its own customary
rules. With this object, safeguards were devised for free-
men, (a) For a slight offence, only a petty sum could be
taken. This was nothing new : the records of John's reign
show that, both before and after 12 15, very small amounts
were often taken : threepence was a common sum. (h) For
grave offences, a larger sum might be assessed, but not out
of proportion to the offence, (c) In no case could the
offender be pushed absolutely to the wall : his means of
livelihood must be saved to him. Even if all other effects
had to be sold off to pay the amount assessed, he was to
retain his " contenement," a word to be afterwards dis-
cussed, (d) Another clause provided machinery for giving
effect to these rules. The amount must be fixed, not
arbitrarily by the Crown, but by impartial assessors, " by
the oath of honest men of the neighbourhood." In the
reissue of 12 16 "honest men" became "honest and law-
worthy (legalium) men," a purely verbal change.^
1 "Very likely there was no clause in Magna Carta more grateful to the mass of
the people than that about amercements." Maitland, Gloucestei- Pleas, xxxiv.
2 Even Coke {Second Insiihite, p. 27) has to confess that for purposes of this
chapter he must abandon the attempt made elsewhere {ibid., p. 4 and p. 45) to
bring villeins into the class of freemen.
3 Adams, Origin, 257, thinks the addition made it clear tliat villeins could not
amerce the liber homo ; but were not the four legaliores homines of each village
described in Assize of Clarendon, villeins ? Harcourt, Stexvard, 221 n., insists that
the clause does not secure *' trial by peers " in the feudal sense, for the jury of
neighbours need not be *' peers of a tenure."
288 MAGNA CARTA
There were apparently two steps in the fixing of amerce-
ments, (a) In the case of a commoner, the penalty under
normal circumstances would be assessed provisionally by
the King's justices on circuit, with the assistance of the
sheriff. It was their duty to see that the amount was pro-
portionate to the gravity of the offence.^ (h) Thereafter,
the sheriff or his serjeants, in full county court, with the
assistance of twelve neighbours, taxed the amercements,
reducing them in accordance with their knowledge of the
wrong-doer's ability to pay.^
The Pipe Rolls afford illustrations of the practice. In the
fourteenth year of Henry 11.^ a certain priest (who, in this
respect, stood on the same footing as a layman) had been
placed "in misericordiam " of lOO marks by William fitz
John, one of the King's justices, but that sum was after-
wards reduced to 40 marks "per sacramentum vicmorum
siiorum," It seems a safe inference that, on the priest
pleading poverty, the question of his ability to pay was
referred to local recognitors with the result stated. This
priest was subsequently pardoned altogether " because of
his poverty." ^
Magna Carta in this chapter, treating of the amercements
of freeholders, merchants and villeins, makes no reference
to the part played by the King's justices, but only to the
functions of the jury of neighbours.^ "^11 this is in marked
contrast with the provisions of chapte\2i, regulating the
treatment to be accorded to earls and Yarons who made
default. ^
(2) Amercement of merchants. The trader is in the same
position as the liher homo, except that it is his " merchan-
dise," not his " contenement," that is protected. The word
^ Harcourt, En^I. Hist. Rev., XXII. 733-4. See also Dial, de Scac, p. 207 n. ;
Maitland, Gloucester Pleas ^ xxxiv. Amercements apparently might also be pro-
visionally fixed by the justices of the bench or the barons of exchequer, who might
(where arrears were still unpaid) reduce their figures of previous years.
'Harcourt, ?;5/^. ^ jyjadox, I. 527.
* See, however, on whole subject, Harcourt, ibid.
^ Reeves, History of English Law, I. 248 (Third Edition) says : ** Upon this
chapter was afterwards framed the writ de moderata inisericordia, for giving remedy
to a party who was excessively amerced."
CHAPTER TWENTY 289
is capable of two somewhat different shades of meaning.
Narrowly interpreted, it may refer to his wares, the stock-in-
trade without which the pursuit of his calling would be
impossible. More broadly viewed, it might mean his busi-
ness itself, his position as a merchant. The difference is of
little practical import: in either view the Charter saves to
him his means of earning a living.^
Some boroughs, indeed, had anticipated Magna Carta by
obtaining in their own charters a definition of the maximum
amercement exigible, or in some cases of the amercing
body. Thus, John's Charter to Dunwich of 29th June,
1200,2 provides that the burgesses shall only be amerced by
six men from within the borough, and six men from with-
out. The capital had special privileges : in his Charter to
London, Henry I. promised that no citizen in misericordia
pecuniae should pay a higher sum than loos. (the amount of
his wer).^ This was confirmed in the Charter of Henry H.,
who declared " that none shall be adjudged for amercements
of money, but according to the law of the city, which they
had in the time of King Henry, my grandfather." * John's
Charter to London of 17th June, 1 199, also referred to this ; ^
and the general confirmation of customs, contained in
chapter 13 of Magna Carta, would further strengthen it. In
all probability, the earlier grant covered trivial offences only
(such as placed the offender in the King's hands de miseri-
cordia pecuniae). The present chapter is wider in its scope,
applying to grave offences also, and embracing merchants
everywhere, not merely the burgesses of chartered towns.
(3) Amercements of villeins. The early history of
villeins as a class is enveloped in the mists that still surround
the rise of the English manor. Notwithstanding the bril-
liant efforts of Mr. Frederic Seebohm ^ to find the origin of
1 Cf. Professor James Tait, Ejt^L Hist. Rev. XXXVII. , 720 ff. , who thinks that any
attempt to exempt merchant " wares " from amercement was inconsistent with the
right to distrain goods for debt, as illustrated by many cases given by Gross, Sel.
Cases in Merchant Law (Selden Society), passim.
2 Rotzili Ckartarum, 51. ^ See Select Charters, 108.
* See Birch, Historical Charters of London^ p. 5.
5 Ibid. , p. 1 1 . • See English Village Community, passim.
T
290 MAGNA CARTA
villeinage in the status of the serfs who worked for Roman
masters upon British farms long before the Teutonic immi-
grations began, an older theory still holds the field, namely,
that the abject villeins of Norman days were descendants
of free-born " ceorls " of Anglo-Saxon stock. On this
theory, most of England was once cultivated by Anglo-
Saxon peasant proprietors grouped in little societies, each
of which formed an isolated village. These villagers were
slowly sinking from their originally free estate during
several centuries prior to 1066 : but the process of their
degradation was completed rapidly and roughly by the
Norman conquerors. The once free peasantry were crushed
down into the dependent villeins of the eleventh and twelfth
centuries.
Whichever theory may be the correct one, the position,
economic, legal, and political, of villeins in the thirteenth
century has been ascertained with certainty. Economically
they were part of the equipment of the manor of their lord,
whose fields they had to cultivate as a condition of being
left in possession of acres, in a sense, their own. The
services exacted, at first vague and undefined, were gradu-
ally specified and limited. They varied from century to
century, from district to district, and even from manor to
manor ; but at best the life of the villein was, as a contem-
porary writer has described it, burdensome and wretched
(graviter et miserabiliter). After his obligations were dis-
charged, little time was left him for the ploughing and
reaping of his own holding. The normal villein possessed
his virgate or half virgate (thirty or fifteen scattered acres)
under a tenure known as villenagium, sharply distinguished
from the freeholder's tenures. He was a dependent dweller
on a manor which he dared not quit without his master's
leave.
It is true that he had rights of a proprietary nature in the
acres he claimed as his own ; yet these were determined,
not by the common law of England, but by " the custom of
the manor," or virtually at the will of the lord. These
rights, such as they were, could not be pled elsewhere than
before the court customary of that manor over which the
CHAPTER TWENTY 291
lord's steward presided with powers wide and undefined.
Politically his position was peculiar : allowed none of the
privileges, he was yet expected to perform some of the
duties, of the freeman. He attended the shire and hundred
courts, and acted on juries, thus suffering still further
encroachments on the scanty portion of time he might call
his own, but preserving for a brighter day a vague tradi-
tion of his earlier liberty.
This chapter extends some measure of protection to
villeins. Two questions, however, may be asked : — What
measure ? and from what motive ? One point is clear : the
villeins were protected from the abuse of only such amerce-
ments as John himself might inflict, not from the amerce-
ments of their manorial lords ; for the words used are " si
inciderint in misericordiam nostravi." A villein in the ,'
King's mercy shall enjoy the same consideration as the /
freeholder or merchant in similar plight — his means of I
livelihood being saved to him. The word now used is
neither " contenement " nor "merchandise," but " wayna-
gium," the meaning of which has been the subject of
discussion. Coke defined it as " the contenement of a vil-
lein; or the furniture of his cart or wain," and Coke has
been widely followed. The word, however, has apparently
no connexion with wains or wagons, but is merely a
Latinized form of the French word " gagnage," of which
Godef roy gives five meanings : (a) gain ; (h) tillage ; (c)
crop; (d) land under the plough; (e) grain. Professor
Tait is inclined to read the word, in its present context, as
equivalent either to " crops " or to " lands under cultiva-
tion," and to translate the clause "saving his tillage." ^
What was the motive of these restrictions? It is usually
1 See JSn^l Hist. Rev., XXXVII. 724, where Mr. Tait argues " for a broader and
less concrete interpretation of the term . . . than has hitherto been put upon it."
The villein was not to be ruined by impounding his .seed-corn or growing crops
any more than by depriving him of his plough or plough team. See also A. F.
Pollard, Engl. Hist. Rev., XXXVIII. 117, and cf. waynagitir. in c. 5, supra. The
Mirror of Justices, p. 169, has a gloss on this passage, in which it is the villein's
"gaigneur" that is saved to him, and this is apparently identified with the
villenagium held by him. Mr. Tait's view has been adopted here ; but the word
has sometimes a more restricted meaning, e.g. in Hoveden, iv. 48, where 100 acres
of land are reckoned to the " waynage " of each plough.
292 MAGNA CARTA
supposed to have been clemency, the humane desire not to
reduce a poor wretch to absolute beggary. It is possible,
however, to imagine a different motive ; the villein was the
property of his lord, and John must respect the vested
interests of others. That the King might do what he
pleased with his own property, his demesne villeins, seems
clear from a passage usually neglected by commentators,
namely, chapter i6 of the reissue of 1217. Four important
words were there introduced — villanus alterius quam
noster : the king was not to inflict crushing amercements on
villeins " other than his own," thus leaving villeins on royal
manors unreservedly in his power. ^
It must not be thought, however, that the position of
the King's villeins was worse than that of villeins of an
ordinary unroyal manor. On the contrary, it has been
clearly shown 2 that the King's peasants enjoyed privileges
denied to the peasants of other lords. Magna Carta pro-
tected a lord's villeins from the King, not from the lord
who owned them. That " great bulwark of the people's
rights " left the bulk of the rural population of England at
the mercy of their lords. The King must not take so much
from any lord's villeins as to destroy their usefulness as
manorial chattels; that was all.^
(4) The difference hetiveen fines and amercements. In the
thirteenth century, these terms were sharply contrasted.
" Amercement " was applied to sums imposed in punish-
1 The view here taken of the motive for protecting villeins is strengthened by the
use of the peculiar phrase, "vas^ttm ho/?nmtm" in chapter 4 {^.v.). Thomson,
Magna Charta^ p. 202, seems completely to have misunderstood this 1 6th chapter
of the reissue of 121 7, construing the four interpolated words in a sense the Latin
will not bear, viz. : ** A villein, although he belonged to another.''''
2 Notably by Professor Vinogradoff in his Villeinage in England^ passim.
^ The gulf which separated villein from freeman in this matter is shown by the
Pipe Roll oi 16 Henry II. (cited Madox, I. 545) ; Herbertus Faber debet j marcam
pro /also clamore qtiem fecit ut liber mm sit rusticus. A villein might be amerced
for merely claiming to be free. It is difficult to reconcile any theory of the
villein's freedom with the doctrine of Glanvill, V. c. 5, who denies to everyone
who had been once a villein the right to "wage his law," even after emancipa-
tion, where any third party's interests might thereby be prejudiced. R. Hoveden,
iv. 46, speaking of the carucage of 1198, explains that for perjury a villein forfeited
his best ox to his lord (not to the King).
CHAPTER TWENTY 293
ment of misdeeds; the law-breaker had no option of
refusing, and no voice in fixing the amount. " Fine," on
the contrary, was used for voluntary offerings made to the
King to obtain some favour or to escape punishment. Here
the initiative rested with the individual, who suggested the
amount to be paid, and was, indeed, under no legal obliga-
tion to make any offer at all. This distinction between
fines and amercements, absolute in theory, could readily be
obliterated in practice. The spirit of the restriction placed
by this chapter and by the common law upon the King's
prerogative of inflicting amercements could often be evaded.
The Crown might imprison its victims for an indefinite
period, and then graciously allow them to offer large pay-
ments to escape death by fever or starvation in a noisome
gaol : enormous fines might thus be taken, while royal
officials were forbidden to inflict arbitrary amercements.
With the gradual elimination of the voluntary element
the word " fine " came to bear its modern meaning, while
"amercement" dropped out of ordinary use.''^
(5) Contenement. This word, which occurs in Glanvill ^
and in Bracton,^ and also (in its French form) in the Statute
of Westminster, I.,^ as well as in Magna Carta, has formed
a text for many commentators from Coke's days to our own.
By comparing the entries from exchequer rolls brought
together by Madox,^ it appears that to save a man's "con-
tenement " was to leave him sufficient for the sustenance of
himself and those dependent on him. The word comes
from the French " contenir," and has many shades of mean-
ing, as capacity, maintenance, appearance, social condition
or grade. A free man is not to be so crushed by an
amercement that he cannot maintain himself in his former
condition.^ Several entries on Exchequer Rolls of Henry
^ C. 55, which supplements this chapter, cancels amercements unjustly inflicted
in the past.
2 IX. 8. 3 III. folio I i6b_ 4 3 Edward I. c. 6. s See II. 208-9.
^ Prof. Tait's conclusions {op. cit.) have here been accepted with some hesitation,
*' Contenement," he urges, " is not a compound from tenement." He admits, how-
ever, following Godefroy, that in one instance the word does mean " tenement."
He does not notice the striking analogy between the use of " contenement" in this
chapter and that of "tenement" in c. 11 supra ; nor does he discuss the evidence
294 MAGNA CARTA
III. and Edward I., collected by Madox,i throw light on
the way in which a " contenement " might be saved to the
man amerced. Thus in 40 Henry III. the officials of the
exchequer, after discussing an offender's failure to pay an
amercement of 40 marks, ordered inquiry to be made, " how
much he was able to pay the King per annum, saving
his own sustenance and that of his wife and children," an
excerpt which illustrates the more humane side of exchequer
procedure. In 14 Edward I. again, the officials of that
day, when ferreting out arrears, found that certain poor men
of the village of Doddington had not paid their amerce-
ments in full. An inquiry was set on foot, and the barons
of exchequer were ordered to fix the dates at which the
various debtors should discharge their debts (evidently an
arrangement for payment by instalments) "salvo contene-
mento suo." ^
These illustrations of the procedure of later reigns,
agreeing closely with the rules laid down by the Great
Charter, show how a man's contenement might be saved
to him without loss to the Crown. Magna Carta apparently
desired that time should be granted in which to pay up
debts by degrees. Meanwhile, the amerced freeman was
not forced to part with what was necessary to maintain him,
with his wife and family, in his proper station in life.
of the contemporary Histoire de Gtiillaume le Marichal^ where the word appears
seven times with various meanings, e.g. capacity, manner of being, conduct, and
equipment. M. Paul Meyer has collected these in his index. Mr. Tait goes too
far when he asserts that to make freehold liable to amercement shows "a com-
plete misconception of that form of punishment," p. 726. There were three
degrees of amercement; and only for the mildest of the three was the forfeit limited
to the culprit's personal estate {de peamia). See supra^ p. 286, n. Again, a man
might be forced to sell his freehold to meet a heavy pecuniary mulct. Under
Henry's Charter, in its final form, no ecclesiastic could be amerced except in
accordance with his "tenement," which suggests an analogy with the saving of
a freeman's ** contenement " in the present passage.
1 See II. 208-9. * ^ See Madox, ibid.
CHAPTER TWENTY-ONE 295
CHAPTER TWENTY-ONE.
CoMiTES et barones non amercientur nisi per pares suos, et
non nisi secundum modum delicti.
Earls and barons shall not be amerced except through
their peers, and only in accordance with the degree of the
offence.
Amercement of earls and barons. It is noteworthy that
the Articuli Baronum contain no provisions corresponding
to this chapter, which forms in one sense a supplement to
chapter 20, and in another to chapter 39 of John's Charter.
How is the omission from the earlier document to be
explained? Was it an oversight? Was the present
clause added at Runnymede as an afterthought for the sake
of symmetry? Had the barons no personal grievances
under this head to redress ? Were they too disinterested to
urge them ? Or was the grievance of so notable a kind and
so hard to remedy that they hesitated till the last moment
before committing themselves to any form of expression ?
There is no contemporary evidence on which to base a con-
clusive answer to these questions; but much may be said
for answering the last of them in the affirmative.
The equally natural question as to what the actual words
of the Charter stood for in the minds of the barons is also
hard to answer. When they mentioned amercement per
pares suos, what exactly did they desire ? Bracton ^ has
a famous gloss on this chapter, in which these words seem
at first sight to be replaced by the phrase " et hoc per
barones de scaccario vel coram ipso rege." Is this to be
taken as an honest paraphrase? or does it represent a
deliberate attempt by Crown lawyers to pervert the plain
words of the Charter to authorize precisely what they had
been originally intended to forbid? — to substitute the
decision of a small knot of royal officials for that of the
community of feudal barons ? While the problem is
MIL folio 1 1 6b.
296 MAGNA CARTA
perhaps insoluble, some suggestions may be founded on
a consideration of the actual practice before and after
1215.1
The plea rolls contain no distinct evidence of two stages
in the amercement of barons, corresponding to those
described in connexion with commoners. It is clear, how-
ever, that the justices on circuit had no power to fix the
iamercements imposed on them : in their case a blank was
usually left to be afterwards filled in at the exchequer.
" For this purpose, a separate roll or schedule was prepared
containing the names of the amerced barons with the
offences for which they were penalised, and this was sent
to the exchequer with the other estreats." ^
This was the course followed at an eyre held at Hertford
in 1 1 98-9: when a list of the amerced was prepared and
definite sums were entered after each ordinary offender's
name, blanks were left after the names of Gerard de
Furnivall and Reginald de Argenton, each of whom was
reserved for special treatment " as a baron," and as such " to
be amerced at the Exchequer for a disseisin." The Pipe
Roll of John's first year shows that this procedure was
carried out.^
Magna Carta, then, had good precedents for insisting
that barons ought not to be amerced by the justices of eyre
in the course of their circuits ; but what exactly did it mean
by demanding amercement " by their peers " ? In asking
amercement per pares suos, were they iherely acquiescing
in John's current practice? Did they desire to substitute
the decision of a full commune concilium, as defined in
chapter 14, for that of the King's professional justices?
Did they merely ask for the presence of a few barons at
^ A valuable volume of evidence has been collected by Harcourt, ^«^. //ist. Rev.
XXII. 733 ff. ; though his conclusions are mainly negative. See also his Steward^
ff. 289.
^Harcourt, ibid., 736. Pike, Hotise of Lords, 256-7, shows how barons were
assessed sometimes — (a) before the barons of exchequer ; or {b) before the full
King's Council ; or {c) at a later date, even before the justices of Common Pleas.
They were never assessed, however, before the justices on circuit.
^See Pike, House of Lords, 255.
CHAPTER TWENTY-ONE 297
the exchequer, when one of their own class was being
amerced? Or, did they refer to a second stage of pro-
cedure in which the amercements of barons should be taxed
or reduced by other barons, just as (in the procedure
referred to in chapter 20) amercements of commoners were
taxed by a jury of neighbours ?
If the last query could be answered in the affirmative,
a clue would be afforded to the interpretation of Bracton's
gloss: — " Comites vero vel barones non sunt amerciandi
nisi per pares suos et secundum modum delicti et hoc per
barones de scaccario vel coram ipso rege." ^ The words
" et hoc " may here refer merely to the first stage in the pro-
cess, the provisional fixing of the amount at the exchequer
secundum modum delicti, while the function of the baron's
" peers " was to " tax " this amount, with reference to the
circumstances of the defaulting baron. If this interpreta-
tion of Bracton be admissible and if he has accurately
paraphrased the substance of this chapter, then the barons
were asking no more for themselves than they had already
asked for their humble dependents. They were unlikely
to ask less.
In the fourteenth century several cases are recorded in
the course of which defaulters, in the hope of escaping with
smaller payments, protested against being reckoned as
barons. Thomas de Furnivall, for example, in the nine-
teenth year of Edward II. complained that he had been
amerced as a baron " to his great damage, and against the
law and custom of the realm," whereas he really held
nothing by barony. The King directed the Treasurer and
Barons of Exchequer " that if it appeared to them that
Thomas was not a baron, nor did hold his land by barony,
then they should discharge him of the said imposed amerce-
ment ; provided that Thomas should be amerced according
to the tenor of the great Charter of Liberties, " ^ that is to
say, as a simple freeholder according to the provisions of
chapter 20. It is clear that Thomas de Furnivall was
confident that a local jury would " tax " him at a lower
figure than that fixed by the Exchequer barons. A few
1 Bracton, f. Ii6b. "^adox, I. 535-8.
298 MAGNA CARTA
years earlier the Abbot of Croyland had made a similar
claim, but without success.^
, At a later date, barons and earls were successful in
securing by another expedient some measure of immunity
from excessive exactions. They had established, prior to
the first year of Henry VI., a recognized scale of amerce-
ments with which the Crown was expected, in ordinary
circumstances, to content itself.^ In the reign of Edward
VI. a duke was normally amerced at ;i£"io, and an earl or
a bishop at loos.^
CHAPTER TWENTY-TWO.
NuLLUS clericus amercietur de laico tenemento suo, nisi
secundum modum aliorum predictorum, et non secundum
quantitatem beneficii sui ecclesiastici.
A clerk shall not be amerced in respect of his lay holding
except after the manner of the others aforesaid ; further, he shall
not be amerced in accordance with the extent of his ecclesiastical
benefice.
Amercement of the clergy. The churchman was to
receive the same favourable treatment as the layman in
all respects, and to enjoy one additional privilege. In
proportioning the amercement to his means, no account
was to be taken of the value of his " church benefice."
There is room, however, for doubt as to the precise nature
of this privilege, which seems to depend for its point on
an antithesis between " lay tenement " and " ecclesiastical
benefice."
In a well-known article of the Constitutions of .Clarendon
1 See Madox, idid, and also Pike, Hoztse of Lords^ 257. Mr. Pike, p. 255,
rightly says that what was originally a privilege had become a burden.
2 See Pike, ibid.
3 Madox, Baronia Anglica^ 106, seems to view these sums as fixing a minimum,
not a maximum. ** If a baron was to be amerced for a small trespass, his amerce-
-ment was wont to be lOOs. at the least ; he might be amerced at more, not at less.
This, I think, was the meaning of the term amerciater ut ba7-o." He adds that a
■commoner for a similar trespass would get off with los., 20s., or 40s.
CHAPTER TWENTY-TWO 299
(c. 9), a contrast is drawn between laicum feudum and
tenementuTn pertinens ad eleemosinam. It is possible that
Magna Carta means to observe the same distinction between
"lay fee" and " frankalmoin," reckoning the former, but
not the latter, in estimating a clerk's ability to pay amerce-
ments.
A more likely interpretation is that the contrast is drawn
between lands owned by a clerk absolutely, and lands
belonging to the church and held by the clerk in liferent.
The plausibility of this conjecture is strengthened by
alterations, apparently of a purely verbal nature, made in
reissues of the Charter. The " de laico tenemento " of
1215 was omitted altogether in 1216; and in 1217, the
provision took this final form : " Nulla ecclesiastica persona
amerciatur secundum quantitatem beneficii sui ecclesiastici,
sed secundum tenementum ^ suum et secundum quantitatem
delicti." The substitution of ecclesiastical " person " — a
word fast acquiring even then a connotation like that of
the " parson " of present-day colloquial speech — for " clerk "
has no significance, but the main antithesis drawn would
seem to be between the " benefice " or mere liferent and
the " tenement " or fief held in perpetuity. In taxing a
clerk's amercement, no account was to be taken of posses-
sions of which he was not really owner.
CHAPTER TWENTY-THREE.
Nec villa nee homo distringatur facere pontes ad riparias,
nisi qui ab antiquo et de jure facere debent.
No village or individual ^ shall be compelled to make bridges
at river banks, except those who from of old were legally bound
to do so.
^Stubbs, SeL Chart., 345, by a curious oversight i^ads ** contenementum," in
the issue of 1217, for which there seems to be no authority.
2 The word *^ villa,'" used at first as synonymous with "manor," came to be
freely applied not only to all villages, but also to chartered towns. Even London
was described as a villa in formal writs. " Homo," though often loosely used, was
the word naturally applied to a feudal tenant. The version given by Coke {Second
Institute, p. 30) reads " liber homo," which is also the reading of one MS. of the
Inspexiimis of 1297 (25 Edward I.), See Statutes of the Realm, I. 114.
300 MAGNA CARTA
The object of this chapter is obvious; to compel the
King to desist from his practice of illegally increasing the
burden of the obligation to keep in repair all bridges over
rivers. John might continue to exact what his ancestors
had exacted; but nothing more. So much lies on the
surface of the Charter, which explains, however, neither
the origin of the obligation nor the reasons that made John
keen to enforce it.
I. Origin of Obligation to make Bridges. The Norman
kings seem to have based their claim to compel their
subjects to maintain bridges upon the ancient tripartite
obligation^ (known as the trinoda necessitas). Three
duties were required of all the men of England in the
interests of the commonweal : attendance on the fyrd or
local militia ; the making of roads, so necessary for military
purposes; and the repairing of bridges and fortifications.
Gradually, as feudal tendencies prevailed, the obligation
to construct bridges ceased to be a personal burden upon
all freemen, and became a territorial burden .^ The present
chapter seems to be a particular application of the general
principle enunciated in chapter i6. " Brigbot " required
special treatment because of the prominence into which it
had been forced by John.
II. The King's interest in the Repair of Bridges. Part
at least of John's motives for making an oppressive use of
this prerogative must be sought in his rights of falconry.
Whenever John proposed to ride a-fowling, with his hawk
upon his wrist, he issued letters compelling the whole
country-side to bestir themselves in the repair of bridges.
Several such writs of Henry III. are extant. The exact
words vary somewhat, but comparison leaves no room for
doubt either as to the nature of the commands conveyed
or the reasons for issuing them. Addressed to sheriffs of
such counties as the King was likely to visit, these letters
1 See I^oL C/aus., 19 Henry III., cited by Moore, History and Law of Fisheries,
p. 8.
2 The Hundred Rolls illustrate the manner of its incidence ; e.g. Omnes tenentes
de Spaldinge debent ad reparacionem pontis illius, quilibet pro rata porcionis terrae
suae contribuere, ita quod quaelibet acra erit par alterius. Rot. Hund., I. 468.
CHAPTER TWENTY-THREE 301
gave instructions for repair of bridges, and a prohibition
against the taking of birds before the King had enjoyed
his sport. Both points are well brought out in a Letter
Close of Henry III., dated 26th December, 1234, which
directed " all bridges on the rivers Avon, Test, and Itchen
to be repaired as was wont in the time of King John, so
that when the lord King may come to th6SS^*pafts, free
transit shall lie open to him for ' revaying ' {ad riviandum)
upon the said rivers." The sheriff is to issue a general
prohibition against any one attempting to " revaye " along
the river banks, previous to the coming of the King.^ The
Latin verb, for which the Old English word " revaye " or
" ryvaye " is an exact equivalent, has been the subject of
misconception ; but conclusive evidence has recently been
adduced to prove that it referred thi the medieval sport of
fowling, that is to the taking of jl^ld birds in sport by
means of hawks and falcons.^ N^ \
These writs prove that the Crown^^imed a preferential
right to this form of sport along the banks of certain rivers ;
and these " preserved " rivers were said to be " in defence "
(in defenso), a phrase which occurs also in a later chapter
of Magna Carta. ^
Two distinct hardships were thus imposed by the King's
exercise of his rights of falconry, one negative and the
other positive. Between the King's intimation and his
arrival at the indicated rivers, the sport of other people was
forbidden, while whole villages had to forsake their ploughs
to reconstruct otherwise useless bridges. A wise king
would be careful to use such rights so as to inflict a mini-
mum of hardship. John knew no moderation, placing
" in defence " not merely a few banks ct a time, but many
1 See J?of. C/aus., 19 Henry III., cited in Moore, History and Law of Fisheries ^
p. 8.
'See Moore, ibid.^ 8-16. Two links in the chain of evidence are worthy of
emphasis: {a) Writs of 13th November and ist December, 1234, order repair of
bridges for the transit of the King "along with his birds." {b) A writ of 28th
October, 1283, contains a licence to 'the Earl of Hereford "during the present
winter season to 'revaye' and take river-fowl throughout the rivers Lowe and
Frome which are in defence."
^I.e. c. 47 {q.v.).
302 MAGNA CARTA
rivers indiscriminately, including those which had never
been so treated in his father's day, and demanding that
all bridges everywhere should be repaired, with the object,
not so much of indulging a genuine love of sport, as of
inflicting heavy amercements on those who neglected
prompt obedience to his commands. Great consternation
was aroused when John at Bristol in 1209 prohibited the
taking of birds throughout the entire realm of England.^
Both grievances were redressed by Magna Carta. The
present chapter promised not to impose the burden on those
from whom it was not legally due.^ Chapter 47, in which
he agreed to withdraw his interdict from all rivers which
had not been previously "in defence," and to disafforest
all forests of his own creation, was entirely omitted in
the Charter of 1216;^ but in 1217 it reappeared in a new
position and expressed in different words. The provision
in the original chapter 47 that related to forests was rele-
gated to the Carta de Foresta, and the other part of that
chapter, relating to falconry, was joined to a clause which
redressed another grievance growing from the same root.
Chapter 19 of Henry III.'s Charter, in its final form,
repeats word for word the terms of the present chapter of
John, while in chapter 20 Henry proceeds to declare " that
no river shall in future be placed in defence except such
as were in defence in the time of King Henry, our grand-
father, throughout the same places and during the same
periods as they were wont in his day."
This express prohibition seems to have prevented the
Crown from extending its prerogatives further in this
direction. Yet Henry HI. had ample opportunities of
harassing his subjects by an inconsiderate use of the rights
^R. Wendover, II. 49 (R.S.)j ^^ Ibi capturam avium per totam Angliam inter-
dixit y
2 Article 1 1 of the Barons had demanded that no villa should be amerced for
failure to make illegal repairs, thus illustrating at once John's policy, and the point
of connection between this provision and the immediately preceding chapters which
dealt with amercements.
* It was, however, included among the subjects reserved for further consideration
in "the respiting clause" (c. 42 of 1216) under the words ''* de ripariis et earum
custodibus y Cf. supra, 143.
CHAPTER TWENTY-THREE 303
still left to him. In many cases dubiety existed as to what
banks had actually been "preserved" by Henry II., and
a vague general command left in cruel uncertainty the
district to be visited. Henry III. made important conces-
sions : after the year 1241, he specified the particular river
along whose banks he intended to sport, and sometimes
announced the exact date at which he expected to arrive.
As no writs appear subsequent to 1247, it is possible that
he was induced to abstain from the exercise of a right
which inflicted hardships out of all proportion to the
benefits conferred on the King.^
The Crown, however, had not renounced its prerogatives,
and several writs still exist to show that Edward I.
occasionally allowed his great nobles to share in the royal
sport. Licences were granted in 1283 to the Earl of Here-
ford and to Reginald fitz Peter, and in the following year
to the Earl of Lincoln. On 6th October, 1373, Edward III.
commanded the sheriff of Oxfordshire to declare that all
bridges must be repaired and all fords marked out with
stakes, for the crossing of the King " with his falcons "
during the approaching winter.^
III. Erroneous Interpretations, It is not surprising that
a pastime so passionately followed as falconry, should have
left its traces on two chapters of Magna Carta, the full
import of which has not been appreciated by commentators,
partly from failure to read them together, but chiefly
through the assumption that the words ad riviandum and
in defenso referred to fishing rather than to fowling.^
It has been confidently inferred that the framers of
Magna Carta, when forbidding additional banks to be put
" in defence, " equally as when demanding the removal of
" weirs " from non-tidal waters,^ intended to preserve public
rights of fishing against encroachment. This is an error :
in the Middle Ages, fishing was a means of procuring food,
not a popular form of sport : to depict John and his action-
^ Moore, idzd.^ g. ^^oore, idz'ci., 12.
3 The Mirror of Justices is cited as first suggesting this. See Moore, ibid., 12-16.
Coke, Second Institute^ 30, misled by the Mif-ror, has misled others.
"* Cf. infra, under c. 33.
304 MAGNA CARTA
loving courtiers as exponents of the gentle art of Isaac
Walton is a ridiculous anachronism.
It is true that the value of fish as an article of diet led
in time to legislation directed primarily to their protec-
tion ; but apparently no statute with such a motive was
passed previous to 1285.^ It is further true that in the
reign of Edward I. it became usual to describe rivers, over
which exclusive rights of fishing had been established by
riparian owners, as being in defenso ; ^ but rivers might be
" preserved " for more purposes than one.
From Edward's reign onwards, however, rights of fishing
steadily became more valuable, while falconry was super-
seded by other pastimes. Accordingly a new meaning was
sought for provisions of Magna Carta, whose original
motive had been forgotten. So early as the year 1283 the
words of a petition to the King in Parliament show that
"fishing" had been substituted for "hawking," in inter-
preting the prohibition referred to in chapter 47 of John's
Charter. The men of York complained that Earl Richard
had interfered with their rights of fishing by placing in
defenso the rivers Ouse and Yore " against the tenor of
Magna Carta." ^ This error, which thus dates from 1283,
has been accepted for upwards of five hundred years by all
commentators on Magna Carta. The credit for dispelling
it is due to Mr. Stuart A. Moore and Mr. H. S. Moore
in their History and Law of Fisheries, published in 1903.^
CHAPTER TWENTY-FOUR.
NuLLUS vicecomes, constabularius, coronatores, vel alii
ballivi nostri, teneant placita corone nostre.
No sheriff, constable, coroners, or others of our bailiffs, shall
hold pleas of our Crown.
^This was 13 Edward I., stat. i, c. 47, cited Moore, ibid., 173.
Udid., p. 6. ^/did.y p. 16.
*Lord Hale (Hargreaves, Law Traces, p. 7) partly anticipated their conclusions,
and he seems to have been followed by decisions of the New York Courts. See
Law Notes (New York) for August, 1 905.
CHAPTER TWENTY-FOUR 305
The main object of this provision is not open to doubt :
men accused of crimes must be tried before the King's
judges and not by local magistrates of whatsoever kind.
Innocent men dreaded the jurisdiction of the local tyrants
whose harshness had earned widespread hatred. The
sheriffs and castellans deserved their bad repute ; the
records of the age overflow with tales of their cruelties and
oppressions. It ought not to be forgotten, however, that
if this chapter contains a condemnation of the local admini-
stration of justice, it testifies to the comparative purity of
the justice dispensed by the King's own judges. So far
there is no difficulty ; but differences of opinion exist as to
certain points of detail.
I. Pleas of the Crown. Pleas were royal or common
according as the interests of the Crown were or were not
involved. This classification has already been discussed
in connection with chapter 17. The present chapter con-
cerns itself only with " pleas of the Crown," a phrase which
had, even in 12 15, considerably altered its original mean-
ing. In the eleventh century it had denoted royal business,
whether relating to judicial procedure or not, embracing
all matters connected with the King's household or his
estates, with the collection of his revenue, or the admini-
stration of his justice, civil as well as criminal. Gradually,
however, the usage of the word altered in two respects,
contracting in one direction, while expanding in another.
It ceased to be applied to financial business and even
to non-criminal, judicial business, and was reserved for
criminal trials held before the King's judges. This pro-
cess of contraction had been nearly completed before the
accession of John.
A tendency in an opposite direction had been for some
time in progress; the distinction drawn in early reigns
between petty trespasses, which were left in the province
of the sheriff, and grave offences, which alone were worthy
of the King's attention, was being slowly obliterated. ^
The central courts extended their activity over all misdeeds,
however trivial, until the whole realm of criminal law fell
^Traces may be found in Glanvill, I. c. i.
U
3o6 MAGNA CARTA
under the description of " pleas of the Crown." In John's
reign this process of expansion was far from complete :
the words then, indeed, embraced grave criminal offences
tried in the King's great courts, but not the petty offences
disposed of in the sheriff's tourn or elsewhere.^
North of the Tweed the same phrase has had a different
history: in modern Scots law its connotation is still a
narrow one; and this is a result of the slow growth of the
• Scottish Crown in authority and jurisdiction, in notable
contrast to the rapidity with which the English Crown
, attained its zenith. The Kings of Scotland failed to crush
their powerful vassals, and pleas of the Scottish Crown,
exclusively reserved for the High Court of Justiciary,
formed a meagre list — the four heinous crimes of murder,
robbery, rape, and arson. The feudal courts of the Scot-
tish nobles long preserved their wide jurisdiction over all
other offences. When the heritable jurisdictions were at
last abolished, in 1748, the old distinction, so deeply rooted
in Scots law, still remained. The sheriff court had no
cognizance, until late in the nineteenth century, over the
four crimes specially reserved for the King's judges. ^
Thus in Scotland the historic phrase " pleas of the Crown "
is, even at the present day, confined to murder, robbery,
rape, and fire-raising, while to an English lawyer it
embraces the entire realm of criminal law.
II. Keeping and Trying Criminal Pleas. The machinery
for bringing criminals to justice, as organized by Henry II.,
was somewhat elaborate. For our present purpose, it may
be sufficient to emphasize two important stages in the
procedure. An interval had always to elapse between the
commission of grave crimes and the formal trial of the
accused, for the coming of the itinerant justices took place
only at intervals of about seven years. Meanwhile, pre-
^ The triumph of royal justice over all rivals in the sphere of criminal law is thus
symbolized by the extension of the phrase "pleas of the Crovi^n," which can be
traced through a series of documents — e.^. [a) the laws of Cnut ; (d) Glanvill, I.
cc. I, 2, and 3 ; {c) the Assizes of Clarendon and Northampton ; {d) the ordinances
of II 94 ; and (e) Magna Carta.
2 The Criminal Procedure {Scotland) Act, 1887 (50 and 51 Victoria, c. 35) gave
him jurisdiction over three of them.
CHAPTER TWENTY-FOUR 307
liminary steps were taken to collect and record evidence,
which might otherwise be lost. The magistrate respons-
ible for these preliminary steps was said to " keep " the
pleas (custodire placita) — that is, to prevent them from
passing out of mind ^ while waiting for the justices who
would formally " hold " or " try " or " determine " them
(placitare or habere or tenere placita).
Before the reign of John, the two functions had been
entrusted to two distinct types of royal officials. The local
magistrates of each district " kept " royal pleas, while only
the King's justices could " hold " them. The process of
differentiation was accelerated in consequence of the
jealousy with which the Crown regarded the increasing
independence of the sheriffs. The elaborate instructions
issued in 1194 to the justices, whom Archbishop Hubert
Walter was despatching through the counties, contain pro-
visions intended to keep the pretensions of sheriffs within
bounds: 2 they were expressly forbidden to act as justices
within any counties in which they had acted as sheriffs
since Richard's coronation.^
It is safe to infer that the " trying " of royal pleas was
the province from which the sheriff was thus to be excluded.
Even with regard to the " keeping " or preliminary stages
of such pleas, the sheriff was by no means left in sole com-
mand. The justices received instructions ^ to cause three
knights and one clerk to be chosen in each county as
" custodes placitorum coronae." It is possible that these
new local officers, specially entrusted with the duty of
" keeping " royal pleas, were intended rather to co-operate
with, than to supersede, the sheriffs in this function ; but,
in any view, the sheriffs had no longer a monopoly of
authority in their bailiwicks. Magistrates, to be after-
wards known as coroners, were thenceforward associated
with them in the administration of the county.^
^Cf. tnfra, 315-6, for details.
^See Forma procedendi, cc. 20 and 21 {Sel. Chart. ^ 260).
^ Ibid., c. 21. '^Ibid., c. 20.
5 The Forma procedendi is usually considered the earliest distinct reference to
the office of coroner. Dr. Gross, however {History of Office of Coroner, 1892, and
Select Cases from Coroners' Rolls, 1896) claims to have found traces of their
3o8 MAGNA CARTA
The ordinance of 1194 seems to have settled subsequent
practice in both respects. Sheriffs, while still free to punish
petty offenders in their half-yearly tourns or circuits,
allowed the coroners to " keep " royal pleas, and the justices
to " try" them. Public opinion of the day approved both
rules. Yet John condoned and encouraged irregularities,
allowing sheriffs to meddle with pleas of the Crown, even
when no coroners were present to check their arbitrary
methods;^ and allowing them to give final judgments,
involving loss of life or limb, without waiting for the
Justices.2 He employed the same men to visit as justices
the very counties they had oppressed as sheriffs. The
notorious Engelard of Cigogne, branded by name in
chapter 50 of Magna Carta, acted as justice in his own
county of Gloucester.^
The Articles of the Barons condemned such practices;
and Magna Carta, in this first of a series of clauses directed
against sheriffs' misdeeds, forbade them under any circum-
stances to try royal pleas.
III. The Intention of Magna Carta. The barons were
merely demanding that the Crown should observe the rules
it had laid down for its own guidance : caprice must give
way to law. Sheriffs must not usurp the functions of
coroners; nor must sheriffs and coroners together usurp
those of King's justiciars. John's opponents associated
these two irregularities, and may have assumed that
expressly to abolish one implied an intention to abolish
both. Some such supposition would explain a peculiar
discrepancy between the Articles and the Charter. While
Article 14 demanded redress of one grievance. Magna
Carta granted redress of a different one. The earlier docu-
ment required that coroners should always be associated
with the sheriff when he meddled with pleas of the Crown :
the Charter forbade sheriffs and coroners to " try " pleas
existence at a much earlier date. Maitland remained unconvinced {£n^. Hist.
Rev.^ VIII. 758, and Pollock and Maitland, I. 519).
^ This inference is drawn from Article 14 of the Barons.
2 This inference is drawn from c. 24 of Magna Carta.
^ See Maitland, Gloucester Pleas, xx.
CHAPTER TWENTY-FOUR 309
of this description. These two provisions are the comple-
ments of each other. Magna Carta would seem to be here
incomplete.
The prohibition against sheriffs trying pleas of the Crown
was repeated in all reissues of the Charter; and, although
not strictly enforced in Henry's reign, soon became
absolute. Thus sheriff Ralph Musard was one of seven
justices of eyre who went on circuit in 1221, but he was
prohibited from sharing the labours of his colleagues when
they sat in Gloucestershire, where he was still sheriff.^
Under Edward I. no one could determine such pleas unless
armed with a royal commission to that effect ; ^ and the com-
mission would take the form either of gaol delivery, of trail-
baston, or of oyer and terminer.^
IV. An Erroneous View, Hallam misunderstood the
object of this provision. Commenting on Henry's Charter
of 1225, he declares that the "criminal jurisdiction of the
Sheriff is entirely taken away by Magna Carta, c. 17."*
This is a mistake : both before and after the granting of
the Charter, the sheriff exercised criminal jurisdiction, and
that of two kinds. Along with the coroners, he conducted
preliminary enquiries even into pleas of the Crown ; while
in his tourn (which was specially authorized to be held
twice a year by chapter 42 of the very Charter quoted by
Hallam) he was made responsible for every stage in the
trial of trivial offences. He heard indictments and then
condemned and punished petty offenders in a summary
manner.^ Several statutes of later reigns confirmed, even
while regulating, the authority of the sheriff to take indict-
ments at his tourns,^ until this jurisdiction was transferred,
^Idid., p. X.
2 See Coke, Second Institute, 30, and authorities there cited.
3 For explanation of these terms, see supra, c. 18.
*See Middle Ages, II. 482 n.
^ Cf. Stephen, History of Criminal Law, I. 83. The mistake made by Hallam
and others may have been in part the result of their neglecting the important
modification undergone by the phrase "pleas of the Crown" between 12 15 and
the present day.
^E.g. 13 Edward I. c. 13, and i Edward III., stat. 2, c. 17.
310 MAGNA CARTA
by an act of the fifteenth century, to the justices of peace
assembled in Quarter Sessions.^
All that Magna Carta did was to insist that no sheriff or
local magistrate should encroach on the province reserved
for the royal justices, namely the final " trying " of such
grave crimes as had now come to be recognized as " pleas of
the Crown." 2 The Charter did not even attempt to define
what these were, leaving the boundary between great and
small offences to be settled by use and wont. In all this,
it was simply declaratory of existing practice, making no
attempt to draw the line in a new place.^
Professor Hearnshaw^ propounds a theory that better
fits the facts. He holds that this chapter defined and con-
solidated the sheriff's authority, giving him a recognized
sphere of action of his own : in 12 15 " leet jurisdiction
came into existence. It was the jurisdiction left by the
Great Charter to the sheriff in his tourn," while chapter 42
of the reissue of 121 7, forbidding the tourn to be held
oftener than twice a year, marked it off " from the ordinary
civil jurisdiction of the three-weekly hundred court."
V. Local Magistrates under John, The urgent need of
restricting the authority of the sheriffs can be abun-
dantly illustrated from contemporary records. Ineffectual
attempts had, indeed, been made more than once to
restrain their evil practices, as in August, 12 13, when
directions were issued from the Council of St. Alban's
commanding the sheriffs, foresters, and others, to abstain
^ I Edward IV. c. 2.
^Contrast Coke, Second Institute, 32, who seems to suggest that one effect of
Magna Carta was to take from the sheriff a jurisdiction over thefts previously
enjoyed by him.
3 Dr. Stubbs, Const. Hist., I. 650, thinks that the Charter indicated a tendency
towards judicial absolutism, only curbed by the growth of trial by jury. Yet the
barons had no intention to enhance the royal power. The attitude of the insurgents
in 121 5 suggests rather that the sheriffs had now become instruments of royal
absolutism to a greater extent than the King's justices themselves. Edward I.,
indeed, deftly turned this chapter to his own advantage, arguing that it cancelled
all private jurisdiction over criminal pleas previously claimed by boroughs or
individuals. See Coke, Second Institute, 31, and cases there cited.
* Leet Jurisdiction^ 340.
CHAPTER TWENTY-FOUR 311
from unjust dealing,^ and, again, some two months later,
when John, at the instance of Nicholas, the papal legate,
promised to restrain their violence and illegal exactions.^
Little or nothing, however, was effected ; and Magna Carta,
in addition to condemning specified evils, contained two
general provisions : chapter 45, which indicated what type
of men should be appointed as Crown officials, and the
present chapter, which forbade local magistrates to encroach
on the province of the King's justices. These local magis-
trates are comprehensively described under four different
names. ^
(i) The Sheriff. No royal officer was more justly hated
than the sheriff. The chapter under discussion affords
strong evidence alike of his importance and of the jealousy
with which his power was viewed. A brief sketch of the
growth of the office is all that is here possible. Long before
the Conquest, in each shire of England, the interests,
financial and otherwise, of the kings of the house of
Wessex had been entrusted to an agent of their own
appointing, known as a scir-gerefa (or shire-reeve). These
officers were continued by the Norman monarchs with
increased powers, under the new name of vice comitesA
In England, during the Anglo-Saxon period, the chief
power over each group of shires had been shared among
three officers — the bishop, the earl, and the sheriff. The
bishop, by the natural differentiation of functions, soon
confined his labours to spiritual affairs; while the policy
of the Conqueror relegated the earl to a position of dignity
severed from the possession of real power. Thus the
sheriff was left without a rival within his shire. For a
period of at least one hlindred years after the Norman
^ See supra, p. 28. 2 ggg ^j^ Coventry, II. 214-5.
' Abuses by sheriffs and other bailiffs continued to be rife after 121 5 as before it.
Many later statutes afford graphic illustrations of the oppressive conduct they sought
to control. In 1275 Edward found it necessary to provide "that the sheriffs from
henceforth shall not lodge with any person, with more than five or six horses ; and
that they shall not grieve religious men nor others, by often coming and lodging,
neither at their houses nor at their manors." See Statute of Westminster, c. i,
confirmed by 28 Edward I., stat. 3, c. 13.
^Cf. supra, pp. 15-16.
312 MAGNA CARTA
Conquest, he wielded an excessive local authority as the
sole tyrant of the county. He was not indeed irresponsible,
but it was difficult for his victims to obtain the ear of
the distant King, who alone was strong enough to punish
him.
To appreciate the full authority enjoyed by a sheriff who
retained the King's confidence, we must remember the
varied nature of his powers. He was not only local magis-
trate, local tax-gatherer and local judge, but he commanded
the troops of his bailiwick. A royal favourite might have
several counties and one or more royal strongholds in his
custody. The military power of Fawkes de Breaute, for
example, must have been enormous, for it embraced the
forces of Northampton, Cambridge, Huntingdon, Bedford,
Buckingham, and Oxford.^ How powerful such men had
become is shown by their pretensions after King John's
death, when they claimed to hold their bailiwicks as matter
of right throughout his son's minority. Preposterous as
this demand seems, Henry's advisers gave effect to it,
when they confirmed the appointment of all John's sheriffs
(with the one exception of the notorious Stephen Harengod),
thus weakening the central government at a time when it
needed all its strength .^
The sheriff, however, had passed the zenith of his power
before the reign of John. That King's father had been
strong enough to show the disobedient sheriff his proper
place, as he did notably in 1170. John, however, had
his own reasons for giving a freer hand to the agents of
his evil will, foreigners and desperadoes, whose services
he rewarded in this way. This recrudescence of the
sheriff's powers must be added to the causes contributing
to the revolt of 12 15.
It has already been explained how in 1194 the sheriff's
powers were restricted. To the next year is usually traced
the origin of the justices of the peace, who gradually took
over the duties of the sheriff, until they practically super-
seded him as the ruling power in the county. In Tudor
iSee G. J. Turner, Trans. R. Hist. Soc, XVIII. 272.
2 On this whole subject see the valuable remarks of Mr. Turner, op. cit., p. 272.
CHAPTER TWENTY-FOUR 313
days a new rival appeared in the Lord Lieutenant, then
first appointed in each shire to represent the Crown in its
military capacity. The fall of the sheriff was thus gradual,
although finally complete. From presiding, as he did in
his golden age, over all the business of the district —
financial, admimstrative, military, and judicial — the sheriff
has become, in England at the present day, a mere honorary
figure-head of the county executive. A high sheriff is still
chosen annually by King George for each county by
pricking at random one name out of a list of three leading
land-owners presented to him for that purpose. He is
responsible, during his year of office, for the execution of all
writs of the superior Courts within his county, including
the execution of criminals, for returning the names of those
elected to serve in the House of Commons, and for many
other purposes; but his responsibility is chiefly theoretical.
The real duties of his office are now performed by sub-
ordinates. What really remains to him is an empty and
expensive honour, usually shunned rather than courted.
In Scotland and America, the sheriff also exists at the
present day, but his position and functions have in these
countries developed in very different directions. In Scot-
land, in opposition to what has happened in England and
America, the sheriff has remained emplxatically a judicial
officer, the judge of the local court of his shire, known as
" the Sheriff Court." He has thus retained intact his
judicial functions, to which such administrative duties as
still remain to him are subordinate. In the United States
of America, on the contrary, the sheriff is a purely executive
official, possessing perhaps more real power, but notably
less honour and social distinction, than fall to the lot of the
English high sheriff. The duties of his office are some-
times performed by him in person ; he may even set out
at the head of the posse comitatus in pursuit of criminals.
Three completely different offices have thus sprung from
the same constitutional root, and all three are still known
by one name.
(2) The constable. Portions of certain counties were
exempted from the sheriff's bailiwick. Districts afforested
314 MAGNA CARTA
were administered by wardens, assisted by verderers, who
excluded the sheriff and coroners; while royal fortresses,
together with the land immediately surrounding them,
were under command of officers known indifferently as
castellans or constables.^ The offices of warden of a par-
ticular forest and warden of an adjacent royal castle were
frequently conferred on the same individual. Indeed,
chapter i6 of the Forest Charter of Henry III. seems to
use the term " castellans " as the recognized name of forest
wardens, whom it forbids to hold " pleas of the forest."
The name constable has at different periods been applied
to officers of extremely different types. The King's High
Constable, a descendant of the horse-thegn of the Anglo-
Saxon kings, was originally the member of the royal
household responsible for the King's stables. At a later
date, he shared with the Earl Marshal the duties of Com-
mander-in-chief. The name of constable came to be
applied also to commanders of small bodies of troops,
whether in castles or elsewhere. At a later date the word
was used in connection with duties of watch and ward :
each hundred had its high constable and each village its
petty constable in the fourteenth and fifteenth centuries. ^
The name is at the present day, confined to members of
the police force.
The word, as used in Magna Carta, denoted the captain
of a royal castle.^ Such an office was one of trust; and
wide powers were conferred upon its holder. He acted
as gaoler of prisoners confided to the safe-keeping of his
dungeons. He had authority, under certain ill-defined
restrictions, to take whatever he thought necessary for
provisioning the garrison — a privilege the exercise of which
frequently led to abuses, guarded against by chapters 28
1 These localities were independent of the ordinary executive authorities of the
county ; partial exemption from the sherififs control was enjoyed also by (a)
chartered boroughs and {d) holders of franchises. The same man might, of course,
be both sheriff and castellan.
2 See H. B. Simpson, £n^. Hist. Rev.^ X. 625, for authorities.
3 Evidence collected by Coke, Second Institute^ 31, proves the identity. See
also Round, Ancient Charters^ No. 55, where Richard I. in 1 159 speaks of
* ' constabularia castelli Lincolniae. "
CHAPTER TWENTY-FOUR 315
and 29 of Magna Carta. He had also, to a limited extent,
judicial authority. Not only did he try pleas for small
debts to which Jews were parties, but he enjoyed a juris-
diction over all petty offences committed within the pre-
cincts of the castle, analogous to that of the sheriff within
the rest of the county. The power of trying and punishing
misdemeanours was not taken away by the Great Charter,
and was confirmed by implication in 1300 by a statute
which directed that the constable of Dover Castle should
not hold, within the castle gate, " foreign " pleas of the
county which did not affect "the guard of the castle."^
The Articles of 1309 complained that constables of the
King's castles took cognizance of common pleas.^ In the
reign of Henry IV. complaint was made that constables
of castles were appointed justices of the peace, and im-
prisoned in one capacity the victims they had unjustly
condemned in another. This practice was put down by
statute in 1403.^
It would appear that at an earlier period the constable
of the hundred sometimes acted as deputy-sheriff. Chapter
12 of the Assize of Northampton provided that when the
sheriff was absent the nearest castellanus might take his
place in dealing with a thief who had been arrested. His
interference outside his own precincts must, however, have
been regarded with great jealousy, and the coroners, after
their appointment in 1194, would naturally act as sub-
stitutes during the sheriff's absence.
(3) The coroners. The coroners of each county, after
their institution in i,^, seem to have shared with the
sheriff most of the powers of which the latter had previously
enjoyed a monopoly. They were appointed by the whole
body of freeholders assembled in the county court,* and
the nature of their duties is explained by the oath of office
^ See Articuli stiper cartas, 28 Edward I. c. 7.
2Stubbs, CoTist. Hist., II. 339.
^ See 5 Henry IV. c. 10. Coke, Second Institute, 30, relates, as an indication
of the authority and pretensions of these constables, that they had seals of their
own "with their portraiture on horseback."
* See Stubbs, Hoveden, Pref. to Vol. IV. xcix.
3i6 MAGNA CARTA
sworn in the same words for many centuries, " ad custodi-
enda ea quae pertinent ad coro7iam." Their duty was to
guard royal interests generally ; and their " keeping " of
royal pleas was merely one aspect of this wider function.
Besides " attaching " those suspected of crimes — that is,
receiving formal accusations and taking such sureties as
might be necessary, it was their duty to make preliminary
investigations; to examine the size and nature of the
victim's wounds in a charge of mayhem; ^ and to keep a
watchful eye on royal windfalls, including deodands,
wrecks, and treasure-trove. They had also to appraise
the value of chattels forfeited to the King. When felons
took refuge in sanctuary, it was the coroner who arranged
for their leaving the country on forfeiting all that they had.
They kept a record of those who had been outlawed, and
received " appeals " of criminal charges.^
Magna Carta forbade the coroner to determine pleas of
the Crown; but, even after 12 15, he sometimes did justice
upon felons caught red-handed. An act of Edward I.^
accurately defined his duties, empowering him to attach
pleas of the Crown and to present criminals for trial, but
forbidding him to proceed further alone.
The coroner's functions, originally so wide and varied,
have been gradually narrowed down, until at the present
day the duty usually associated with his office is the hold-
ing of inquests on dead bodies where there are suspicious
circumstances.^ He is still responsible for treasure-trove
and he is also competent to act as the sheriff's substitute
in case of illness or absence during the year of office.
(4) The bailiffs. The mention by name of three classes
of local officers is supplemented by the addition of an
indefinite word sufficiently wide to cover all grades of
^ See Bracton, f. 122b.
2 In 1 197 Richard's Assize of Measures appointed six custodientes in each county
and town. These were coroners over one class of offences, the use of false weights
and measures. Cf. infra^ under c. 35.
^ Statute of Westminster, I. c. 10.
^Cf. Coke, Second Institute, 31, "In case when any man come to violent or
untimely death, super visum corporis^
CHAPTER TWENTY-FOUR 317
Crown officials. The term " bailiff " may be applied to
every individual to whom authority of any sort has been
delegated by another. It would include the men who
actually served writs, or distrained the goods of debtors;
and also generally all local officials of every description,
holding authority directly or indirectly from the Crown.
The district over which his office extended was called his
"bailiwick," a term often applied to the county considered
as the sphere of the sheriff's labours.^
CHAPTER TWENTY-FIVE.
Omnes comitatus, hundrede, wapentakii, et trethingii, sint
ad antiquas firmas absque ullo incremento, exceptis domi-
nicis maneriis nostris.
All counties, hundreds, wapentakes, and trithings (except our
demesne manors) shall remain at the old rents, and without any
additional payment.
This provision, directed against the sheriffs, shows a
determination to get to the root of the disease, instead of
merely attacking the symptoms. The rents at which the
counties (or parts of them) were farmed out to the sheriffs
must no longer be arbitrarily raised, but were to remain at
the old figures which had become stereotyped from long
usage. To understand how such increases would injuri-
ously affect the inhabitants of the county, some explanation
is necessary. Centuries before the Norman Conquest,
England had been already mapped out into shires on lines
substantially the same as those which still exist. Each
county had been subdivided into smaller districts known as
" hundreds " in the south, and as " wapentakes " in the
Danish districts of the north ; while intermediate divisions
existed, exceptionally, in some of the large counties such as
York and Lincoln, each of which had three " trithings " or
ridings.
^Mr. G. J. Turner, speaking of the minority of Henry III., thinks "the term
'bailiff' as applied to a county at this period meant 'sheriff.'" Transactions^
P- 274.
3i8 MAGNA CARTA
In commenting upon chapter 24, it has been explained
how the Anglo-Saxon Kings entrusted their interests in
each shire to an officer called a sheriff, and how a similar
officer under the Norman Kings became the chief magis-
trate in the county. His financial duties, however, long
remained the most important. Even before 1066, the sheriff
had ceased to be a mere intermediary, who lifted the King's
rents and paid over, pound by pound, the yearly varying
sums he might receive. He had become a firmarius, buy-
ing for a yearly rent the right to appropriate to his own uses
the revenues of the county. The Crown got the exact sum
stipulated for, known as the firma comitatus ; while the
balance, if any, remained with the sheriff. In plain words,
the sheriff speculated in the returns : it was his business,
by fair means or foul, to make sure of a handsome
surplus.
Authorities differ as to the exact list of items purchased
by the firma comitatus ; but the two chief sources of revenue
were the profits of justice in the local courts, and the rents
of royal manors.
William I. sharply raised the farms, and his successors
endeavoured, whenever possible, to increase them further.
Now, it might seem at first sight that these additional
burdens concerned exclusively the Crown and the sheriff,
but such was by no means the case. The sheriff took care
to pass on the burden to the shoulders of those subject to his
authority. His rule tended always to be oppressive, but
his unjust exactions would be doubled when the amount of
the firma had recently been raised.
Under the vigilant rule of Henry II., some measure of
relief was obtained by the shires from the misdeeds of their
local tyrants, since that far-seeing King knew that his own
best interests called for curtailment of the sheriffs' preten-
sions. He punished their excesses and deprived them of
office. John, on the contrary, appointed men of a less
reputable type, and gave them rope. In return, he wrung
more money from them. Not content with exacting the
annual firma and the additional sum known as " increment,"
which had now become stereotyped as a fixed and recog-
CHAPTER TWENTY-FIVE 319
nized payment,^ John from 1207 onwards exacted a third
payment under name of proficuum, and allowed his sheriffs
to inflict new severities to recoup themselves for their addi-
tional outlay.
Magna Carta made no attempt to abolish the practice of
farming out the shires, but forbade alike the increase of the
farm and the exaction of proficuum. The barons here made
an innovation which was unfair to John. If it benefited the
men of the counties in dealing with their sheriffs, it gave
the sheriffs an undeserved advantage over the exchequer.
The total value of the various assets included in the. firma
comitatus had greatly increased in the past, and would
probably continue to increase in the future. Therefore, it
was unfair to bind the Crown by a hard-and-fast rule which
would practically make a present of this future " unearned
increment " to the sheriff. To stereotype the firma was to
rob the Crown, which required increased revenues to meet
the increased cost of its expanding duties.^
Although this chapter was omitted from all reissues, the
Crown, during Henry III.'s minority, forbore to exact the
proficuum, reverting to the practice prior to the seventh
year of his father's reign. After he had been declared of
age, however, increased sums were again taken. ^ There
was, indeed, no valid reason why the unearned increment
should go to the sheriff rather than to the King : it was
sufficient to provide against the fixing of the amounts too
high. The Articuli super cartas, accordingly, while con-
ceding to the counties the right of electing their own
sheriffs, declared that neither the bailiwicks and hundreds
^ These extra payments appear under various names, e.g. augmentum or incre-
mentum in Domesday Book (cf. Ballard, Domesday Inquest, 75). The Pipe Roll
for 1 166 (p. 11) records 200 marks paid zs, gersuma for Norfolk and Suffolk. See
evidence collected by Adams, Origin, 237 n. Huge sums were sometimes paid :
Archbishop Geoffrey in 1 194 purchased the shrievalty of York for ;!^2000. Ramsay,
Angevin England, 345.
2 Cf. Sir James Ramsay, Angevin Empire, 476, who describes this provision as
*'an impossible requirement." Dr. Stubbs' paraphrase is not entirely happy:
"the ferms of the counties and other jurisdictions are not to be increased." See
Const. Hist., I. 575.
3 See Turner, Trans. R.H.S., XVIII. 289.
320 MAGNA CARTA
of the King, nor those of great lords ought to be put to farm
at too high rates. The evil, however, continued under a
new form ; sheriffs, while only paying a moderate farm
themselves, sublet parts of their province at much higher
rates, thus appropriating the increment denied to the
exchequer, while the bailiffs who had paid the increase
could not " levy the said ferm without doing extortion and
duress to the people."^ Three successive acts prohibited
this practice, declaring that hundreds and wapentakes must
either be kept in the sheriff's own hands, or sublet, if at all,
at the old fixed farms only.^
One exception to the scope of its own provisions was
deliberately made by Magna Carta — an exception of an
important and notable nature; the demesne manors of the
Crown were left exposed to arbitrary increases of their
annual rents. Now, the chief items contained in the firma
were, as already explained, the rents of these manors and
the profits of the local courts. It would thus appear, in the
light of this exception, that the aim of Magna Carta was to
prevent an increase under the second head — to prevent, that
is, the local courts being made the instruments of extortion ;
and this apparently was the precise object of chapter 42 of
the reissue of 12 17.
That chapter struck at one of the most fertile of the
sheriffs' expedients for swelling the profits of their office.
It was their practice to summon the various district courts
with unnecessary frequency and at inconvenient times and
places, fining every suitor who failed to attend. The
Charter of 121 7 reaffirmed the ancient usage ^ : no county
court should meet in future oftener than once a month ; no
sheriff or bailiff should make his " tourn " through th(
hundreds oftener than twice a year, to wit at Easter an(
Michaelmas, and that only at the accustomed place; vie\
of frankpledge should only be held once a year at Michael-
mas, and the sheriff must not seek " occasions," but content
^ These are the words of the statute of 1 330, cited below.
-4 Edward III. c. 15 ; 14 Edward III. c. 9 ; 4 Henry IV. c. 5.
^For this usage see Cnut, II. 18 (Liebermann, Gesetze^ I. 321) ; Leges Henricii
7 and 8 [ibid., 553) ; Writ of Plenry I. {ibid.^ 524).
CHAPTER TWENTY-FIVE T\ 321T
I
himself with what he was wont to get for taking his " vie^^*?^
under Henry II.; all liberties must be respected; and any
district in which the courts meet by custom less frequently
than is normal, shall have the benefit of such exceptional
local usage. ^
In a curious case^ that came before the justices in 1226,
this clause was pleaded as a defence against a charge of
impeding the sheriff of Lincoln in the performance of his
duties of holding "counties," " thethings " (or courts of
ridings), and wapentakes : the sheriff, against custom, was
holding county courts oftener than once in five weeks and
for more than one day at a time, and was holding a wapen-
take in Ancaster oftener than twice a year, and not according
to the charter of liberty.
In another plea (1231)^ juries testify that since the
making of " carta de Runemede " (here evidently used for
the Charter of 12 17) the sheriff has come into the hundred
twice instead of once a year (as the old custom was) to take
view of frankpledge and to make attachments of pleas of the
Crown.
After 12 1 7, in absence of express royal gf*ant or prescrip-
tion to the contrary, the rule formulated in Henry's second
reissue of Magna Carta fixed the times of holding the
" tourn " of the sheriff, and this was extended also to the
" leet " jurisdiction, which in the liberties took the place of
the tourn."*
CHAPTER TWENTY-SIX.
Si aliquis tenens de nobis laicum feodum moriatur, et
vicecomes vel ballivus noster ostendat litteras nostras
patentes de summonicione nostra de debito quod defunctus
nobis debuit, liceat vicecomiti vel ballivo nostro attachiare,
et inbreviare catalla defuncti, inventa in laico feodo, ad
1 See supra, p. 150. ^Bracton's Note-book^ Plea 1730.
3 Ibid., No. 513.
*See Hearnshaw, Leet Jurisdiction^ 79, 80, who reminds us, however (p. 147),
that "even Magna Carta can be prescribed against."
X
322 MAGNA CARTA
valenciam illius debiti, per visum legalium hominum, ita
tamen quod nichil inde amoveatur, donee persolvatur nobis
debitum quod clarum fuerit ; et residuum relinquatur execu-
toribus ad faciendum testamentum defuncti; et, si nichil
nobis debeatur ab ipso, omnia catalla cedant defuncto, salvis
uxori ipsius et pueris racionabilibus partibus suis.
If any one holding of us a lay fief shall die, and our sheriff or
bailiff shall exhibit our letters patent of summons for a debt
which the deceased owed to us, it shall be lawful for our sheriff
or bailiff to attach and catalogue chattels of the deceased, found
upon the lay fief, to the value of that debt, at the sight of law-
worthy men, provided always that nothing whatever be thence
removed until the debt which is evident ^ shall be fully paid to
us ; and the residue shall be left to the executors to fulfil the
will of the deceased ; and if there be nothing due from him to
us, all the chattels shall go to the deceased, saving to his wife
and children their reasonable shares.
The primary object of this chapter was to regulate the
procedure to be followed in attaching the personal estates of
Crown tenants who were also Crown debtors. Incidentally,,
it throws light on the right of bequeathing property.
I. Nature of the Grievance. When a Crown tenant died
it was almost certain that arrears of scutages, incidents, or
other exactions remained unpaid. The sheriff and bailiffs
of the district, where deceased's estates lay, were in the habit
of seizing everything they could find on his manors, under
excuse of securing the interests of their royal master. They
attached and sold chattels out of all proportion to the sum
actually due. A surplus would often remain in the sheriff's
hands, which he refused to disgorge.
Magna Carta sought to make such irregularities impos-
sible, by defining the procedure to be followed. The sheriff
and his bailiffs were forbidden to touch a single chattel of a
deceased Crown tenant, unless they came armed with legal
warrant in the form of royal letters patent vouching the
existence and the amount of the Crown debt. The officers
of the law were allowed to attach only as many chattels as
might reasonably be expected to satisfy the debt due to the
exchequer; and everything so taken must be carefully
^ Cf. the use of the phrase " a Hquid debt" in Scots law.
CHAPTER TWENTY-SIX 323
inventoried. All this was to be done " at the sight of lawful
men," respectable, if humble, neighbours specially sum-
moned for that purpose, whose function it was to form a
check on the actions of the sheriff's officers, to prevent them
from appropriating anything not included in the inventory,
to assist in valuing each article and to see that no more
chattels were distrained than necessary. A saving clause
protected the interests of the Crown by forbidding the
removal from the tenant's fief of any chattels, even those
not so attached, until the full ascertained amount had been
paid to the exchequer. Not till then could a division take
place among the deceased man's relatives or legatees.
These provisions should be read in connection with
chapter 9,^ which provided that diligence for Crown debts
must proceed against personal estate before the debtor's
freehold was distrained, and laid down other equitable rules
applicable alike to a deceased Crown debtor and to a living
one.
II. The Right to Bequeath. The main interest of this
chapter, for the historian of law and institutions, lies in a
different direction ; in the light thrown on the right of
making Wills in 12 15. The early law of England had
difficulty in deciding how far it ought to acknowledge the
claims made by owners of property, both real and personal,
to direct its destination after death. Various influences
were at work, prior to the Norman Conquest, to make the
development of this branch of law illogical and capricious.^
Of the law of bequests in the twelfth century it is possible
to speak with greater certainty ; definite principles had by
that time received recognition. All testamentary rights
over land or other real estate were then denied, not, as has
sometimes been maintained, in the interests of the feudal
lord, but rather of the expectant heir.^ Many reasons con-
tributed to this result. For one thing, it had become neces-
sary to prevent churchmen from using their spiritual
1 Cf. what is there said of the sheriff's oppressions.
'The subject is discussed by Pollock and Maitland, II. 312-353. See also
Holdsworth, III. 418 ff. ; Makower, ConsL Hist. Church, 427 ff.
3 See Pollock and INTaitland, II. 324.
324 MAGNA CARTA
influence to wring bequests from dying men to the impover-
ishment of the heir. Churchmen, in compensation as it
were for the obstacles thus opposed to their thirst for the
land of the dying, ultimately, but not before the reign of
Henry III., made good their claim to regulate all Wills
dealing with personal estate; that is money, goods, and
chattels.
Under Henry II. no such right had been admitted. The
Assize of Northampton (1176) directed that heirs should
divide the chattels according to the provisions made in the
Will, without any reference to the supervision of the bishop
or his clergy. Glanvill twice gives a writ directing the
sheriff to uphold the Will of a testator ; but no trace of any
similar writ appears in the Registers of the early years of
Henry HI.: "the state has had to retreat before the
church."!
This victory of the ecclesiastical courts was probably won
shortly after 12 15. John's Charter makes no admission of
any right of the church in the " proving " of Wills ; but it
does admit (in chapter 27) the church's right to " superin-
tend" the division of the goods of intestates, an insidious
privilege, which was used as a lever during the minority
of Henry (a ward of Holy See), and thus helped to give
the courts Christian an excuse for deciding also as to
the validity of Wills. It was apparently in John's reign
that the practice of appointing executors to carry out the
Will of the deceased became general. Henry II. in his
own case had entrusted this duty to individuals whom he
named, but did not describe as " executors," a word, how-
ever, used in its technical sense in King John's Will.^
John claimed that his subjects could not make valid
Wills without his consent, which had, as usual, to be paid
for. Such, at least, is the inference to be drawn from the
existence of writs granting licences to make a Will, or
confirming one that had been made.^ The King's interfer-
iMaitland, CoIL Papers, II. 139. 2Holdsworth, III. 418 ff.
^On 30th August, 1 199 {New Ry77ier, I. 78) John confirmed the testament of
Archbishop Hubert Walter; and on 22nd July, 1202 {ibid., I. 86), he granted
permission to his mother, the dowager Queen Eleanor, to make a will.
CHAPTER TWENTY-SIX 325
ence in this province seems to have been regarded as an
illegal encroachment.
Magna Carta declares that all the chattels (or the residue
after paying Crown debts) " shall go to the deceased " for
"the executors to fulfil the will of the deceased," but
immediately adds the saving clause, that " all the chattels "
means only what remains after deducting the " reasonable
shares " of wife and children. This seems to exclude, by
implication, the King's right to interfere on the plea that
he had not licensed a Will, while it keeps alive an ancient
rule that a testator could only dispose of part of his pecunia
(or personal estate), his widow and children having absolute
claims to the rest.
The Charter did not define these " reasonable shares " ;
but custom had already fixed them at the same proportions
of the whole as the law of Scotland observes at the present
day. When a Scotsman dies, leaving wife and children,
his movable or personal estate falls into three equal parts,
known respectively as the widow's part (or jus relictae), the
" bairns' " part (or legitim, the legitima portio of the Roman
law), and " the dead's part." ^ If he attempts to dispose of
his entire estate, wife and children may claim their legal
rights, and " break the Will." Where a wife survives, but
no children (or a child and no wife), the division is into two
equal portions. Magna Carta recognizes a similar divi-
sion ; and we know from Glanvill that, if the dead man's
Will had attempted to defeat the just claims of wife or
children, the writ de rationahili parte honorum would give
them relief.^
The conception of a " dead's part " or portion to be dis-
pensed in charity and good works for the benefit of the
deceased's soul was, of course, in great measure due to the
influence of the church, which was not unwilling to stimu-
late the belief that one of the best methods of affecting this
was to leave money to itself. Under Henry III. the bishop
of the diocese made good his claim to " prove " Wills (that
^ Cf. " tota pars sua de pecunia sua " in Burton Abbey Surveys (cited by Round,
Engl. Hist. Rev. , XX. 279) ; Bateson, Borough Customs, II. xcvi.
2 Glanvill, VII. 7.
326 MAGNA CARTA
is to determine whether they were vaHd), and to control the
" executors " in carrying out the dead man's instructions.
Where the testator's intentions were ambiguous, the
" ordinary " would see to it that deceased's soul did not
suffer through giving too little to the church.
The reissue of 1216 makes no alteration on this chapter
of John's Charter : that of 1217 omits " et pueris," probably
through a clerical blunder, for the words were restored in
1225. As mere disuse does not abrogate an English statute,
this provision remained in force until repealed by implica-
tion by the Wills Act of 1837.^
Long subsequent to the thirteenth century, the laws of
England and Scotland as to the rights of succession of wife
and children seem to have remained identical : but, while
Scots law is unaltered to the present day, English law has,
by slow steps, the details of which are obscure, entirely
changed. The rule that acknowledged the children's right
to one third of the personal estate was gradually relaxed,
while the testator became sole judge what provision he
ought to make for his sons, until at last a purely nominal
sum of money was all that was required. The law of Eng-
land, at the present day, does not compel a father to leave
son or daughter even the proverbial shilling. The phrase
" to cut off a son with a shilling," which still lives in popular
usage, may perpetuate the tradition of an intermediate stage
of English law, where some provision, however inadequate,
had to be made, if the Will was to be allowed to stand.
CHAPTER TWENTY-SEVEN.
Si aliquis liber homo intestatus decesserit, catalla sua per
manus propinquorum parentum et amicorum suorum, per
visum ecclesie distribuantur, salvis unicuique debitis que
defunctus ei debebat.
If any freeman shall die intestate, his chattels shall be dis-
tributed by the hands of his nearest kinsfolk and friends, under
supervision of the church, saving to every one the debts which
the deceased owed to him.
^ See Z«w/ J/o^as/Wif, Oct. 1905.
CHAPTER TWENTY-SEVEN 327
Here the Great Charter proceeds to remedy an evil con-
nected with intestate succession, a natural sequel to the
subject of testate succession.^ In light of subsequent
history, the words most worthy of notice are probably " per
visum ecclesiae," which appear also in the Articles of the
Barons. There were good reasons for welcoming the inter-
vention of the bishop's court as a substitute for the scramble
that often took place for an intestate's chattels; but the
jurisdiction thus gained by the church was quickly put to
other uses.
The placing of this powerful weapon in the hands of the
church was only incidental to the main purpose of this
chapter. This was (while safeguarding the interests of
creditors) to secure to the deceased's kinsmen and friends
the right to make an equitable division of his chattels. By
implication the Charter says " hands off " to John, and
indeed to any lord superior, whether the King or another.
In the Middle Ages all classes of men, good and bad
alike, exhibited an extreme horror of dying intestate. ^
Several causes contributed towards this frame of mind.
Churchmen, from motives not unmixed, inculcated the
belief that a dying man's duty was to leave part at least of
his personal estate for religious and charitable objectsJ
The bishop or priest, who had power to withhold extreme
unction from dying men, was in a strong position to force
advice upon penitents who believed the church to hold the
keys of heaven. Motives of a more worldly nature lent
their weight. If a man died intestate, his lord seized his
chattels. Henry I. in his Coronation Charter renounced
this right over Crown tenants under certain circumstances :
if a baron or " man," cut off by war or infirmity (the words
have a grudging, hesitating sound), had given no instruc-
tions for disposal of his pecunia, his wife and children and
legitimi homines (or vassals) should divide it " for his soul "
as seemed best to them.^ Stephen, in his second or Oxford
iQn whole subject, see Holdsworth, III. 418 ff. ; Makower, Const. Hist,
Church, 427 ff.
- Pollock and Maitland, II. 354.
^ See Appendix and supra, p. 98. Also Bateson, Borough Customs, II. cxlii-iii.
Cf. Cnut, II. cc. 70 and 78 (Liebermann, Gesetse, 357-365).
328 MAGNA CARTA
Charter,^ gave up all such claims, as regards the property of
prelates and clerks, who were confirmed in their rights of
making Wills.
These promises were not kept : in Glanvill's day, the
King, like other feudal lords, appropriated the goods of
intestates.2 Henry II. continued to treat intestacy, especi-
ally in the case of clerks, as an excuse for forfeiture.^
Magna Carta contained a clear pronouncement against this
practice. The kinsmen and friends of the deceased, with-
out royal interference, were to divide the chattels under
supervision of the church : the King's courts were excluded.
No scheme of intestate succession was set forth ; but where
wife and children survived, the tripartite division was
clearly implied. In the distribution of the dead's part, the
prelates allowed themselves liberal discretion : something
went to the poor, but more might be spent on masses, while
a portion would be retained as remuneration for trouble
expended.
In 1 216 this provision of John's Charter was withdrawn.
Why? Had a suspicion crossed the mind of William
Marshal that it conferred a dangerously elastic privilege
upon the church ? Did the legate Gualo refuse to trust the
English prelates with authority? Did the young King's
1 See Appendix and supra, p. 102. ^ Glanvill, VII. 16.
^See Pollock and Maitland, II. 354. Examples are readily found: "When
Archbishop Roger of York died in 1 182, Henry II. enjoyed a windfall of ;i^l 1,000,
to say nothing of the spoons and saltcellars " (Pollock and Maitland, I. 504).
Royal prerogatives in the twelfth century were elastic. Henry II. used them
freely, but on the whole fairly. His sons stretched every doubtful claim to its
utmost limits. The Crown was the legal heir of all Jews (cf. c. 10) and apparently
of all Christian usurers as well, at least of such as died unrepentant (see Pollock
and Maitland, II. 486), and the making of a will was a necessary condition of a
usurer's repentance. (See Dialogus de Scaccario, 224-5 i^"^-) The King, further,
took the goods of all who died a felon's death (cf. c. 32) and of men who committed
suicide (itself a felony). Madox (I. 346) cites an entry from the Pipe Rolls of
1 1 72, recording 60 marks due to the exchequer as the value of the chattels of an
intestate ; and, two years later, mention is made de pecunia Gilleberti qui obiit
intestatus. There is nothing to show whether such men were, or were not,
usurers. The Pope was another competitor for the personal estates of intestate
clerks. In 1246 he issued an edict making this demand: even Henry III.
(dependent and ally of Rome as he was) protested, and the edict was withdrawn.
See Pollock and Maitland, II. 357.
CHAPTER TWENTY-SEVEN 329
advisers, conscious of their urgent need of money, deter-
mine to reserve what rights the indefinite earHer law allowed
them of taking part in the scramble for the coin and chattels
of intestates ?
Irregularities continued during Henry's reign : Bracton ^
thought it necessary to urge that intestacy was not a crime.
But his direct condemnation of the feudal lords' practice
of seizing chattels is confined to cases of sudden death.
Yet it was neither King nor barons, but the church that
triumphed : the rule, enunciated in John's Charter, though
omitted from all reissues, settled the practice of later years. ^
The personal estate of intestates was administered " under
supervision of the church," and the same supervision was
ultimately extended over the Wills and estates of men who
had died testate.
CHAPTER TWENTY-EIGHT.
NuLLUS constabularius, vel alius ballivus noster, capiat
blada vel alia catalla alicujus, nisi statim inde reddat
denarios, aut respectum inde habere possit de voluntate
venditoris.
No constable or other bailiff of ours shall take corn or other
provisions from any one without immediately tendering money
therefor, unless he can have postponement thereof by permission
of the seller.
This chapter is the first of several that redressed abuses
springing from the exercise of the royal right of purvey-
ance.
I. Purveyance in General. The Norman and Angevin
Kings of England were compelled by their administrative
duties and induced by the pleasures of the chase to move
constantly from district to district. The difficulties must
have been great of finding sufficient food for the retinues
surrounding the King in peace or war. It was to the
^F. 60. 2 Pollock and Maitland, II. 355. Cf. snp-a, p. 324.
330 MAGNA CARTA
interests of the community that the work of government
should not be brought to a stand-still for want of supplies.
No opposition was made when the King arrogated to him-
self the privilege of appropriating, under fair conditions,
the necessaries his household might require. Such a right,
not unlike that enjoyed in modern times by the commander
of an army encamped in an enemy's country, was allowed
to the Kings of England in their own land in time of peace.
This was known as purveyance.^ Unfortunately, the condi-
tions under which supplies might be requisitioned were left
vague : the privilege was subject to abuse. In theory it was
a right of pre-emption ; the provisions seized were to be
paid for at the market rate : but practice tended to differ
lamentably from theory. In the absence of a neutral
arbitrator to fix the value of the goods, the unfortunate seller
was thankful to accept any pittance offered by royal officials,
who might subsequently, indeed, charge a higher rate
against the Crown. Payment was often indefinitely delayed
or made not in coin but in exchequer tallies, " a vexatious
anticipation of taxation," since these could only be used in
payment of Crown dues.
Magna Carta did not abolish purveyance, and placed no
restrictions upon its use for the legitimate purpose of
supplying the King's household. Some slight attempt to
control its exercise was made sixty years later in the Statute
of Westminster I.; but without producing much effect.-
The Articles of 1309 ^ complained that the King's purveyors
took great quantities of corn, malt, and meat without pay-
ing even by exchequer tallies. The grievances connected
with purveyance continued, throughout four centuries, as a
fertile source of vexation to the people and of friction
between parliament and the King. An attempt, made by
the House of Commons to induce James I. to surrender this
prerogative for a money grant, ended in failure, with the
abandonment of the abortive treaty known as " the Great
Contract." In the general re-settlement of the revenue,
liowever, at the Restoration, purveyance and pre-emption,
^ See Blackstone, Commentaries^ I. 287, for an often-quoted definition.
«3 Edward I. c. 32. ^Stubbs, Const, Hist., II. 339.
CHAPTER TWENTY-EIGHT 33i
which had fallen into disuse during the Commonwealth,
were abolished.^ Yet in the following year a new statute ^
virtually revived one branch of the right under essential
modifications : when royal progresses were necessary in the
future, warrants might be issued from the Board of Green
Cloth, authorizing the King to use such carts and carriages
as he might require, at a fair rate of hire specified in the Act
of Parliament.
II. Branches of Ptirveyance restricted by Magna Carta.
A practice tolerated because of its absolute necessity, when
confined to providing for the needs of the King's household,
became intolerable when claimed by every castle-warden,
sheriff, and local bailiff, for his own personal or official
needs. Discretionary authority was vested by John in a
class of officials least qualified to use it, unscrupulous
foreign adventurers hired to intimidate the native popula-
tion, responsible to no one save the King, and careful never
to issue from their strongholds except at the head of their
reckless soldiery. The Great Charter contained a few
moderate provisions for checking the abuses of purvey-
ance.
(i) Provisioning of castles. Commanders of fortresses
were left free by Magna Carta to help themselves to such
corn and other supplies as they deemed necessary for their
garrisons. Immediate payment, however, must be made in
current coin (not in exchequer tallies) for everything they
requisitioned, unless the owner consented to postpone the
date of payment. The Charter of 1216 made a slight modi-
fication in favour of castellans. Payment for goods taken
from the town where the castle was situated might be legally
delayed for three weeks, a term extended in 12 17 to forty
days. Such relaxation was perhaps necessary to meet the
case of a warden with an empty purse called on to provide
against an unexpected siege or other emergency ; but the
peaceful townsmen, over whose dwellings the dark walls of
a feudal stronghold loomed, would not dare to press unduly
for payment. Under Henry's Charters, as under that of
John, immediate payment had to be tendered to owners
1 12 Charles II. c. 24, ss. II-12. * 13 Charles II. c. 8.
332 MAGNA CARTA
who lived elsewhere than in this neighbouring town.^
(2) Requisitioning horses and carts. The provisions of
chapter 30, modified in subsequent reissues, sought to pro-
hibit sheriffs from commandeering wagons that were the
property of freemen. (3) Appropriation of timber. The
succeeding chapter confined the King and his officers to
the use of such wood as they could obtain from the royal
demesnes.^
III. Branches of Purveyance not mentioned in Magna
Carta. A wide field was left alike for the use and the abuse
of this prerogative, after due effect had been given to these
moderate provisions. Two minor aspects of purveyance
came into prominence in later history.
(i) Requisition of forced labour. Hallam explains how
the King's rights of pre-emption were extended, by
analogy, to his subjects' labour. "Thus Edward III.
announces to all sheriffs that William of Walsingham had
a commission to collect as many painters as might suffice for
'our works in St. Stephen's chapel, Westminster, to be at
our wages as long as shall be necessary ' ; and to arrest and
keep in prison all who should refuse or be refractory ; and
enjoins them to lend assistance. Windsor Castle owes its
massive magnificence to labourers impressed from every
part of the kingdom. There is even a commission from
Edward IV. to take as many workmen in gold as were
wanted, and employ them at the King's cost upon the
trappings of himself and his household." ^ Perhaps, how-
ever, such demands did not form a legal branch of
purveyance, but were merely instances of illegal royal
encroachments.
(2) Billeting of soldiers in private houses. This practice,
^ The Statute of Westminster I. (3 Edward I. c. 7) enacted " that no constable
or castellan from henceforth take any prise or like thing of any other than of such
as be of their own town or castle, and that it be paid or else agreement made
within forty days, if it be not ancient prise due to the king, or the castle, or the
lord of the castle," and further (c. 32) that purveyors taking goods for the King's
use, or for a garrison, and appropriating the price received therefor from the
exchequer, should be liable in double payment and to imprisonment during the
King's pleasure.
2 For details, see under cc. 30 and 31. ^jj^llam. Middle A^es, III. 221.
CHAPTER TWENTY-EIGHT 333
which may be considered a branch of purveyance, has
always been pecuHarly abhorrent to pubHc opinion in Eng-
land. It is as old as the reign of John ; for, when that King
visited York in 1201, he complained bitterly that the citizens
neither came out to meet him nor provided for the wants of
his crossbow-men. His threats and demands for hostages
were with difficulty turned aside by a money payment of
;^ 1 00.1 Charles I. made an oppressive use of this preroga-
tive, punishing householders who refused to pay illegal
taxes by quartering his dissolute soldiery upon them, a
practice branded as illegal by the Petition of Right in
1628.2
CHAPTER TWENTY-NINE.
NuLLUS constabularius distringat aliquem militem ad
dandum denarios pro custodia castri, si facere voluerit
custodiam illam in propria persona sua, vel per alium
probum hominem, si ipse eam facere ntn possit propter
racionabilem causam ; et si nos duxerimus vel miserimus
eum in exercitum, erit quietus de custodia, secundum
quantitatem temporis quo per nos fuerit in exercitu.
No constable shall compel any knight to give money in lieu
of castle-guard, when he is willing to perform it in his own
person, or (if he himself cannot do it from any reasonable
cause) then by another responsible man. Further, if we have
led or sent him upon military service, he shall be relieved from
guard in proportion to the time during which he has been on
service because of us. v^
Castle-guard, or the liability to serve in the garrison of a
royal fortress, formed part of the feudal obligations of the
owners of certain freeholds. This service was sometimes
due in lieu of attendance in the army; more usually the
tenant who owed garrison duty owed knight's service as
well. 3 It was probably this duplication of duties that pre-
^ See Rotuli de oblatis etjinihtts, 119. ^See 3 Charles I. c. I.
3 See the examples collected in Pollock and Maitland, I. 257. See also in
Rotuli de oblatis et finibus^ 107, how in 1200 Ralph de Bradel offered John
40 marks and a palfrey to be relieved of " the custody of the work of the castle of
Grimsby."
334 MAGNA CARTA
vented castle-guard from hardening into a separate tenure.^
John preferred to commute personal service of castle-guard
for money payments (analogous to the scutage paid in lieu
of knight's service), and to man his feudal towers with
soldiers of fortune rather than with rebellious Englishmen.
Captains of royal castles were, therefore, in the habit of
demanding money even from those who offered personal
service. What was worse, when the freeholder followed
John on distant service, he was mulcted in a money pay-
ment because he had not stayed at home to perform garrison
duty during the same period. Both abuses were forbidden
in 1 2 15.2 In certain circumstances, however, this prohibi-
tion would have deprived the King of what was equitably
due to him. Suppose he had granted two fiefs to the same
tenant — one by simple knight's service, the other by castle-
ward. A double holding implied double service; the
tenant could not in fairness plead that the service of one
knight, rendered abroad, operated as the full discharge of
the services of two knights due from his two separate fiefs.
Castle-guard must in such a case be performed by an
efficient deputy, or else the usual compensation be paid.
The reissue of 12 17 amended John's Charter to this effect.
Service with the army abroad operated as a discharge of
castle-guard at home, but not where the tenant owed two
services for two distinct fiefs.^
CHAPTER THIRTY.
NuLLUS vicecomes, vel ballivus noster, vel aliquis alius,
capiat equos vel carectas alicujus liberi hominis pro cariagio
faciendo, nisi de voluntate ipsius liberi hominis.
No sheriff or bailiff of ours, or other person, shall take the
horses or carts of any freeman for transport duty, against the
will of the said freeman.
^ Cf. stipra, p. 57 n.
2 Adams, Origin^ 238, contrasts the principle of this chapter with that of c. 12,
where no option is allowed the vassal of offering service in lieu of scutage — a
breach of strict feudal custom.
"^ Defeodo pro quo fecit servicimn in exe^-citti. This variation in the charter of
121 7 seems to have escaped Dr. Stubbs' attention. See Select Charters ^ 346.
CHAPTER THIRTY 335
The Charter here returns to the subject of purveyance,
one branch of which is practically abolished, except as
affecting villeins. No carts or horses belonging to a free-
man were to be requisitioned by any sheriff or bailiff for the
King's use without the owner's consent ; that is to say, they
could not be requisitioned at all. Protection, however, was
limited to freemen; the inference is that the horses and
implements of villeins were left at the disposal of the Crown.
The relative chapter of the reissue of 1216 partially restored
this branch of purveyance ; consent of the owner, even when
a freeman, need not be obtained, provided hire was paid at
rates that were fixed : lod. per diem for a cart with two
horses, is. 2d. for one with three. ^ The prerogative,
though restored, was not to be abused.
In 12 17 it was again slightly restricted in favour of the
upper classes. No demesne cart of any " parson " (ecclesi-
astica persona), or knight, or lady, could be requisitioned by
the bailiffs. The " demesne " carts were, of course, those
that belonged to the owner of the manor sLy opposed to the
carts of the villeins : the rights of villeins, if they had any,
must not stand against the rights of the Crown. Yeomen
and small freeholders were also left exposed to this annoy-
ing form of interference. Abuses continued. Purveyors
would lay hands on all horses and carts in the countryside —
far more than they required — choosing perhaps the season
of harvest or some equally busy time. The owners, who
urgently required them for their own purposes, had to pay
ransom to regain possession. Edward I. enacted that per-
petrators of such deeds should be "grievously punished by
the marshals," if, as members of his household, they were
amenable to the summary jurisdiction of his domestic
tribunal, or, if not members, then they should pay treble
damages and suffer imprisonment for forty days.^
^The rate fixed by 13 Charles II. c. 8, for the hire of carts or carriages requisi-
tioned by the King, was 6d. per mile. This hire included six oxen, or alternatively
two horses and four oxen, to each vehicle.
2 See 3 Edward I. c. 32.
336 MAGNA CARTA
CHAPTER THIRTY-ONE.
Nec nos nee ballivi nostri capiemus alienum boscum ad
castra, vel alia agenda nostra, nisi per voluntatem ipsius
cujus boscus ille fuerit.
Neither we nor our bailiffs shall take, for our castles or for
any other work of ours, wood which is not ours, against the will
of the owner of that wood.
Purveyance of timber growing elsewhere than on royal
manors is here prohibited in absolute terms. In marked
contrast with the limited restrictions placed upon other
branches of purveyance, this branch is taken away, not
merely from local officials, but from the King himself.^
There was an obvious reason for greater stringency in this
case : the King's own extensive demesne woods furnished
timber in abundance, whether for building purposes or for
firewood, leaving him no excuse for taking, especially if for
nothing, the trees of other people.
The purveyors of James I., shortly after his accession,
transgressed this provision of Magna Carta by requisitioning
timber for repairing the fortifications of Calais. A decision
against the Crown was given by the Barons of Exchequer
in the second year of James's reign, and a proclamation was
issued, bearing date 23rd April, 1607, disclaiming any right
to such a prerogative. The guilty purveyors were brought
before the Star Chamber.^
CHAPTER THIRTY-TWO.
Nos non tenebimus terras illorum qui convicti fuerint de
felonia, nisi per unum annum et unum diem, et tunc red-
dantur terre dominis feodorum.
1 Cf. Sir James Ramsay, Angevin Empire^ p. 476, who considers that chapters
28 and 30, in the branches of prerogative with which they respectively deal,
*' leave the king's personal right open."
^ See Coke, Second Institute, 36.
CHAPTER THIRTY-TWO 337
We will not retain beyond one year and one day, the lands of
those who have been convicted of felony, and the lands shall
thereafter be handed over to the lords of the fiefs.
I. The Crown's Claim to Property of Felons, The
Crown had established certain rights, not too well defined,
in the property of criminals formally indicted and sentenced
for felony. John, here as elsewhere, took advantage of the
vagueness of the law to stretch prerogative to its limit.
Magna Carta, therefore, attempted to define the exact boun-
daries of his rights. Custom gave the felon's land to his
feudal lord, and his chattels to the lord who tried him. The
Crown encroached on the rights of both, claiming the real
estate of felons, as against mesne lords, and their personal
estate, as against lords who had jurisdiction.
(i) The felon's lands. No difficulty arose when Crown
tenants were convicted, for the King was lord of the fief as
well as lord paramount, and claimed the whole lands as
escheat. When the condemned man was the tenant of a
mesne lord, however, a conflict of interests occurred, and
here a distinction, which gradually became hard and fast,
was drawn between treason and felony.^ Treason was an
offence against the person of the sovereign, and it was
probably on this ground that the King made good his claim
to seize as forfeit the entire estate, real and personal, of
every one condemned to a traitor's death. With regard to
ordinary felons, what looks like a compromise was arrived
at. The King secured the right to lay waste the lands in
question and to appropriate everything he could find there
during the space of a year and a day ; after which period
he was bound to hand over the freehold thus devastated to
the lord who claimed the escheat. Such was the custom
during the reign of Henry II. as described by Glanvill,^ who
makes it perfectly clear that, before the lands were given up
at the expiration of the year, houses were thrown down and
* Pollock and Maitland, II. 500, consider that the present chapter had a distinct
influence in accentuating this twofold classification of crimes.
^Glanvill, VII. c. 17. Cf. Bracton, folio 129, for a graphic description of
"waste," which included the destruction of gardens, the ploughing up of meadow
land, and the uprooting of woods.
338 MAGNA CARTA
trees rooted up. The lord, when at last he entered into
possession of the escheated lands found a desert, not a
prosperous manor. ^
Coke has attempted to give a more restricted explanation
of the Crown's rights in this respect, maintaining that the
" year and day " was not an addition to, but a substitute for,
the earlier right of "waste," that the King renounced his
barbarous claims in return for the undisputed enjoyment of
the ordinary produce for one year only, and agreed, in
return to hand over the land with all buildings and appur-
tenances intact.^ The authorities he cites, however, are
inconclusive, and the weight of evidence on the other side
leaves little room for doubt. Not only does the phrase*
" year, day, and waste " commonly used, create a strong
presumption ; but Glanvill's words in speaking of the earlier
practice are quite free from ambiguity, while the document
known as the Praerogativa Regis is equally explicit for a
period long after Magna Carta.^ Waste, indeed, was a
question of degree, and the Crown was not likely to be
scrupulous in regard to felons' lands, when it allowed
wanton destruction even of Crown fiefs held in honourable
wardship.*
Wide as were the legal rights of the Crown, John
extended them illegally. When his officers had once
obtained a footing in the felon's land, they refused to sur-
render it to the rightful lord after the year and day had
expired. In 1205, Thomas de Aula paid 40 marks and a
palfrey to get what he ought to have had for nothing,
namely, the lands escheated to him through his tenant's
^ Is it possible .that the origin of '* year and waste " can be traced to the difficulty
of agreeing on a definition of **real" and "personal" estate respectively? The
Crown would claim everything it could as "chattels" — a year's crops and every-
thing above the ground.
^ Second Institute, p. 36.
2 See Pollock and Maitland, I. 316. "The apocryphal statute praerogativa
regis which may represent the practice of the earlier years of Edward I." Bracto
(folio 129) while stating that the Crown claimed both, seems to doubt the legality
of the claim.
*Cf. c. 4.
CHAPTER THIRTY-TWO 339
felony. 1 Magna Carta prohibited such abuses, and settled
the law for centuries. ^ The Crown long exercised its
rights, thus limited, and Henry III. sometimes sold his
" year, day, and waste " for considerable sums. Thus, in
1229 Geoffrey of Pomeroy was debited with 20 marks for
the Crown's rights in the lands of William de Streete and
for his corn and chattels. This sum was afterwards dis-
charged, however, on the ground that the King, induced to
change his mind, doubtless by a higher bid, had bestowed
these rights on another.^
(2) The felon^s chattels. From an early date the King
enjoyed, like other owners of courts, the right to the goods
of the offenders he condemned. When Henry II. reorgan-
ized the system of criminal justice, and formulated, in the
Assizes of Clarendon and Northampton, a scheme whereby
all grave offenders should be formally indicted, and there-
after reserved for the coming of his own justices, he estab-
lished a royal monopoly of jurisdiction over felons; and
this logically implied a monopoly over their chattels — an
inference confirmed by the express terms of article five of
the earlier Assize. As the list of " pleas of the Crown "
grew longer, so this branch of royal revenue increased pro-
portionately, at the expense of the private owners of " courts
leet." The goods of outlaws and fugitives from justice
likewise fell to the exchequer — the sheriff who seized them
being responsible for their appraised value.*
* Such at least is the most probable explanation of an entry on the Ftpe Roll of
6 John (cited Madox, I. 488) ; although it is possible that Thomas only bought in
•* the year day and waste."
2 Magna Carta is peculiar in speaking of year and day, without any reference to
waste. If it meant to abolish " waste " it ought to have been more explicit. Later
records speak of '■'■annum et vastuju,''^ e.g. the Memoranda Roll, 42 Henry III.
(cited Madox, I. 315), relates how 60 marks were due as the price of the "year
and waste " of a mill, the owner of which had been hanged.
^ Pipe Roily 13 Henry III., cited Madox, I. 347. In Kent, lands held in gavel-
kind were exempt alike from the lord's escheat and the King's waste, according to
the maxim, "The father to the bough, the son to the plough." See, e.g. praero:
gativa regis, c. 1 6. See also Gloucester Pleas ^ 1 14, where apparently the King's
rights over half a hide were sold for 20s.
* Madox, I. 344-8, cites from the Pipe Rolls many examples.
340 MAGNA CARTA
The magnates in 12 15 made no attempt to interfere with
this branch of administration, tacitly acquiescing in Henry
II.'s encroachments on their ancestors' criminal jurisdic-
tions and perquisites. Under Henry HI. and Edward I.,
the forfeited goods of felons continued to form a valuable
source of revenue. In 1290 the widow of a man who had
committed suicide, and therefore incurred forfeit as a felo de
se, bought in his goods and chattels for ;^300, a high price,
in addition to which the Crown specially reserved its " year,
day and waste." ^
H. Indictment, Conviction, and Attainder, The Crown
could not appropriate the property of men merely suspected
of crime, however strong might be the presumption of guilt.
Mere accusation was not enough ; a formal judgment was
required. The Charter refers to the lands of a " convicted "
offender, and conviction must be distinguished from indict-
ment on the one hand, and from attainder on the other;
since these formed three stages in the procedure for deter-
mining guilt.
(i) Indictment. It has been already shown ^ how Henry
of Anjou tried to substitute, wherever possible, indictment
by a jury for private appeal in criminal suits. The Assize
of Clarendon authorized such indictments to be taken before
sheriffs, and we learn from Bracton that, immediately the
formal accusation had been made, the sheriff became respon-
sible for the safety of the accused man's property, both real
and personal. With the help of the coroners and of law-
worthy men of the neighbourhood he must have the chattels
appraised and inventoried, and hold them in suspense until
the " trial," providing therefrom in the interval " estovers,"
that is, sufficient sustenance for the accused and his family.^
If the prisoner was acquitted or died before conviction,
the lands and chattels were restored to him or to his rela-
tives, the Crown taking nothing. Reginald of Cornhill,
sheriff of Kent, was discharged in 1201 from liability for the
appraised value of the goods of a man who, after indictment
for burning a house, had died in gaol non convictus. As
^This case is cited by Madox, I. 347, from 18 Edv/ard I.
^ Supra^ p. 88. ' See Bracton, II. folio 123, and folio 137.
CHAPTER THIRTY-TWO 341
the Pipe Roll states, his chattels did not pertain to the
King.i
(2) Conviction, Only the justices could " try " the plea,
that is, give sentence according to success or failure in the
test appointed for the accused man to perform .2 Prior to
1 2 15 the usual test was ordeal of water in the ordinary case,
or of the red-hot iron in the case of men of high rank and of
women. If the suspected person failed, sentence was a mere
formality; he had "convicted" himself of the felony. As
a consequence of the condemnation of ordeal by the Lateran
Council of 1 2 15, the verdict of a petty jury became the
normal " test " that branded an offender as convictus. This
was long looked on as an innovation, and accordingly the
law refused to compel the accused, against his will, to trust
his fate to this new form of trial. He might refuse to " put
himself upon his country," and by " standing mute" make
his " conviction " impossible, saving himself from punish-
ment and depriving the King of his chattels and " year and
day." For centuries those responsible sfirank from the
obvious course of treating silence as equivalent to a plea
of guilty; but while liberty to refuse a jury's verdict was
theoretically recognized, barbarous measures were in reality
adopted to compel consent. The Statute of Westminster
in 1275 ^ directed that all who refused should be imprisoned
en le prison forte et dure. This statutory authority for
strict confinement was liberally interpreted by the agents
of the Crown, who treated it as a legal warrant for revolting
cruelties. Food and drink were virtually denied, a
little mouldy bread and a mouthful of impure water only
being allowed upon alternate days ; and at a later date the
prisoner was slowly crushed to death under great weights
"as heavy, yea heavier than he can bear." Brave men,
guilty, or mayhap innocent, but suspicious of a corrupt
jury, preferred thus to die in torments, that they might
save to their wives and children the property which would
upon conviction have fallen to the Crown. The fiction was
carefully maintained that the victim of such barbarous treat-
1 Pipe RolU 2 John, cited Madox, I. 348. ^ Qf^ supra, c. 24.
33 Edward I. c. 12.
342 MAGNA CARTA
ment was not subjected to " torture, " always illegal at
common law, but merely to peine forte et dure, sl perfectly
legal method of persuasion under the Statute of 1275.
This procedure was not abolished until 1772; then only
was an accused man for the first time deprived of his right
to " have his law " — his claim to ordeal as the old method
of proving his innocence. Until that date, then, a jury's
verdict was treated as though it were still a new-fangled
and unwarranted form of " test " usurping the place of
the ordeal, although the latter had been virtually abolished
early in the thirteenth century. ^
(3) Attainder. Coke in commenting on this passage
draws a further distinction between "conviction," which
directly resulted from a confession or a verdict of guilty,
and " attainder " which required a formal sentence by the
judge. In his age, apparently, it was the attaint that
implied forfeiture; looking as usual at Magna Carta
through seventeenth-century glasses, he seems surprised
to find " convicted " used where he would have written
"attainted." Yet this distinction, if recognized in 1215,
must have been immaterial then. It was under the Tudor
sovereigns that the doctrine of the penal effects of attainder
was elaborated. When sentence was passed on a felon, a
blight fell on him : his blood was impure, and his kindred
could inherit nothing that came through him. The Crown
reaped the profit.^
Statutes of the nineteenth century modified the harshness
with which this rule bore on the felon's innocent relations : ^
finally the Forfeiture Act of 1870* abolished "corruption
of blood " and deprived the Crown of all interest in the
estates of felons, alike in escheats and chattels. Thus the
word "attainted" has become practically obsolete. A
^The Act 12 George III. c. 20, made standing mute equivalent to a plea of
guilty. A later Act, 7 and 8 George IV. c. 28, made it equivalent to a plea of not
guilty. See Stephen, Hist. Crim. Law^ I. 298.
* This fiction of corrupt blood was apparently based in part on a false derivation
of the word "attainder." See Oxford English Dictionary.
3 E.g. 54 George III. c. 145, and 3 and 4 William IV. c. 106, s. 10.
* 33 and 34 Victoria, c. 23.
CHAPTER THIRTY-TWO 343
criminal who is fulfilling the term of his sentence is known,
not as a man attainted, but simply as a " convict," the same
word as was used in Magna Carta.
CHAPTER THIRTY-THREE.
Omnes kydelli de cetero deponantur penitus de Tamisia,
et de Medewaye, et per totam Angliam, nisi per costeram
maris.
All kydells for the future shall be removed altogether from
Thames and Medway, and throughout all England, except upon
the sea shore.
The object of this provision is not open to doubt ; it was
intended to remove from rivers all obstacles likely to
interfere with navigation. Its full importance can only be
understood when the deplorable state of the roads is kept
in view. The water-ways were the great avenues of com-
merce; when these were blocked, townsi^ien and traders
suffered loss, while those who depended on them for
necessaries, comforts, and luxuries, shared in the general
inconvenience. Magna Carta mentions only one kind of
impediments, namely, " kydells " (or fish-weirs), not because
of the purpose to which these were put, but because they
were the form of obstruction that called for repressive
measures at the moment. This word seems to have been
used by the framers of Magna Carta in a wide general
sense, embracing all fixed contrivances or " engines "
intended to catch fish, and likely by their bulk to interfere
with the free passage of boats. ^
It has been gratuitously assumed that the motive for
prohibiting " kydells " must have been of a similar nature
to the motive for constructing them ; and that therefore
the object of the present chapter was to prevent any mono-
poly in rights of fishing. Law courts and writers on
jurisprudence for many centuries endorsed this mistaken
^The Oxford English Dictionary defines it as ** a dam, weir, or barrier in a
river, having an opening in it fitted with nets or other appliances for catching fish."
For weirs in Domesday Book, see Ballard, D. Inqtiest, 175-6.
w
344 MAGNA CARTA
view, and treated Magna Carta as an absolute prohibition
of tlie creation of " several " (or exclusive) fisheries in tidal
waters. 1 Although this legal doctrine has been frequently
and authoritatively enunciated, it rests on a misconception.
The Great Charter sought to protect freedom of navigation,
not freedom of fishing ; and this is obvious from the last
words of the chapter : kydells are to be removed from
Thames and Medway and throughout all England " except
upon the sea-shore." It would have been a manifest
absurdity to allow monopolies of taking fish in the open
seas, while insisting on freedom to fish in rivers, the banks
of which were private property. The sense is clear : no
objection was taken to " kydells " so long as they did not
interfere with navigation.
The erroneous view, however, had much to excuse it,
and acquired plausibility from the circumstance that the
destruction of obstacles to the free passage of boats inciden-
tally secured also free passage for salmon and other migra-
tory fish ; and that later statutes, when legislative motives
had become more complicated, were sometimes passed with
both of these objects in view. The change is well illus-
trated by a comparison of the words of two statutes of 1350
and of 1472 respectively. The first of these repeats the
substance of this chapter, and thus explains its object: —
" Whereas the common passage of boats and ships in the
great rivers of England be oftentimes annoyed by the
inhancing of gorces, mills, weirs, stanks, stakes, and
kydells." 2 Here there is no allusion to fish or rights of
fishing. The later Act, while confirming, under penalties,
previous statutes for the suppression of weirs, not only
states its own intention as twofold, namely, to protect
^Blackstone, Commentaries, IV. 424, declared that this chapter "prohibited for
the future the grants of exclusive fisheries." Cf. e.g. Thomson, Magna Charta,
214, and Norgate, John Lackland, 217. See also Malcolmson v. O'Dea (1862),
10 H. of L. Cas., 593, and Neill v. Duke of Devonshire (1882), 8 App. Ca. at
p. 179, — cases cited in Moore, History and Law of Fisheries, p. 13, where the
fallacy is exposed. For an unsuccessful attempt to extend the principle to Scotland,
after the Act of Union, see an interesting review of the first edition of this work in
Jurid. Rev. for March, 1905.
* 25 Edward III., stat. 3, c. 4.
CHAPTER THIRTY-THREE 345
navigation of rivers, and " also in safeguard of all the fry
of fish spawned within the same," but retrospectively and
unwarrantably attributes a like double motive to Magna
Carta.i
So far as Thames and Medway were concerned, this
provision contained nothing new. To the Londoners,
indeed, the keeping open of their river for trade was a
matter of vital importance. The right to destroy kydelli
had been purchased from Richard I. for 1500 marks, and
a further sum had been paid to John to have this confirmed.
These charters (dated 14th July, 1197, and 17th June, 1199)
" granted and steadfastly commanded that all kydells that
are in the Thames be removed wheresoever they shall be
within the Thames; also we have quit-claimed all that
which the Warden of our Tower of London was wont yearly
to receive from the said kydells. Wherefore we will and
steadfastly command that no warden of the said Tower, at
any time hereafter, shall exact anything of any one, neither
molest nor burden nor make any demand qf any person by
reason of the said kydells." John's Charter went further
than that of Richard, making it clear that the prohibition
referred to Medway as well as to Thames, and granting
the right to inflict a penalty of ;^'io upon anyone infringing
its provisions.^
Magna Carta confirmed this provision and extended it
to all rivers, and this was repeated in the reissues of
Henry IIL The citizens of London, not content with a
clause in a general enactment, purchased for 5000 marks
^ 12 Edward IV. c. 7. Apparently the earliest statute which refers to weirs as
causing injury to fish was one passed in 1402, namely, 4 Henry IV. c. 1 1 ; see Moore,
Fisheries, p. 175.
2 It seems to have been generally assumed that these charters conferred positive
as well as negative privileges on the citizens, including rights of administration and
jurisdiction over the waters of Thames. See Noorthouck, N'e7v History of I^ondon
(1773), 36. Lufifman, Chartcj-s of London (1793), 13, says of Richard's grant in
1 197 : "By this charter the citizens became conservators of the river Thames."
This is an anachronism, but Patent Rolls of 33 Edward I., 5 Edward III.,
8 Edward III., contain Commissions of Conservancy. See Moore, op. cit., p. 176.
In 1393 the statute of 17 Richard II. c. 9 granted authority to the Mayor of
London to regulate weirs and generally to "conserve" the Thames from Staines
downwards, and the Medway.
346 MAGNA CARTA
three new charters exclusively in their own favour. One
of these, dealing with kydells in Thames and Medway,
was issued by Henry on i8th February, 1227, in terms
almost identical with those of Richard and John.^
CHAPTER THIRTY-FOUR.
Breve quod vocatur Precipe de cetero non fiat alicui de
aliquo tenemento unde liber honiio amittere possit curiam
suam.
The writ which is called praecipe shall not for the future be
issued to anyone, regarding any tenement whereby a freeman
may lose his court.
The grievance here dealt with lay at the heart of the
quarrel of 12 15, and the remedy adopted proved a vital
factor in the history of royal jurisdiction in England.
In extorting from John a solemn promise to restrict the
use of this particular writ, the barons gained something
of infinitely greater value than a petty reform of court
procedure ; they committed their enemy to a reversal of a
line of policy vigorously pursued for half a century. The
process by which the jurisdiction of the King's courts was
undermining that of the feudal courts was now to be
arrested.^ Magna Carta, by this apparently inoffensive
clause, was grappling in reality with an urgent problem
of the day, fraught with tremendous practical issues alike
for King and barons. This can only be understood in con-
nection with the technical details on which it hinges.
I. Royal Writs and Feudal J tir is dictions. In pleas of
disputed titles to land, feudal theory gave sole jurisdiction
to the lord of the^^L_ No principle was more absolutely
established than this : no person, neither King nor Em-
peror, had any right to interfere, except on the one ground
of failure of justice. Not even Henry II. dared to
repudiate this universal rule; but he adopted expedients
^See Rotuli Cartaruniy ii Henry III.
2 The Histoire des dticSy 149, paraphrases this chapter thus: "Toutes hautes
justices vaurrent-ils avoir en lor tierres." Miss Norgate, Minority y 11, has not
grasped the significance of this clause.
CHAPTER THIRTY-FOUR 347
to render it inoperative. If Glanvill may be trusted,
Henry was strong enough to obtain acquiescence in his
prohibition of any plea, concerning ownership of a lay fee,
being tried in a seignorial court without the licence of a
royal precept.^
Henry also invented, or adopted from precedents of the
Carolingian Emperors, two types of writ, the virtual effects
of which were to evoke causes from the lords' courts to his
own, without too open an infringement of feudal principle.^
These were the Writ of Right and the Writ Praecipe, .j
The Writ of Right proceeded on the principle that a lord X^i^t^
superior was bound to see that his vassals dispensed
justice to their rear vassals. When a freeholder, the
tenant of a mesne lord, complained to the King that justice
was refused him, the King formally commanded the remiss
lord " to do full right " to the complainant, and added the
threat that, unless he did so, the King himself would. The
writ, known as a hieve de recto tenendo, was thus issued
to the owner of a feudal court ; professed {o afford him an
opportunity of obedience by trying the plea in his court;
and avoided conflict with feudal theory by justifying the
proposed royal interference as ^^ pro defectii justitiae," It
afforded, however, excellent opportunities for the insidious
encroachments of the royal courts at the behest of powerful
kings, who retained in their own hands the right to define i
what constituted a failure to do justice.^ Jix^
The Writ Praecipe in its origin and antecedents diffej:ed*'''M
fundamentally from the Writ of Right : it was addressed Qj/^
to the sheriff, not to the owner of a franchis^-rtwas a more
direct violation of feudal rights, for it made no allegation
of failure of justice but simply ignored the lord's jurisdic-
tion, bidding the sheriff command the tenant to restore the
land in question to the demandant^ or else to appear before
the royal court to explain his reasons for disobedience.^
No opportunity was afforded the mesne lord of hearing the
plea. The whole procedure, almost without disguise or
excuse, was an open transference of the dispute from the
* Glanvill, XII. 25. See stcpra, p. 89. ^g^yj^ner, Schwurgerichte, jS^.
5 The form of the writ is given in Glanvill, XII. 3.
348 MAGNA CARTA
v"- manorial court to that of the King.^ The writ, which on
the surface reads merely as a summary and final command
to hand over the estate to another, is really an " original
writ " commencing a litigation in the King's court. Dne
important effect of its issue was that all proceedings
instituted in inferior tribunals must immediately stop.
The feudal lord, in whose court baron the plea would
naturally have been decided, was thus robbed by the King
of his jurisdiction. With it, he lost also authority over
his tenants, and numerous fees and perquisites. The writ
praecipe was thus an ingenious device for " evoking " a par-
ticular cause from the manorial court to the King's court.^
I The two types of writ, praecipes and writs of right, at
[first contrasted as alternative methods of bringing a plea
I under royal jurisdiction, came in time to have entirely
t different relations. The person to whom the preliminary
writ was issued, whether sheriff or lord of a franchise court,
ceased to be of much importance, when the writ had become
a mere formality. The^^gssetUial feature of a Writ of Right
came to be that it dealt with ownership as opposed'to mere
possession : all royal writs that originated pleas involving
title to land w^ere then reckoned Writs of Right, which
now embraced an important species of the originally
opposed genus of writs praecipe.^ Thus, in one place,
writs praecipe and writs of right overlapped each other.*
The motives of Henry II., in instituting his legal reforms,
were probably mixed; and it is not easy to determine
whether he favoured his new writs most because they really
iCf. Stubbs, ConsL Hist., I. 576.
^Glanvill, I. 6, gives the form oi z. praecipe : Rex vicecomiti salutem^ Praecipe A.
quod sine dilatione reddat B. tmam hidam terrae in villa ilia, unde idem B.
giieritur quod praedictus A. ei deforceat : et nisi fee erit, summone eum per bonos
simimonitores quod sit ibi coram me vel Justiciariis meis in crastino post octabas
clausi Paschae apud loctim ilium, ostensuitis quare non fecerit. Et habeas ibi
summonitores et hoc breve. Teste Ranulpho de GlanviUa apud Clarendon.
^Brunner, Schwurgerichte, 411 ; Maitland, Col. Papers, II. 129.
*Coke, Sec. Inst., 40, gives three varieties oi praecipe : {a) praecipe quod reddat ;
{b) quod permittat ; {c) quodfaciat. The first group includes one variety of Writs
of Right and the various Writs of Entry. Writs of Right, on their part, are of three
kinds : (i) writ of right patent, (2) ^nxxt praecipe, (3) little writ of right, applicable
to villeins on ancient demesne.
CHAPTER THIRTY-FOUR 349
stimulated the flow of justice in the feudal courts, or because
they afforded facilities for sapping their strength. While
reforming the entire administration of justice in England,
the King hoped, by the same means, to destroy gradually
the feudal privileges of his magnates. He intended to
draw into his own courts all pleas relating to land. Ques-
tions of property were to be tried before his justices, by
combat or, at the tenant's option, by the grand assize;
questions of possession (without any option) by the appro-
priate petty assize. The barons showed no desire to dis-
pute the Crown's assumption of a monopoly over the petty
assizes; indeed they cordially acquiesced in this by the
terms of chapter 18 of the Charter. The grand assize
was another matter; they refused to be robbed of their
right to determine, in their own courts baron, proprietary
actions between their own tenants. Indeed, for such whole-
sale extension of the King's jurisdiction over pleas of land,
Henry II. had absolutely no precedent. He had made the
Crown strong and then used its power for Uis own aggran-
dizement. The King's courts had increased their authority,
as a distinguished American historian has expressed it,
" by direct usurpation, in derogation of the rights of the
popular courts and manorial franchises, upon the sole
authority of the King." ^
While undermining the feudal courts, Henry was devis-
ing improved methods of dispensing justice in his own.
Efforts were being successfully made, as has been shown, 2
to substitute the grand assize for trial by combat ; and the
desire for the more rational mode of proof favoured the
King's policy of removing important litigations to his own
court. The assize procedure must be taken along with the
writ of right and the writ praecipe as parts of one scheme
of reform.
II. The Intention of Magna Carta. The present chapter
1 See Bigelow, HzsL of Procedure, 78. Glanvill, read between the lines,
supports this view. Thus in I. c. 3, he speaks of the King's courts as normally
dealing with ** pleas of baronies" ; in I. c. 5, he speaks of what he evidently con-
siders an abnormal expansion of this jurisdiction to any plea anent a free tenement,
if the Crown so desired.
2 See «//;-«, under c. 18.
350
MAGNA CARTA
says nothing of the Writ of Right, but guards against the
abuse of the Writ Praecipe, without attempting to interfere
with its employment within its legitimate sphere, that is to
say, in settling disputes as to Crown fiefs. John might
keep his own court, and issue praecipes to his own tenants;
but let him respect the rights of other feudal lords and not
use his writs as engines of encroachment upon manorial
jurisdictions. For the future, such writs must not be issued
" concerning any tenement whereby a freeman may lose his
court." Writs praecipe might be freely used for any other
purpose, but not for this. This one purpose, however,
was exactly what had specially recommended them to King
Henry.
The present chapter must, therefore, be regarded as one
of the most reactionary in the Charter : the barons had
forced John to promise a complete reversal of the deliberate
policy of his father.
Here, then, under the guise of a small change in legal
procedure, was concealed a notable triumph of feudalism
over the centralizing policy of the monarchy — a backward
step, which, if given full effect to, might have ushered in a
second era of feudal turbulence such as had disgraced the
reign of Stephen. We are told on high authority that
John's acknowledgment of "the claims of the feudal lord
to hold a court which shall enjoy an exclusive competence
in proprietary actions " was one which " Henry II. would
hardly have been forced into."^ That may well be; but
John had already more than once rejected this proposal
with vehemence. In 12 15, he could no longer strive against
the inevitable, and agreed under compulsion to provisions
which he had no intention to keep. The concession,
although insincere, was nevertheless an important one.
The substance of chapter 34 was repeated, with some trivial
verbal alterations, in all future issues of Magna Carta.^
1 Pollock and Maitland, I. 151.
2 The version of 1216 speaks of a "free tenement," where that of 1215 spoke
merely of a '* tenement." The addition makes no change, since in no case could
the King's courts try pleas affecting villeins of mesne lords. Perhaps the object is
to make it clear that there was no interference with the King's rights over holdings
of his own villeins.
CHAPTER THIRTY-FOUR 351
Why did the barons, it may be asked, while attacking
the writ praecipe^ allow the writ of right patent to go un-
scathed ? History is silent; but inferences may be drawn.
The barons had no legal ground for condemning the legiti-
mate use of the writ of right even when it deprived a baron
or other freeman of his court. Feudal theory sanctioned
this procedure, unless where it was abused; and it was
difficult to define abuse of the procedure. If " pro defectu
justitiae " was honestly alleged, the King had a right to
interfere, well grounded in feudal law. The interference,
too, even where unwarranted, was of a subtle nature,
and difficult to guard against. Finally, encroachments
initiated by this procedure had not been attempted before
1215 to any noticeable extent: the barons had no pre-
monition of the new uses to which the writ of right would
be put, after the channel of royal aggression by way of
the praecipe had been closed. The writ of right patent
was a cumbrous process, and its short day of usefulness
came after the granting of Magna Carta.
III. Expedients for evading Magna Carta. One question
remains: was this provision observed in practice? The
answer is that its letter was stringently observed, but its
spirit was evaded. Writs praecipe that deliberately evoked
suits, other than those of Crown tenants, to the King's
courts ceased to be issued, but the sphere denied to the
writ was made as narrow as possible; and methods were
devised for reducing seignorial courts practically to im-
potence, without direct violation of the terms of the Great
Charter.
(i) The letter of the law. The Chancery, in obedience
to Magna Carta, ceased to issue this particular form of writ
in such a manner as to cause a freeman " to lose his court."
It was still issued to Crown tenants; but strictly denied
to under-tenants, who were thus left to find redress at the
feudal court of the magnate from whom they held their
land.i
*In translating the reissue of 1225, the Statutes at Large expand the word
^'■praecipe'''' into ^^ praecipe in capite,^* for which there is no authority in any known
text of Magna Carta, though it appears in Coke's version of Henry's charter [Sec.
352 MAGNA CARTA
The measure thus forced on the Crown in the selfish
interests of the baronage, inflicted hardship on tenants of
mesne lords : the court baron was now their only source
of feudal justice, and in that court they could not get the
benefit of the improved methods of royal procedure. In
particular, the grand assize was a royal monopoly. The
magnates, indeed, desired to adopt it, but they had difficulty
in getting together twelve knights willing to act as jurors.^
Whatever hopes the barons entertained of overcoming such
difficulties were disappointed : in 1259 the Provisions of
Westminster declared that freeholders should not be com-
pelled to swear against their will " since no one can make
them do this without the King's warrant." ^ It was the
deliberate policy of Edward I. to exaggerate all such diffi-
culties, putting every obstacle in the way of private courts,
until he reduced their jurisdictions to sinecures.^
(2) Evasion of its spirit. While the letter of Magna
Carta was strictly kept, its spirit was evaded. It was
impossible to give loyal effect to an enactment that went
Insf., 38). Authorities differ as to what constitutes z. praecipe in capite. Brunner,
Schwtirgerichte, sec. xx., declares it to be so called "because it begins with the
word Praecipe " ; yet all praecipes so begin, even Writs of Entry, which are cer-
tainly not condemned by Magna Carta. Coke [Sec, Inst., 38) seems (inconsistently
with his own version of Magna Carta) to identify the praecipe in capite with a class
of writs NOT prohibited in the Charter, namely, with those professing to deal with
estates held directly under the Crown : no one ought to have it without taking
oath ** that the land is holden of the King in capite.''^ He cites illustrations from
the reign of Edward I. Adams {Origin, 104), speaks of an **/« capite^^ clause
inserted m praecipes to evade the prohibition of Magna Carta. See also Holdsworth,
in. 10.
^ Such an attempt seems to have been made in 1207 by Walter de Lacy, Earl of
Ulster, who set up in his Irish fief what is described as nova assisa, against which
John protested. See Rot. Pat., I. 72, for writ dated 23rd May, 1207. In one
case John acquiesced in grand assizes being held in feudal courts : on 4th May,
1 201, he granted licence to Hubert Walter to hold them for his tenants in gavel-
kind. See New Rymer, I. %-}^.
^See article 18 {Select Charters, 404). Cf. chapter 29 of the Petition of the
Barons {^^/(?r/ Charters, 386), and Pollock and Maitland, I. 182: "The voice of
the nation, or what made itself heard as such, no longer, as in 1215, demanded
protection for the seignorial courts. "
^ A partially successful attempt was made to revive feudal jurisdictions as late as
the reign of Edward III. See Stubbs, Const. Hist., II. 638-9.
CHAPTER THIRTY-FOUR 353
directly counter to the whole stream of progress. Manorial
justice was falling into disrepute, while royal justice was
becoming more efficient and more popular. Under-tenants,
deprived of access to the King's court by the direct road
of the writ praecipe^ sought more tortuous modes of
entrance. Legal fictions were devised. The problem was
how to evade Magna Carta without openly infringing it :
the King's justices and would-be litigants in the King's
courts formed a tacit alliance for this end, but had to pro-
ceed by wary steps, in the teeth of opposition from the
powerful owners of seignorial courts. Three methods were
adopted by the Crown : —
(a) Magna Carta had not condemned the writ praecipe,
but only its abuse; and abuse was sometimes difficult
to define. That writ remained the normal procedure in
cases of Crown holdings,^ and a liberal interpretation of
this exception would sometimes pass unchallenged, though
there seems no ground for supposing that any recognized
legal fiction of this nature came into use. Then, besides
the later developments of the praecipe (to be afterwards
described), the King claimed, in spite of Magna Carta, to
grant ex gratia speciali the very writ complained of .^
(b) When the use of the writ praecipe was barred, the
King could fall back on the more cumbrous procedure
instituted by writ of right, the potentialities of which
were developed after 12 15. Coke ^ cites an instance from
the 34th year of Edward I., where a demandant admitted
that the lands in dispute were not held of the King in capite
but of his brother Edmund, and therefore he could not
proceed by way of praecipe, but he might, if he so desired,
proceed by writ of right patent in the King's curia. This
substitution of the writ of right for the praecipe is described
by Professor Maitland * as " a victory of feudalism conse-
crated by the Great Charter."
When a tenant, whose title was challenged in his lord's
court, applied to the King for a grand assize, the plea was
practically certain, by one avenue or another, to reach the
1 See, e.^. Madox, I. 793. 2 Bracton, 404b.
^Sec. Inst., 38. ^ColL Papers^ II. 129.
Z
354 MAGNA CARTA
Curia Regis.^ The rule that no one need defend his
liberum tenementuni unless summoned by a royal writ also
worked towards the same end. But many difiiculties lay
in the path of the writ of right.^ The Petition of the Barons
of 1259 (chapter 29)^ illustrates one attempt to make the
most of these. Moreover, the whole procedure was dilatory^
expensive, and inelastic, and it was gladly abandoned, after
the invention of less direct but more convenient methods of
effecting the same purpose.
(c) The procedure which rendered recourse to the writ
of right unnecessary was instituted by one of various writs
developed from the older praecipe and known as writs of
entry. These writs instituted procedure in the King's
court on the averment of some recent flaw in the tenant's
title, which could be settled without opening up the whole
matter of the ownership. This was a subterfuge, for the
settling of the special point virtually decided the general
question of ownership without appeal. Although probably
not invented for the express purpose of defeating this
chapter of the Great Charter, these writs were soon applied
to that purpose. One of the most useful of their number
was the writ of cosinage, devised by William of Raleigh,
extending to others than the dispossessed heir the simple
procedure of the petty assizes. As early as 1237, it was
decided in the King's court that such a writ did not violate
the Charter.* Writs of entry were thus, from the point
of view of the magnate with his private court, wolves in
sheep's clothing. They professed to determine a question
of possession, but really decided a question of ownership.
At first, the pleas to which they could be applied were few
and special. Steadily, new forms of action were devised
to cover almost every conceivable case. The process of
evolution was a long one, commencing soon after 12 15, and
virtually concluding with chapter 29 of the Statute of Marl-
'See Brunner, SchwitrgeHchte, 406; Maitland, Coll. Papers, II. 129.
2 See Glanvill, XII. 7. ^ SeL Chart., 386-7.
*See Bracton's Note-book, plea 1215, where the writ in question is cited at
length : it contains the sentence, " nee tollat aliati curiam S7tam nhi locum habere
possit breve de recto.''''
CHAPTER THIRTY-FOUR 355
borough, or rather with the liberal construction which Crown
lawyers placed upon that statute in the following reign.
Edward I., at the height of his power, and eager to set
his house in order, shrank from an open breach of the
Great Charter, gladly adopting subtle expedients to oust
mesne lords from rights secured to them by the present
chapter. In Edward's reign the legal machinery was
brought to perfection, so that thereafter no action relating
to freehold was ever again tried in the courts baron of the
magnates, but, in direct violation of the spirit of Magna
Carta, decided in the courts of the King.i
The demandant had no need to infringe the prohibi-
tion against the older form of writ praecipe when he
could obtain another writ, equally effective. A writ of
entry was, indeed, to a peaceable demandant, preferable to
a writ praecipe, which could only be issued to one prepared
to offer battle, the option of accepting lying with his
adversary. Crown tenants, even, who could obtain the
original writ praecipe, came to prefer the modern substitute;
and clause 34 of Magna Carta was virtually obsolete.
IV. Influence on later legal development. One of the
indirect effects of the clause was of a most unfortunate
nature. The necessity it created for effecting reforms by
a tortuous path did great and lasting harm to the form of
English law. Legal fictions have indeed their uses, by
evading technical rules of law in the interests of substantial
justice. The price paid for this relief, however, is usually
a heavy one. Complicated procedures and underhand
expedients have to be invented, and these lead in turn to
new legal technicalities of a more irrational nature than
the old ones. It would have been better in the interests
of scientific jurisprudence, if so desirable a result could have
been effected in a more straightforward manner. The
authors of Magna Carta must bear the blame.^
^ Technical details are given by Pollock and Maitland, II. 63-7. The whole
family of writs were known as "writs of entry sur disseisin''^ ; and these were
applied to still wider uses after 1267 on the authority of the Statute of Marl-
borough, as " writs of entry sur disseisin on the posty See also Maitland, Preface
to Sel. Pleas in Manorial Courts, p. Iv.
2 Cf, Pollock and Maitland, 1. 1 5 1 , and SeL Pleas in Manoi-ial Courts, already cited.
356 MAGNA CARTA
CHAPTER THIRTY-FIVE.
Una mensura vini sit per totum regnum nostrum, et una
mensura cervisie, et una mensura bladi, scilicet quarterium
Londonie, et una latitude pannorum tinctorum et russet-
orum et halbergectorum, scilicet due ulne infra listas; de
ponderibus autem sit ut de mensuris.
Let there be one measure of wine throughout our whole realm ;
and one measure of ale ; and one measure of corn, to wit, " the
London quarter"; and one width of cloth (whether dyed, or
russet, or " halberget "),^ to wit, two ells within the selvedges;
of weights also let it be as of measures.
This chapter confirmed the provisions of various ordin-
ances that sought to regulate the sale of commodities.
Assizes of bread and beer were issued from time to time,
and also assizes of weights and measures, and of wines.
Richard's Assize of Cloth, for example, of 20th November,
1 197, was, according to modern conceptions of the proper
sphere of government, partly commendable and partly ill-
advised. It strove, on the one hand, to overcome the
inconvenience experienced by traders, who met with vary-
ing standards as they moved their wares from place to place.
What was of more importance, the Assize sought to obviate
frauds perpetrated upon buyers under shelter of ambiguous
weights and measures. The London quarter must, there-
fore, be used everywhere for corn ; and one measure for
wine or beer : so far, good. On the other hand, the ordin-
ances of Richard went further than modern ideas of laisses
faire would tolerate. In particular, freedom of trade was
interfered with by the regulations reported by Roger of
Hoveden.2 No cloth, he tells us, was to be woven except
of a uniform width, namely, " two ells within the lists." ^
^This word, unknown to Ducange, seems to be connected with the "hauberk"
or coat-of-mail. It may mean thick cloth worn under a coat-of-mail.
•^R. Hoveden, IV. 33-4.
'At a later date cloth of an alternative standard width was also legalized, viz.,
of one yard between the "lists." Hence arose the distinction between "broad-
cloth" (that is, cloth of two yards) and "streits" (that is, narrow cloth of one
CHAPTER THIRTY-FIVE 357
Dyed cloths, it was provided, should be of equal quality
through and through, as well in the middle as at the
outside. Merchants were prohibited from darkening their
windows by hanging up, to quote the quaint language of
the ordinance, " cloth whether red or black, or shields
(scuta) so as to deceive the sight of buyers seeking to
choose good cloth." Coloured cloth was only to be sold
in cities or important boroughs. Here we have a sumptuary
law meant to ensure that the lower classes went in modest
grey attire. Six lawful men were to be assigned to keep
the Assize in each county and important borough. These
custodians of measures must see that no goods were bought
or sold except according to the standards ; imprison those
found guilty of using other measures ; and seize the chattels
of defaulters, for the King's behoof. If the custodes per-
formed their duties negligently they were to suffer amerce-
ment of their chattels.^ Richard's Assize of Measures was
supplemented in 1199 by John's Assize of Wine, which
tried to regulate the price of wines of various qualities,^ an
attempt not repeated in Magna Carta.
The author who gives us the text of the ordinance of
1 197, tells us that its terms were too stringent, and had to
be relaxed in practice.^ This was done in 1201 : the King's
justices seized cloth that was less than the legal width.
They compromised, however, by accepting money " to the
use of the King and to the damage of many " ; thus
Hoveden denounces what he regards as an unlawful bargain
between justices and traders for evading the strict letter of
the ordinance.
The justices, indeed, were often more intent on collecting
fines for its breach than on enforcing the Assize. In 1203,
two merchants of Worksop were amerced each in half a
yard) (see Statute i Richard III. c. 8). The word " broadcloth " has, long since,
changed its meaning, and now denotes material of superior quality, quite irrespec-
tive of width. See Oxford English Dictionary, under "Broadcloth."
^Cf. supra, c. 20, for "amercements," and supra, c. 24, for "custodes" of
pleas (or coroners).
2 See R. Hoveden, IV. icx). <
'See Hoveden, IV. 172, and Stubbs, Const. Hist., I. 616.
358 , MAGNA CARTA
mark for selling wine contrary to the Assize, while the
custodians of measures of the borough were mulcted in
one mark for performing their duty negligently — an exact
illustration of the words of the ordinance.^ In the same
year, a fine of one mark was imposed on certain merchants
"for stretching cloth," in order, presumably, to bring it
to the legal width. ^ Merchants frequently paid heavy fines
to escape the ordinance altogether.^
When the barons in 12 15 insisted upon John enforcing
his brother's ordinance, they took a step in their own
interests as buyers, and against the interests of the trade
guilds as sellers. Although this provision was repeated
in subsequent charters, evasion continued. One example
may suffice : in the second year of Henry III.'* the citizens
of London paid 40 marks that they might not be questioned
for selling cloth less than two yards in width. Here is an
illustration of the practice of the judges to which Hoveden
had objected, and which Magna Carta had apparently failed
to put down. Sometimes, however, Richard's Assize of
Measures^ and John's Assize of Wine were enforced.
In 1 2 19, a Lincolnshire parson, with a liberal conception
of his parochial duties, had to pay 40s. for wine sold extra
Assisam.^ Parsons, apparently, might engage in trade,
but only if they conformed to the usual regulations.
* See Pipe Roll^ 4 John, cited Madox, I. 566.
* See ibid.
^In 1203 the men of Worcester paid lOOs. *^ ut possint emere ct vendere pannos
tinctos sicut solebani tempore Regis Henrici " ; and the men of Bedford, Beverley,
Norwich and other towns made similar payments. See Pipe Roll, 4 John, cited
Madox, I. 468-9.
* See Pipe Roll, cited Madox, I. 509.
^ Gloucester Pleas, No. 501.
^ Pipe Roll, 3 Henry III., cited Madox, I. 567.
CHAPTER THIRTY-SIX 359
CHAPTER THIRTY-SIX.
NiCHiL detur vel capiatur de cetero pro brevi inquisicionis
de vita vel membris, sed gratis concedatur et non negetur.
Nothing in future shall be given or taken for a writ of inquisition
of life or limbs, but freely it shall be granted, and never denied.
This chapter has an important bearing upon trial by
combat, and none at all upon habeas corpus, to which it is
often supposed to be closely related. The writ upon which
emphasis is here laid had been invented by Henry II. to
obviate the judicial duel, by allowing the accused to refer
the question of guilt or innocence to the verdict of his
neighbours.
I. Trial by Combat prior to the Reign of John, The
crucial moment in judicial proceedings during the Middle
Ages arrived, as has already been explained,^ when the
**test" or "trial" (lex), appointed by the court, was
attempted by one or both of the litigants. The particular
form of proof to which the warlike Norman barons were
attached was the duellum, and it was only natural that such
of the old Anglo-Saxon aristocracy as associated with them
on terms of equality should adopt their prejudices. " Com-
bat " became the normal mode of deciding pleas among the
upper classes. From the first, however, it seems not to have
been competent for property of less than los. in value, ^ and
it soon came to be specially reserved for two classes of dis-
putes— civil pleas instituted by writ of right, and criminal
pleas following on "appeal." The present chapter is con-
cerned with the latter only.
An " appeal " in this connection was entirely different
from the modern appeal from a lower to a higher court. It
was a formal accusation of treason or felony made by a
private individual on his own initiative, and was usually
followed by judicial combat between the appellant and
ap^llee, each of whom fought in person. Such a right
was necessary in an age when the government had not
* See supray pp. 84-6. ' See Leges Henrici primi^ c. 69, §§ 15-16.
36o MAGNA CARTA
yet assumed a general responsibility for bringing ordinary
criminals to justice. The wronged person, not the magis-
trate, was the avenger of crime ; and this explains several
peculiarities — why, for example, when the accused had
uttered " that hateful word craven," ^ thus confessing him-
self vanquished and deserving a perjurer's fate, the victori-
ous accuser was entitled to his vengeance, even in face of
a royal pardon. When Henry of Essex, constable and
standard-bearer of Henry H., in 1163, had been worsted
in the combat, the royal favour could not shield him, though
the King's connivance enabled him, by becoming a monk,
and therefore dead in law, to escape actual death by hang-
ing.2 At an early date the procedure resembled even more
closely a legalized private revenge : " the ancient usage
was, so late as Henry IV. 's time, that all the relations of
the slain should drag the appellee to the place of execu-
tion." 3
The evils of trial by combat are obvious. From the first
it was dreaded by the traders of the boroughs, who paid
heavily for charters of exemption. Their aversion spread
to the higher classes, and was shared by Henry H. To
that statesman, endowed with the instincts of a reformer,
despising obsolete and irrational modes of procedure, and
devoid of reverence for tradition, trial by combat was
abhorrent. He would gladly have abolished it, but followed
the more subtle policy of undermining its vitality. For
this purpose, he used four expedients, which are of great
interest, in respect that they throw light on the process by
which trial by jury superseded trial by battle. * (i) Every
facility was afforded the parties to a civil suit to forego the
duellum voluntarily. Henry placed at their disposal, as a
substitute, a procedure which his ancestors had reserved for
the service of the Crown. Litigants might refer their rival
claims to the oath of a picked body of local neighbours:
the old recognitors thus developed into the jurata. This
' See Bracton, folio 531. ^ See Jocelyn of Brakelond, 50-2.
'Blackstone, Commentaries, IV. 316. Cf. Bateson, Borough CustomSy I. 73,
II. XXV., II. xxxiv.
*Cf. supra, p. 88, and also p. 272.
CHAPTER THIRTY-SIX 361
course was possible, however, only where both parties con-
sented, and it had many features in common with a modern
arbitration. (2) In pleas relating to the title and possession
of land, Henry went further, granting to the tenant the
option of a peaceful settlement even when the demandant
preferred battle. The assisa, like the jurata, applied only
to civil pleas. (3) Attempts were made to discourage trial
by combat in criminal pleas also, by discouraging private
" appeal," its natural prelude. The corporate voice of the
accusing jury was made to supersede the individual com-
plaint of the injured party. Only the near blood relation,
or the liege lord, of a murdered man was allowed to prove
the offender's guilt by combat; while a woman's right of
appeal was kept within narrow limits.^ (4) A wide field
was still left for private appeal arid battle; but Henry
endeavoured to narrow it by a subtle device. In appeals of
homicide, where the accusation was not made bona fide, but
maliciously or without probable cause, the appellee was
afforded a means of escaping the duelluni : he might apply
for the writ that forms the subject of this chapter.
II. The Writ of Life arid Limb. The writ here referred
to, better known to medieval England as the writ de odio
etatiaj^ was intended to protect from duel men unjustly
appealed of homicide. Many an appealed man was glad
to purchase escape by assuming the habit and tonsure of
a monk ; ^ but Henry desired to save innocent men from
the risk of failure in the duellum, without this subterfuge.
If the accused asserted that his appellant acted " out of
spite and hate" (de odio et atia), he might purchase from
the chancery a wTit to refer this preliminary plea to the
^ See under c. 54.
*In identifying the writ spoken of by Magna Carta as that "of life and limbs"
with the well-known writ de odio et atia, most authorities rely on a passage in
Bracton (viz., folio 123). There is still better evidence. The Statute of West-
minster, II. c. 29, ordains : '* Lest the parties appealed or indicted be kept long
in prison, they shall have a writ de odio et atia like as it is declared in Magna Carta
and other statutes." Further, in 1231, twelve jurors who had given a verdict as to
whether an appeal was false, were asked quo waranto fecerunt sacramentwn ilhid
de vita et menibris^ without the King's licence. See Bracton's Note-book, case 592.
' Madox, I. 505, has collected instances.
362 MAGNA CARTA
verdict of twelve recognitors. If his neighbours upheld
the plea, further proceedings were quashed: the duellum
was avoided.^ A similar privilege was afterwards extended
to those guilty of homicide in self-defence, or of homicide
by misadventure.2 Soon every man appealed of murder,
whether guilty or not, alleged as matter of course that he
had been accused maliciously, mere " words of common
form." Virtually, the main issue of guilt or innocence,
not merely the preliminary pleas, came to be determined by
the neighbours' verdict,^ which was treated as final. No
further proceedings were necessary : none were allowed.
The duellum had been elbowed aside, although it was not
abolished until 1819.*
III. Subsidiary Uses of the Writ. This inquest of life
and limb has been claimed as the direct antecedent of the
procedure which became so valuable a bulwark of the sub-
ject's liberty, under the name habeas corpus. This is a
mistake ; the modern writ of habeas corpus was developed
out of an entirely different writ, which had for its original
object the safe-keeping of the prisoner's body in gaol, not
his liberation from unjust confinement.^
The opinion generally, though erroneously, held, is not
without excuse; for the writ mentioned by Magna Carta
was put to a subsidiary use, which bears superficial resem-
blance to that of the habeas corpus. Considerable delay
might occur between the appellee's petition for the writ
of inquisition and the verdict upon it. In the interval, the
man accused of murder had no right to be released on bail,
a privilege allowed to those suspected of less grave crimes.
This was hard where the accused was the victim of malice,
or guilty only of justifiable homicide. Prisoners, in such
^ Cf. Pollock and Maitland, II. 585-7, and Thayer, Evidence, 68.
* Feudal courts adopted a similar procedure in malicious appeals (although the King
objected to their doing so without royal licence). Inquests were held shortly after
the abolition of ordeal (1215) in the court of the Abbot of St. Edmund. See
Bracton's Note-book^ case 592.
3 See Pollock and Maitland, II. 586. * 59 George III. c. 46.
*The early history of habeas corpus is traced by Prof. Jenks, Law Quarterly
Review^ VIII. 164. The writ de odio was obsolete prior to the invention of the
habeas corpus.
CHAPTER THIRTY-SIX 363
a plight, might purchase royal writs that would save them
from languishing for months or years in gaol. The writ
best suited for this purpose was that de_odio et_atia, since
it was already applicable to presumably innocent appellees
for another purpose.^
As trial by combat became rapidly obsolete, the original
purpose of the writ was forgotten, and its once subsidiary
object became more prominent. Before Bracton's day,
this change had taken place : the writ had come to be
viewed primarily as an expedient for releasing upon bail
homicides per infortunium or se defendendo. Bracton, in
giving the form of the writ,^ declares it to be iniquitous
that innocent men should be long detained in prison :
therefore, he tells us, an inquisition is wont to be made, at
the request of sorrowful friends, whether the accusation is
bona fide or has been brought de odio et alia. This pleasing
picture of a king moved to pity by tearful friends of accused
men scarcely applies to John, who listened only to suitors
with long purses : the writs that liberated homicides had
become a valuable source of revenue. Sheriffs were repri-
manded for releasing prisoners on bail without the King's
warrant, but, in spite of heavy amercements, they continued
their irregularities. Thus, in 1207, Peter of Scudimore
paid to the exchequer 10 marks for setting homicides free
upon pledges, without warrant from the King.^ In that
year, John repeated his orders, strictly forbidding man-
slayers to be set free upon bail until they had received
judgment in presence of the King's justices. ^
To John, then, the fees to be received for this writ, con-
stituted its greatest merit ; w^hereas the barons claimed, as
mere matter of justice, that it should be issued free of charge
to all who needed it. John's acceptance of their demands
was repeated in all reissues, and apparently observed in
practice. The procedure during the reign of Henry III. is
described by Bracton in a passage already cited. After the
writ de odio had been received, an inquest, he tells us, must
^Cf. Brunner, Schwurgerichtey 47 1. 2 ggg fQjjQ 12^.
3 See Pipe Roll, 8 John, cited Madox, I. 566.
*See Rot. Pat., I. 76 ; Madox, I. 494. The date is 8th Nov., 1207.
364 MAGNA CARTA
be held speedily, and if tlie jury decided that the accusation
had been made maHciously, or that the slaying had been in
self-defence or by accident, the Crown was to be informed
of this. Thereafter, from the chancery would be issued a
second writ (known in later days as the writ tradias in bal-
lium)j directing the sheriff, on the accused finding twelve
good sureties of the county, to " deliver him in bail to those
twelve " till the arrival of the justices.
It should be noted that the provision granting gratuitous
writs was not construed as forbidding payments made by an
accused man for a special form of " trial." Prof. Maitland
has shown how " occasionally a person pays money to the
King that he may have an inquest, and it would seem that
he might still buy the right to be tried by a body constituted
in some particular way. He might pay to be tried by the
jurors of two hundreds, or of three hundreds, and because of
local enmities such a payment may sometimes have been
expedient."^ A certain Reginald, Adam's son, in 1222,
offered one mark for a verdict of the three neighbouring
counties (it was a Lincolnshire plea), as to whether the
accusation was made because of " the ill-will and hate "
which William de Ros, appellant's lord, bore to Reginald's
father " vel per verum appellum." ^
A long series of later statutes enforced or modified this
procedure. These have been interpreted to imply frequent
changes of policy, sometimes abolishing and sometimes re-
introducing the writ and the procedure which followed it.^
This is a mistake ; the various statutes wrought no radical
change, but merely modified points of detail ; sometimes
seeking to prevent the release of the guilty on bail, and
sometimes removing difficulties from the path of the inno-
^ Gloucester Pleas, xli., where cases are cited.
2 See Bracton's Note-book, case 1 34, and cf. case 1 548.
2 Stephen, Hist. Crim. Law, I. 241 (following Foster, Ci-im. Casesy 284-5),
considers that it was abolished by 6 Edward I., stat. i, c. 9. Coke, Second
Institute, 42, thought it was abolished by 28 Edward III. c. 9 (which, however,
seems not to refer to this at all), and restored by 42 Edward III. c. I (abolishing
all statutes contrary to Magna Carta). Coke, ibid., and Hale, Pleas of the Crown,
II. 148, considered that the writ was not obsolete in their day. Cf. Pollock and
Maitland, II. 587 n.
I
CHAPTER THIRTY-SIX 365
cent. The Statute of Westminster, I., for example, after a
preamble, which animadverted on sheriffs impanelling
juries favourable to the accused, provided that inquests
" shall be taken by lawful men chosen by oath (of whom two
at least shall be knights) which by no affinity with the
prisoners nor otherwise are to be suspected." ^ The Statute
of Gloucester, on the other hand, ordered the strict confine-
ment, pending trial, of offenders whose guilt was apparent.^
The Statute of Westminster, II., once more favoured
prisoners, providing by chapter 12 for the punishment of
false appellants or accusers, and by chapter 29 that " lest the
parties appealed or indicted be kept long in prison, they
shall have a writ of odio et atia, like as it is declared in
Magna Carta and other Statutes."^ The writ in question
was in use in 1314,^ and seems never to have been expressly
abolished, but to have sunk gradually into neglect, as
appeals became obsolete and gaol deliveries were more
frequently held.
IV. Later History of Appeal and Battle. The right of
private accusation was restricted, not abolished, by Henry
II. and his successors. It could not be denied to an injured
man who was not suspected of abusing his right. Prosecu-
tions by way of indictment and jury trial supplemented,
without superseding, private prosecutions by way of appeal
and battle. The danger of a second prosecution might
hang over the head of an accused man after he had " stood
his trial " and been honourably acquitted. It was unfair
that he should be kept in such prolonged suspense; and,
accordingly, the Statute of Gloucester provided that the
right of appeal should lapse unless exercised within year
and day of the commission of the offence.^ To obviate all
^Edward I. c. Ii. ^6Edwa.Yd I., stat. i, c. 9.
3 13 Edward I. cc. 12 and 29. *See Rot. Pari., I. 323.
^ 6 Edward I. c. 9. Appeals were extremely frequent towards the close of the
Plantagenet period, especially in the days of " the Lords Appellant." The pro-
ceedings on appeal sometimes took place before the Court of the Constable and
Marshal and sometimes before Parliament. In neither case were they popular.
One of the charges brought against Richard II. was that "in violation of Magna
Carta " (that is, probably, of chapter 39) persons maliciously accused of treasonable
words were tried before constable and marshal, and although "old and weak,
366 MAGNA CARTA
risk of a double prosecution, it was necessary that the Crown
should delay to prosecute until the year and day had
expired. This rule was followed in 1482. Such immunity
from arraignment for twelve months would have produced
a worse evil, by facilitating the escape of criminals from
justice. After experience of its pernicious effects, the rule
was condemned by the act of parliament which instituted the
Star Chamber.^
This remedied the more recent evil, but revived the old
injustice : the same statute enacted that acquittal should not
bar appeal by the wife or nearest heir of a murdered man.
Thus, once again, a man declared innocent by a jury might
find himself exposed to a second prosecution. In 181 7 the
British public was startled to find that a long-forgotten
procedure of the dark ages still formed part of the law of
England. The body of a Warwickshire girl, Mary Ash-
ford, was discovered in a pit of water under circumstances
that suggested foul play. Suspicion fell on Abraham
Thornton. After indictment and trial at Warwick Assizes
on a charge of rape and murder, he was acquitted. The
girVs brother, William Ashford, not satisfied by what was
apparently an honest verdict, tried to secure a second trial,
and claimed the appeal of felony, which the judges did not
refuse. Ashford's attempt to revive this obsolete procedure
was met by Thornton's revival of its equally obsolete
counterpart. Summoned before the judges of King's
Bench, he offered to defend himself by combat, throwing
down as " wager of battle " a glove of approved antique
pattern. Lord Ellenborough had to admit his legal right
to defend himself against the appeal "by his body," and
Thornton successfully foiled the attempt to force him to a
second trial, as Ashford, a mere stripling, declined the
maimed or infirm," yet compelled to fight against appellants "young, strong, and
hearty." See /^of. ParL^ III. 420, cited Neilson, Trial by Combat, 193. On the
other hand, Statute i Henry IV. c. 14, provided that no appeals should be held
before Parliament, but certain appeals might come before constable and marshal.
Cf. Harcourt, Steward, 369.
^See 3 Henry VII. c. i, s. II : the injured party, with the right of appeal, was
*' oftentimes slow and also agreed with, and by the end of the year all is forgotten
which is another occasion of murder."
I
CHAPTER THIRTY-SIX 367
unequal contest with an antagonist of atliletic build.^ The
unexpected revival of these legal curiosities led to their final
suppression. In 1819 a Statute abolished proof by battle
alike in criminal and in civil pleas : the right of appeal fell
with it.2
CHAPTER THIRTY-SEVEN.
Si aliquis teneat de nobis per feodifirmam, vel per
sokagium, vel per burgagium, et de alio terram teneat per
servicium militare, nos non habebimus custodiam heredis
nee terre sue que est de feodo alterius, occasione illius feodi-
firme, vel sokagii, vel burgagii ; nee habebimus custodiam
illius feodifirme, vel sokagii, vel burgagii, nisi ipsa feodi-
firma debeat servicium militare. Nos non habebimus
custodiam heredis vel terre alicujus, quam tenet de alio per
servicium militare, occasione alicujus parve serjanterie
quam tenet de nobis per servicium reddendi nobis cultellos,
vel sagittas, vel hujusmodi.
If anyone holds of us by fee-farm, by socage, or by burgage,
and holds also land of another lord by knight's service, we will
not (by reason of that fee-farm, socage, or burgage,) have the
wardship of the heir, or of such land of his as is of the fief of
that other ; nor shall we have wardship of that fee-farm, socage,
or burgage, unless such fee-farm owes knight's service. We will
not by reason of any small ^ serjeanty which anyone may hold
of us by the service of rendering to us knives, arrows, or the like,
have wardship of his heir or of the land which he holds of
another lord by knight's service.
In these provisions the Charter reverts to the subject of
wardship, laying down three rules, which will be better
understood when their sequence is altered, the second being
taken first, (i) Ordinary wardship. The reason for claim-
ing wardship from lands held in chivalry, namely, that a
^See Ashford v. Thornton, i B. and Ald.^ 405-461.
^See 59 George III. c. 46.
' Pollock and Maitland, I. 304, read *■'■ parva " as an untechnical word. Round,
Serjeanties^ 35-6, finds in this chapter the origin of the distinction between '* grand "
and "petty" serjeanties, and compares the distinction made in c. 14 between
greater and lesser barons.
368 MAGNA CARTA
boy could not perform military service, did not apply to fee-
farm, socage, or burgage. There was much looseness of
usage, however; and of this John took advantage. The
Charter stated the law explicitly; wardship was not due
from any such holdings, except in the anomalous case where
lands in fee-farm expressly owed military service.^ As
petty serjeanties (although mentioned in the present chapter
in a different connection) are not expressly said to share this
exemption, it may be inferred that the barons admitted
John's wardship over them, as over great serjeanties. By
Littleton's time, the law had changed : petty serjeanties
were then exempt.^
(2) Prerogative wardship. When the heir of a tenant-in-
chivalry held military fiefs of different mesne lords, each of
these lords enjoyed wardship over his own fief. This was
fair to all parties : but, if the ward held one estate of the
Crown, and another of a mesne lord, the King claimed
wardship over both; and that, too, even when the Crown
fief was of small value.^ Such rights were known as " pre-
rogative wardship," and, thus limited, were in 12 15 perfectly
legal, however inequitable they may now seem, (a) Fee-
farm, socage, and burgage. John, however, pushed this
right further, and exercised prerogative wardship over fiefs
of mesne lords, not merely by occasion of Crown fiefs held
in chivalry, but also by occasion of Crown fiefs held by any
tenure. It was outrageous to claim prerogative wardship
in respect of fee-farm, socage, or burgage lands, which were
exempt even from ordinary wardship. John was made to
promise amendment.^ (h) Small Serjeanties^ were in a
different position. Magna Carta did not abolish the
Crown's rights of ordinary wardship over these, but forbade
*Cf. supra, pp. 55-7 and 6i-2. ^ II. viii. s. 158.
'Cf. Glanvill, VII. c. 10. "When any one holds of the King in capite the
wardship over him belongs exclusively to the King, whether the heir has any other
lords or not ; because the King can have no equal, much less a superior. " Yet the
King is not to have such wardship *' because of burgage."
*Cf. Petition of Barons (1258), c. 2; Prro. of West. {1259), c. 12. Glanvill,
VII. c. 10, had laid it down that burgage tenure could not give rise to prerogative
wardship.
' See supra, p. 56.
CHAPTER THIRTY-SEVEN 369
that this should form an excuse for prerogative wardship
over the wider fiefs of other lords. ^
Prerogative wardship (even in the limited form admitted
by Magna Carta) might involve a double hardship on the
mesne lord. Suppose that the common tenant held lands
from a mesne lord on condition of say, five knights' service,
as well as his Crown fief. The King seized both fiefs on his
death, nominally as a compensation for the loss of military
service, which the minor heir could not render. Yet, when
a scutage ran, the King demanded from the mesne lord
payments in proportion to his full quota without allowing
for the fees of five knights taken from him by prerogative
wardship. This is no imaginary case : the barons in 1258
complained of the practice and demanded redress.^
CHAPTER THIRTY-EIGHT.
NuLLUS ballivus ponat de cetero aliquem ad legem simplici
loquela sua, sine testibus fidelibus ad hoc inductis.
^ See Bracton, folio 87b. The Note-booh,, case 743, contains a good illustration.
The motive for these restrictions was to prevent injustice to mesne lords. It was
probably, however, an indirect consequence of Magna Carta that a similar rule
came to be applied where no mesne lord was injuriously affected. In 1 231 a certain
Ralf of Bradeley died, who had held two separate freeholds of the Crown, (i) a
small fee by petty serjeanty of twenty arrows a year, and (ii) land of considerable
value held in socage. The Crown took possession of both estates, on the assump-
tion that wardship over the petty serjeanty brought with it a right of wardship over
the socage lands also (although these would have been exempt if they had stood
alone). The King sold his rights for 300 marks. Ralf's widow claimed the ward-
ship of the socage lands, on the ground that these were of much greater value than
those held by serjeanty. Her argument was upheld, and the 300 marks refunded
to the disappointed purchaser. See Pipe Roll, 5 Henry III., cited Madox, I.
325-6.
^ See Petition of the Barons, Article 2 {Select Charters, 383). C. 53 of Magna
Carta reverts to prerogative wardship, granting redress, although not summary
redress, where John, or his father or brother, had illegally extended it by occasion
of socage, etc. See also supra, p. 368. Round, Eng. Hist. Rev., XXVIII. 156,
cites from Cal. lug. post mortem. III. 406-7, an interesting case of prerogative
wardship decided against the Crown in 1301. Orpen, Ireland, II. 234, cites two
Charters in which John renounces prerogative wardship. C. 43 infra (amended by
c. 38 of 1217) guards against another abuse of prerogative wardship.
2 A
370 MAGNA CARTA
No bailiff for the future shall, upon his own unsupported
complaint, put anyone to his " law," without credible witnesses
brought for this purpose.
The exact nature of the abuse here condemned has been
much discussed by commentators. Bailiffs (the word is
probably used here in its widest sense i) were wont to abuse
their authority : henceforth they shall put no man to his
" lex " on their own initiative. The word lex, in its
technical sense, applied to any form of judicial test,
such as compurgation, ordeal, or combat, the precise
meaning required in each particular case being determined
by the context.^ In the present chapter it seems to have
this technical meaning of a judicial " proof " or " trial " of
any sort:^ henceforward no bailiff should have power
" simplici loquela sua " ^ to put anyone to a " lex" of any
kind. Authorities differ as to the exact nature of the
irregularities which this clause was meant to suppress.
I. Medieval Interpretations, Ignorance of the exact
* Cf. supra, c. 24. It possibly includes sheriffs and their officers. The same
men, apparently, were described as King's Serjeants and sheriff's Serjeants ; one
Roll records fines for a man buried '''■sine visu servientum vicecomitis,^'' and for a
robber hanged " sine visu servientis regis'''' {Pipe Roll, 31 Henry II). The word
may also include the stewards who presided in manorial courts. If so, the un-
qualified " ballivus " of this passage should, perhaps, be contrasted with the ^^noster
ballivus^^ of cc. 28 and 30. Coke, Second Institute, 44, following the doubtful
Mirror of Justices, extends it to all King's justices and ministers.
*Dr. Stubbs {Const, Hist., I. 576) translates *'/^jt" in this passage by "com-
purgation or ordeal." Pollock and Maitland (II. 604 n.) explain that the word
*'does not necessarily point to unilateral ordeal; it may well stand for trial by
battle." Thayer {Evidence, 199-200) extends it to embrace judicially appointed
tests of every kind — battle, ordeal of fire or water, simple oath, oath with com-
purgators, charter, transaction witnesses, or sworn verdict. Bigelow {Placita
Anglo- Normanica, 44) cites from Domesday Book cases where litigants offered
proof omni lege or omnibus legibus, that is, in any way the court decided. Some-
times lex had a more restricted meaning ; in the Customs of Newcastle-on-Tyne
{Select Charters, 112) it seems to mean compurgation as opposed to combat. For
its various meanings see also Harcourt, Steward, 232.
^In c. 55 **/!?j:" would seem to bear a meaning more akin to the broader con-
ception of "law" in modern jurisprudence; while in c. 39 its denotation is
subject of controversy.
*Cf. the phrases ^^ per simplex vcrbtim smim'' (Ford wick) and '•'■per vocetii
suam simplice?n" (Hereford) in Bateson, Borottgh Customs, I. 181. Cf. ibid., IL
xxxii.
CHAPTER THIRTY-EIGHT 371
nature of the abuse prohibited may well be excused at the
present day, since it had become obscure within a century
of the granting of the Charter. Some legal notes of the
early fourteenth century, containing three alternative sug-
gestions, have come down to us.^
(i) The first interpretation discussed, and apparently
dismissed, in these notes, was that Magna Carta by this
prohibition wished to ensure that no one should serve on a
jury {in juratam) unless he had been warned by a timely
summons. This far-fetched suggestion is clearly erroneous.
(2) The next hypothesis raised is that the clause prevented
the defendant on a writ of debt (or any similar writ) from
winning his case by his unsupported oath, where compur-
gators ought to have sworn along with him. Exception
was, in this view, taken to the bailiff treating favoured
defendants in civil pleas with unfair leniency.
(3) A third opinion is stated and eulogized as a better
one, namely, that the Charter prohibited bailiffs from show-
ing undue favour to plaintiffs in civil pleas. The defendant
on a writ of debt (or the like) should not, in this interpreta-
tion of Magna Carta, be compelled to go to proof at all (that
is, to make his " law ") unless the plaintiff had brought
" suit " against him (that is, had raised a presumption that
the claim was good, by production of preliminary witnesses
or by some recognized equivalent). 2
II. Modern Interpretations. If the chapter is read in a
broad sense as prohibiting abuses of a generic kind, it is
possible that more than one of its modern exponents may
be substantially correct, in spite of apparent contradictions,
(i) One theory would read the clause as forbidding magis-
trates to show undue favour to defendants of certain classes.
Crown officials, under John, it is pointed out, favoured
Jews against Christians with whom they went to law. The
* These appear as an Appendix to the Year Book of 32-3 Edward I. (p. 516) ;
but the handwriting is supposed to be of the reign of Edward II.
^Cf. supruy p. 83. The necessity for such "suit" was not legally abolished
until 1852 (by Statute 15 and 16 Victoria, c. 76, s. 55). In 1343 it had been
decided that the " suit" must be in existence, but need not be produced in court ;
and that if they did appear they could not be examined. See Thayer, Evidencey
13-15-
372 MAGNA CARTA
Hebrew defendant in a civil suit " might purge himself by
his bare oath on the Pentateuch, whereas in a similar case a
Christian, as the law then stood, might be required to wage
his law twelve-handed — i.e. with eleven compurgators." ^
Magna Carta, it has been suggested, struck at this pre-
ferential treatment of Jewish litigants, trebly hated as aliens,
capitalists, and rejectors of Christ. If so, the attempt
failed; for in 1275 a certain Hebrew, named Abraham, was
allowed " to make his law single-handed on his Book of the
Jewish Law " in face of the plaintiff's protest that this was
contrary to the custom of the realm. ^
(2) On the other hand, the clause is sometimes made to
prohibit undue favour shown to demandants in civil suits to
the prejudice of defendants. A " suit " of witnesses (secta-
tores) had to be produced in court by the plaintiff before
any " trial " (lex) could take place at all. Bailiffs were
forbidden to allow, through slackness, favour, or bribery,
this rule to be relaxed. This interpretation, which was
adopted by the author of the Mirror of Justices, and by the
writer of the notes appended to the Year Book already cited,
found favour with Chief Justice Holt in 1700.^
(3) A closely allied explanation treats the clause not as
forbidding undue favour towards one party to an action, but
rather as preventing bailiffs from favouring themselves.
When it suited them, the King's officials were wont to
dispense with the wholesome rule that demanded " suit " or
its equivalent before a plea could be entertained. This
practice was by no means confined to England, and has
been discussed by Dr. Brunner.*
^See Rigg's Sel. Pleas Jezvish Exch.^ xii., and cf. sttpra, c. 10.
2 Rigg, ibid., 89, where the case is cited.
3 See City of London v. Wood (12 Modern Reports, 669). Holt held the clause
of Magna Carta to mean that the plaintiff, unless he had witnesses, could not put
a defendant to his oath. Pollock and Maitland, II. 604, seem to concur, to the
extent at least of counting this as one of the abuses condemned by c. 38 : * * The
rule %hich required a suit of witnesses had been regarded as a valuable rule ; in
12 1 5 the barons demanded that no exception to it should be allowed in favour of
royal officers."
*See his Schwurgerichte^ 199-200. Cf. ibid., 178 and 409-74. For a similar
practice in Galloway, see G. Neilson on **Surdit de Sergaunt," Scot. Antiq., XI.
CHAPTER THIRTY-EIGHT 373
(4) It is perhaps only another aspect of the same explana-
tion to regard the clause as directed mainly against unfair
treatment of accused men in criminal prosecutions. No one
ought to be put to his " lex," in the sense of " ordeal," on
mere grounds of vague suspicion or on the unsupported
statement of a royal bailiff. After 1166, at least, the voice
of an accusing jury of neighbours was a necessary pre-
liminary, under normal circumstances, before any one could
be put to the ordeal in England. Magna Carta confirmed
this salutary rule : no bailiff should put any one to the
ordeal except after formal indictment, due evidence of which
was presented at the diet of proof .^
III. Nature of the grievance. As already suggested, it
seems not unlikely that two or more of these theories may
require to be combined in order to furnish a complete
explanation of the clause under discussion. Magna Carta
may well have condemned alike the practice of compelling a
man to defend a civil action unsupported by suit, and of
sending him to the dreaded ordeal without indictment by
his neighbours.
To the criminal aspect of the matter, the Assize of
Clarendon (1166) seems to supply the key. Article 4
of that ordinance prescribes the procedure for try-
ing robbers, thieves, and murderers : " the sheriff shall
bring them before the justices; and with them they shall
bring two law-worthy men of the hundred and of the village
where they were apprehended, to bear the record of the
county and of the hundred, as to why they had been appre-
hended ; and, there, before the justices they shall make their
law." This "law" is elsewhere in the ordinance clearly
155. The Leges Quatuor BurgoT^m would seem to guard against an evil of an
opposite kind when (c. 76) they forbid the provost or bedells of a town ^prepositus
vel precones) to "bring witnesses to a claim against anyone," but direct that the
defendant shall acquit himself pei' legetti. This peculiar law would seem to be
entirely unknown to previous commentators on this difficult passage of Magna
Carta.
^ This reading is supported by Pollock and Maitland, I. 130 n. There is no
necessary inconsistency between the view here cited, and that already cited from
ibid., II. 604. The same clause of Magna Carta may have been aimed at
irregularities of two kinds, in civil and criminal pleas respectively.
374 MAGNA CARTA
identified with ordeal ; ^ and the purport of the whole was
that accused men could not be put to ordeal except in
presence of two lawful men who had been present at the
indictment and had come before the justices specially to
bear witness thereof. In other words, the sheriff's own
report of the indictment " sine testihiis fidelibus ad hoc
inductis " was not sufficient. The " county " and the
" hundred " which had heard the prisoner accused, must
send representatives to bear record of the facts.^
The ordeal was a solemn affair, and every precaution
must be taken against its abuse. Sheriffs or other royal
bailiffs must be present, as well as members of the accusing
jury. Lords of feudal courts, claiming this franchise,
required apparently royal warrant for its exercise.^ Prac-
tice, however, was loose : the King's justices would seem to
have had a right to put suspects to the ordeal ex officio
without the intervention of the accusing jury :* sheriffs and
others, with the Crown's approval or connivance, exercised
a similar privilege. In condemning these practices. Magna
Carta would appear to have been, to some extent, modifying
previous usage. ^ It was not enough thereafter that indict-
ment should precede ordeal ; members of the presenting
jury, who had made the accusation at the first diet, must
accompany the sheriff before the justices at the final diet,
there to bear testimony both as to the nature of the crime
and as to the fact of the indictment. Before anyone could
■• See Article 12 where ** eat ad aquam " is contrasted with *' non habeat legem "
of Article 13 {Select Charters^ 144).
*The ^^ ad portandum recordationem comitatus et hundredV of the ordinance is
exactly opposed to the *' simplex loquela sua'''' of the Charter.
2 Thus in 1 166 (the year of the Assize of Clarendon) the *' Soca " of Alverton was
amerced because of a man placed ^^ ad aquam sine servient e''^ {Pipe Roll, 12
Henry II., p. 49). In 11 85 the '^villata" of Preston paid 5 marks for putting a
man ^^ ad aquam sine waranto'''' {Pipe Roll, 31 Henry II., cited Madox, I. 547).
In the same year a certain Roger owed half a mark for being present at an ordeal
'■^ sine visu servientum regis'''' : and heavy fines were exacted from those who had
put a man ** injuste ad aquajn " {ibid.).
<See Miss Bateson, Eng, Hist. Rev., XVII. 712.
^Miss Bateson {Borough Customs, II. xxxi.) speaks of the "right of accusation
' ex officio ' which belonged to the King's officers until Magna Carta, Art. 38,
deprived them of it. "
CHAPTER THIRTY-EIGHT 375
be put "to his law," the sheriff's formal report must be
corroborated by the testimony of representative jurors.
The Charter of 1216 repeated this provision without
alteration. In 121 7 a change occurred, which was un-
doubtedly a consequence of the virtual abolition of the
ordeal by the Lateran Council in 12 15. The framers of
Henry's second reissue found leisure to adjust points of
administrative detail. The simple reference to ordeal was
inappropriate now that new forms of trial were taking its
place. The justices, indeed, scarcely knew what test to
substitute for ordeal. They seem sometimes to have
resorted to compurgation and sometimes to battle; but the
sworn verdict of neighbours was fast occupying the ground
left vacant. The Charter of 12 17, then, made it clear that
the provisions applied in 1215 to ordeal were to be extended
to other tests. The " ad legem " of John's Charter became
in the new version " ad legem manifestam. nee ad juramen-
tum." A " manifest law " might well mean either ordeal
or any other actual physical test such as "battle," ^ while
^* juramentum " points to the sworn testimony of the jury,
which was slowly taking the place of the discredited ordeal. ^
CHAPTER THIRTY-NINE.
NuLLUS liber homo capiatur vel imprisonetur, aut dis-
seisiatur, aut utlagetur, aut exuletur, aut aliquo modo
destruatur, nee super eum ibimus, nee super eum mittemus,
nisi per legale judicium parium suorum vel per legem terre.
No freeman shall be taken or [and] imprisoned or disseised
or exiled or in any way destroyed, nor will we go upon him nor
send upon him, except by the lawful judgment of his peers or
[and] by the law of the land.^ ^' •-^^-.-^--■•■'.-^ -•.:.,,.., ^
^ See Thayer, Evidence^ 37 n., for a case of 1291, where ^* ad legem manifeslatn "
can only mean trial by combat. Cf. legem apparentem purgandtts est in Glanvill,
XIV. ff. 112-114.
2 Westminster I. (c. 12) described men refusing to put themselves on a jury's
ve.-cict, ^^ come ceaus qui refusent la commtcne ley de la terre.^^
3 The usual English rendering has here been followed : Mr. Harcourt (Steward,
219) was possibly right in holding that " interpretation under the guise of translation
376 MAGNA CARTA
This chapter occupies a prominent place in law-books,
and is of considerable importance, although its value has
sometimes been exaggerated.^
I. Its Main Object, It has been usual to read it as a
guarantee of trial by jury to all Englishmen ; as absolutely
prohibiting arbitrary commitment ; and as solemnly under-
taking to dispense to all and sundry an equal justice, full,
free, and speedy .^ The traditional interpretation has thus
made it, in the widest terms, a promise of law and liberty
and good government to every one.^^ A careful analysis of
the clause, read in connection with its historical genesis,
suggests the need for modification of this view. It was in
accord with the practical genius of the Charter that it should
here direct its energies, not to the enunciation of vague
platitudes, but to the reform of a specific abuse. Its object
was to prohibit John from resorting to what is sometimes
whimsically known in Scotland as " Jeddart justice."^ It
is in this ease an inevitable snare." ' This does not, however, absolve the com-
mentator from explaining the text. The Articles of the Barons (29) add "z'z"
{^^ nee rex eat vel niittat super eum vV suggesting the fuller contemporary '''■per
vim et anna "). This shows the inadequacy of the translation contained in the
Statutes at Large^ *'nor will we pass upon him nor condemn him." The Statutes
of the Reahn, I. 117, suggest "deal with him" as an alternative. Coke, as
explained infra, originated the error which thus connected " going " and " sending "
with legal process.
*For a valuable discussion of alternative interpretations, see Adams, Origin,
256-274 ; also Pike, House of*Lords, c. X. Mr. Harcourt's learned 'discussions
[Steward, cc. VII. and VIII.) are worthy of careful study, though they are more
useful in suggesting difficulties than in finding solutions. •
2 See, e.g. Coke, Second Institute, 55. *
. ^Thus Blackstone, Commentaries, IV. 424 : "It protected every individual of
the nation in the free enjoyment of his life, his liberty, and his property, unless
declared to be 'forfeited by the judgment of his peers or the law of the land."
Hallam, Middle Ages, II. 448, speaking of cc. 39 and -'40 together, says they
"protect the personal liberty and property of all freemen by giving- ^security from
arbitrary imprisonment and arbitrary spoliation." Creasy, Eng. Const., p. 151 n. :
"The ultimate effect of this chapter was to give and to guarantee full protection
for person and property to every human being that breathes English air."
* The same grim tradition applied to Lidford as to Jedburgh :
"I oft have heard of Lydford law.
How in the morn they hang and draw,
And sit in judgment after."
See Neilson, Trial by Combat, 131, and authorities there cited.
CHAPTER THIRTY-NINE 377
forbade him for the future to place execution before
judgment. Three aspects of this prohibition may be
emphasized.
(i) Judgment must precede execution. In some cases
John proceeded, or threatened to proceed, by force of arms
against recalcitrants as though assured of their guilt, with-
out waiting for legal procedure. ^ Complaint was made of
arrests and imprisonments suffered " without judgment "
(absque judicio) ; and these are the very words of the
" unknown charter " — " Concedit Rex Johannes quod non
capiet homines absque judicio.""^ The Articles of the
Barons and Magna Carta expand this phrase. Absque
judicio becomes nisi per legale judicium parium suorum vel
per legem terrae, thus guarding, not merely against execu-
tion without judgment, but also against John's subtler
device for attacking his enemies by a travesty of judicial
process. The Charter asks not only for aJ^Judgmenf^^U^if
for a " judgment of peers " and " according to the law of the
land." Two species of irregularities were condemned by
these words ; and these will be explained in the two follow-
ing subsections.
(2) Per judicium parium : every judgment must be
delivered^bytke accused man's "equals."^ The need for
" a judgment of peers " was recognized at an early date in-r
England.^ It was not originally a class privilege of the
aristocracy, but a right shared by all grades of free-holders ;
^Mr. Bigelovv considers that such cases were numerous. S&q Procedm-e, 155:
"The practice of granting writs of execution without trial in the courts appears to
have been common."
^See Appendix.^
^ Mr. Harcourt [Steward, 218 ff. ) has much to say on this phrase : for him a man's
"peers" need not be his equals in rank (p. 220) ; while "judgment" is a vague
word embracing widely opposed procedures : e.g. (p. 248), " In common parlance
of the time a resolution of the King in Council to make war on a subject was a
judicmm.^' He further instances, as examples of legal processes accepted in 121 5
as equivalent to "judgment," the procedure for Crown debts under c. 9 ; outlawry
under c. 42 ; the petty assizes under c. 19 ; and the special procedure in cc, 52, 56
and 59 (see ibid., 220-3). Mr. Harcourt's conclusions are not clearly formulated,
and some of them appear to be not well founded.'
*The earliest known reference occurs in the Leges Henrici (c. 31) : Umisquisque
per fares suos judicandzcs est et ejitsdem provinciae.
378 MAGNA CARTA
whatever their rank, they could not be tried by their
inferiors.^ In this respect English custom did not differ
from the procedure prescribed by feudal usage on the Con-
tinent of Europe.^ Two applications of this general prin-
ciple had, however, special interest for the framers of Magna
Carta : the " peers " of a Crown tenant were his fellow
Crown tenants, who would normally deliver judgment in
the Curia Regis; while the "peers" of the tenant of a
mesne lord were the other suitors of the Court Baron of the
manor; In either case, judgments were given per pares
curiae. John, resorting wholesale to practices used spar-
ingly in earlier reigns, had set these rules at defiance. His
political and personal^ enemies were e^dledv o^ deprived of
j! their estates^by^ the judgment -M._a._tr^^ composed
l^entirely of Crown nominees. Magna Carta promised a
-—-return to the ancient practice.
The varied meanings conveyed by the word " peers " to
a medieval mind, together with the nature of judicium
parium, may be further illustrated by the special rules
applicable to four exceptional classes of individuals : — (a)
Jews of England and Normandy enjoyed under John's
Charter of loth April, 1201, the right to be judged by men
of their own race ; for them a judicium parium was a judg-
ment of Jews.^ (h) A foreign merchant, by later statutes,
obtained the right to a jury of the " half tongue " (de medie-
tate linguae), composed partly of aliens of his own country.^
(c) The peers of a Welshman seem, in some disputes with
the Crown, to have been men drawn from the marches :
such at least is the plausible interpretation of the phrase " in
^Cf. Pollock and Maitland, I. 152. As there was no "peerage" in England
(cf. su/>ra, p. 186) until long after John's reign, it is obvious that the judumm
paritwi of Magna Carta must be interpreted in a broader sense than any mere
** privilege of a peer " at the present day. Freeholders holding of the same mesne
lord were '* peers of a tenure."
2 See Stubbs, Const. Hist.^ I. 578 n., for foreign examples oi judicium parium.
3 " If a Christian bring a complaint against a Jew, let it be adjudged by his
peers of the Jews." See Rot. Chartarnm, p. 93, and stipj'a, p. 227 n. Harcourt,
however {ibid., 228), trax\s\Q.ies pares Judei as "justices or custodes of the Jews."
* See Carta Mercatoria, c. 8 ; 27 Edward III. stat. 2, c. 8 ; and 28 Edward III.
c. 13 ; also Thayer, Evidence, p. 94.
CHAPTER THIRTY-NINE 379
Tnarchia per judicium parium suorurn," occurring in later
chapters of Magna Carta, and granting to the Welsh redress
of wrongful disseisins. ^ (d) A Lord ]\Iarcher occupied a
peculiar position, enjoying rights denied to barons whose
estates lay in more settled parts of England. In 1281 the
Earl of Gloucester, accused by Edward I. of a breach of
allegiance, claimed to be judged, not by the whole body of
Crown tenants, but by such as were, like himself, lords
marchers.^ These illustrations show that a " trial by peers "
had a wider and less stereotyped meaning in the Middle
Ages than it has at the present day.^
(3) Per legem terrae. No freeman could be punished
except " in accordance with the law of the land." The
precise meaning of these often-quoted words ought, per-
haps, still to be regarded as an open question. Two mean-
ings are possible : one, narrow and technical ; the other, of
a loose and popular bearing. The more technical has
already been explained.^ Thus interpreted, the words of
John's Charter promised a threefold security to all the free-
men of England. Their persons and property were pro-
tected from the King's arbitrary will by the rule that
execution should be preceded by a judgment — by a judg-
ment ot peers^by aTjudgment according to the appropriate
time-honoured "lest," battle, compurgation, or ordeal.^
^ See infra, cc. 56, 57, and 58. Under c. 59 the barons of England were called
peers of the King of Scots.
^ See Placitortim Abbrevatio, p. 201, cited Pollock and Maitland, I. 393 n.
'See also a passage in the Scots Acts of Parliament (I. 318) attributed to
David : "No man shall be judged by his inferior who is not his peer; the earl
shall be judged by the earl, the baron by the baron, the vavassor by the vavassor,
the burgess by the burgess ; but an inferior may be judged l:>y a superior."
*See supra, p. 84, and cc. 18, 36, and 38.
^See Thayer, Evidence, 200-1, for a discussion of the phrase '■'■lex terrae^ See
also Bigelow, History of Procedure, 155 n. : "The expression ^ per legem terrae^
simply required judicial proceedings, according to the nature of the case ; the
duel, ordeal, or compurgation, in criminal cases ; the duel, witnesses, charters, or
recognition in property cases." The words occur at least twice in Glanvill, each
time apparently with the technical meaning. In II. c. 19, the penalty for a false
verdict includes forfeiture by jurors of their law {^^ legem terrae ajuittentes") ;
while in V. c. 5, a man born a villein, though freed by his lord, cannot, to the
prejudice of any stranger, wage his law {'^ ad aliquam legem terrae faciendam ").
The stress placed on the accused's right to the time-honoured forms of lex is well
38o MAGNA CARTA
Much weight, however, must be allowed to the arguments
of those who contend for interpreting '" lex terrae " more in
accordance with the vague and somewhat meaningless " law
of the land " of popular speech at the present day. The
phrase, they argue, was not confined to methods of pro-
cedure, but referred to the entire tone and substance of the
law.i Advocates of both theories can point to other parts
of Magna Carta where " lex " is used in the sense they claim
for it in the present passage; for its purport was, in 12 15,
ambiguous. In chapters 18, 36, and 38, it refers primarily
to procedure, whereas chapters 9, 45, 52, 56, and 59 suggest
a broader interpretation.
Magna Carta is undoubtedly a loosely drawn document,
and it is always possible that both meanings were in the
minds of the framers. If so, the older, more technical
signification was gradually forgotten, and " the law of the
land " became the vague and somewhat meaningless phrase
of the popular speech of to-day. It was only natural that
this change of emphasis should be reflected in subsequent
statutes reaffirming, expanding, or explaining Magna
Carta. An important series of these, passed in the reigns
of Edward III. and Richard II., shows how the per legem
terrae of 12 15 was read in the fourteenth century as equiva-
lent to " by due process of law, " and how the Great Charter
was interpreted as prohibiting the trial of men for their lives
and limbs before the King's Council on mere informal and
irresponsible suggestions, sometimes made loosely or from
malicious and interested motives. ^
illustrated by the difficulty of substituting jury trial for ordeal. It has already
been shown that the right of "standing mute," that is, virtually, of demanding
ordeal, was only abolished in 1772. See supra, p. 342. Five and a half centuries
were thus allowed to pass before the criminal law was bold enough, in defiance of
a fundamental principle of Magna Carta, to deprive accused men of their "law."
^Mr. Harcourt {Steward, 220 ff.) has vehemently, and Prof. Adams {Origin,
2.(i(> ff.) judicially and moderately, maintained this view. Mr. Adams is influenced
by his failure to discover any instance of '■^ per legem terrae " in the technical sense,
but ^^ per legem Angliae " occurs in Set. Civil Pleas (Selden Society), No. 104,
where the reference is to ordeal of water.
2 It would seem, however, from the words of these statutes that for this purpose
the provisions of chapters 36 and 38 were used to supplement those of the present
chapter, if they were not confused with them. See 5 Edward III. c. 9 ; 25
I
CHAPTER THIRTY-NINE 381
The Act of 1352, for example, after reciting this provision
of Magna Carta, insisted on the " indictment or presentment
of good and lawful people of the same neighbourhood where
such deeds be done." Coke,i founding apparently on these
fourteenth-century statutes, makes " per legem terrae *"
equivalent to " by due process of law " and that again to
"by indictment or presentment of good and lawful men,"
thus finding the grand jury enshrined in Magna Carta.
The framers of the Petition of Right ^ read the same words
as a prohibition, not only of imprisonment " without any
cause showed " but also of proceedings under martial law,
thus interpreting the aims of King John's opponents in the
light of the misdeeds of King Charles.
Anachronisms such as these must be avoided. What-
ever may have been the exact grievances that bulked most
largely in the barons' minds in 12 15, their main contention
was obvious. John was no longer to take the law into his
own hands : the deliberate judgment of a competent court
of law must precede any punitive measures to be taken by
the King against freemen of his realm.
(4) The meaning of " vel." The peculiar use of the word
" vel " introduced an unfortunate element of ambiguity.
No proceedings were to take place " without lawful judg-
ment of peers or by the law of the land " — " or " thus
occurring where " and " might naturally be expected.
Authorities on medieval Latin are agreed, however, that
" vel " is sometimes equivalent to et.^ Comparison with the
Edward III. stat. 5, c. 4 ; 37 Edward III. c. 18; 38 Edward III. c. 3 ; 42
Edward III. c. 3 ; 17 Richard II. c. 6. See also Stubbs, Const. Hist., II. 637-9,
for the series of petitions beginning with 1351.
^ Second Institttte, p. 46, 2 ^ Charles I. c. I.
^ Pollock and Maitland, I. I52n., read the word as having both meanings in this
passage. Cf. Gneist, Engl. Const., chapter xviii. Mr. Pike, Hotise of Lords,
170, takes a different view : "King John bound himself in such a manner as to
show that judgment of peers was one thing, the law of the land another. The
judgment of peers was ... a very simple matter and well understood at the time.
The law of the land included all legal proceedings, civil or criminal, other than
the judgment of peers." The present writer rejects this antithesis, because the
two things may be, and indeed must be, combined. The "trial" by a law and
the "judgment" by equals were complementary of each other. The peers
382 MAGNA CARTA
terms of chapter 52 and with those of the corresponding
Article of the Barons places the matter almost beyond
doubt. The 25th of the Articles of the Barons had provided
that all men disseised by Henry or Richard should " have
right without delay by judgment of their peers in the king's
court," giving no hint of any possible alternative to.
judicium parium. Chapter 52 of the Charter, in supple-
menting the present chapter, describes the evils complained
of in both chapters as acts of disseisin or outlawry by the
King " sine legale judicio parium suorum," leaving no room
for ambiguity.
II. The Scope of the Protection afforded. The object of
the barons was to protect themselves and their friends
against the King, not to set forth a scientific system of
jjlisprudence : the judicium parium was interposed as a
barrier against measures instituted by the King, not
against appeals of private individuals. Pleas following
upon accusations by the injured party were held in 147 1
not to fall within the words of Magna Carta. ^ This was a
serious limitation ; but as against the Crown the scope of
the protection afforded by the Great Charter was very wide
indeed. Care was taken that the three-fold safeguard
should cover every form of abuse likely to be practised by
John.2
(i) Capiatur velimprisonetur. These words are followed
in the text by a string of other verbs, each of which is
introduced by " aut" (" aut disseisiatur/' etc.). The con-
trast between " vel " and " aut " strengthens the suggestion
that " vel " is used in this chapter conjunctively. The mean-
ing would then be that no one could be arrested and
imprisoned (that is, no one could be detained as a prisoner)
without trial. If " vel,*' on the other hand, were to be read
disjunctively while the two words it connects were literally
appointed the test and decided whether it had been properly fulfilled. See also,
on opposite sides, Harcourt, Steward^ 219 ff., and Adams, OHgin^ 262.
^See, e.f;. Pike, House of Lords, 217, citing Littleton in Year Booh, Easter,
10 Edward IV., No. 17, fo. 6.
2 This chapter applied only to abuses of criminal process : cf. c. 21 for amerce-
ments and civil process.
CHAPTER THIRTY-NINE 383
interpreted and enforced, orderly government would be at
an end.^ Arrest normally precedes judgment, although
judgment must precede permanent imprisonment following
on arrest.
(2) Aut disseisiatur. Avarice was a frequent motive of
John's oppressions : the machinery of justice was an engine
for transferring land and money to his treasury. Crown-
tenants frequently found their estates appropriated by the
Crown as escheats. That this was a grievance to which the
barons attached supreme importance is shown in many
ways : by the care taken in the 25th Article of the Barons
and in chapter 52 of the Charter to provide procedure for
restoring " disseised " ^ estates, and by the terms of writs
issued by John after the treaty at Runnymede, for the
immediate restoration of " lands, castles, and franchises
from which we have caused any one to be disseised injuste
et sine judicio," ^
Later versions of Magna Carta (beginning with that of
12 1 7) are careful to define the objects to be protected from
disseisin : " free tenements, franchises, and free customs." ^
(a) Liherum tenementum. " Free " tenements w'ere free-
holds as opposed to the villenagiutn that passed into the
modern copyhold. None of the possessions thus protected
were more highly valued by the barons than their feudal
strongholds.^ Castles claimed by great lords as their own
property are mentioned in many writs of the period, while
chapter 52 of Magna Carta gives them a prominent place
1 The wording of the 29th Article of the Barons, if not merely due to careless
draftsmanship, seems, however, against this conjunctive interpretation. Cf. Adams,
Origin, 262.
2 For this word cf. supra, c. 18.
' See I\of. Clans., I. 215. Mr. Pike {House of Lords, p. 170) maintains, indeed,
that the prevention of disseisins ^* sine judicio'''' was the chief, if not the sole,
object of the chapter under discussion: "The judgment of peers had reference
chiefly to the right of landholders to their lands, or to some matters connected
with feudal tenure and its incidents." This goes too far : the barons by no means
confined the safeguard afiforded by the judicium parium to questions of land.
Pollock and Maitland, I. 393, countenance a broader interpretation.
* De libero teneinento suo vel libertatibus vel liberis consuetudinibus suis.
^Cf. supra, p. 151.
384 MAGNA CARTA
among the " disseisins " to be restored, (b) " Lihertates "
covered feudal jurisdictions, immunities, and privileges of
various sorts, of too intangible a nature to be appropriately
described as " holdings." (c) Consuetudines had two
meanings, a broad general one and a narrower financial
one.^ As the Charter of 12 17 uses a proprietary pronoun
(no freeman shall be disseised of his free customs), it pro-
bably refers to such rights as those of levying tolls and
tallages. These vested interests were of the nature of
monopolies; and Coke, in treating this passage as a text
on which to preach the doctrine that monopolies have always
been illegal in England, aims wide of his mark. Com-
menting on the words "de libertatihus," he declares that
generally all monopolies are against this Great Charter,
because they are against the liberty and freedom of the
subject and against the law of the land." ^ In this error he
has been assiduously followed.^
(3) Aut utlagetur, aid exuletur, aut aliquo rnodo
destruatur. The declaration of outlawry, which could only
be made in the county court, was a necessary preliminary
to the forfeiture of the outlaw's lands and goods. Tine
expedient recommended itself peculiarly to John's genius;
it was his policy to terrify those with whom he had quar-
relled, until they fled the country ; to summon them three
times before the county court, knowing that they dared not
face his corrupt and servile officers; and finally to have
them formally outlawed and their property seized. Such
had been the fate of Robert Fitz Walter and Eustace de
Vesci, in the autumn of 1212."* The outlawed man was
outside the pale of society; anyone might slay him at
pleasure; in the grim phrase of the day, he bore " a wolf's
head " (caput lupinum), and might be hun^Blike a noxious
beast. A reward of two marks was offerea for each out-
law's head brought to Westminster. This sum was paid
^ Cf. suj>ra, p. 246. 2 Second Instihite, p. 47.
'See, e.g. Creasy, Hist, of Const. ^ p. 151 n. : ** Monopolies in general are
against the enactments of the Great Charter." See also Taswell-Langmead, Eng.
Const. Hist., 108.
* See supra, p. 25.
CHAPTER THIRTY-NINE 385
in 1 196 for the head of WilHam of Elleford.^ The word
" exiled " explains itself ; and commentators have very pro-
perly noted the care taken to widen the scope of the clause by
the use of the words " or in any other way destroyed." ^
(4) " Nee super eum ibimus, nee super eum mittemus,"
These words have been frequently misinterpreted. Read in
the light of historical incidents of the immediately preceding
years, they leave no room for ambiguity. Their object was
to prevent John from substituting violence for legal pro-
cess : he must never again attack per vim et arma men
unjudged and uncondemned.
The meaning is plain. Yet Coke, following his vicious
method of assuming the existence, in Magna Carta, of a
warrant for every legal principle of his own day, misled
generations of commentators. He maintained that John
promised to refrain from raising, in his own courts, actions
in which he was personally interested. In elaborating this
error, he drew a distinction between the court of King's
Bench, otherwise known as coram rege, because the King
was in theory present, and other courts to which he had
" sent " a writ delegating authority. Ibimus, he seems to
think, applied in the former case; mittemus in the latter.
To quote his words, " No man shall be condemned at the
King's suit, either before the King in his bench, where the
pleas are coram rege (and so are the words, nee super eum
ibimus, to be understood) nor before any other commis-
sioner, or judge whatsoever (and so are the words, nee super
eum mittemus, to be understood), but by the judgment of
his peers, that is, equals, or according to the law of the
land."^ Coke is in error; it was the use of brute force.
1 See Pi'/'e Rolls, ^Mris^rd I., cited by Madox, I. 20i.
^ E.g. Coke, Sec. Jv> » 48- For the early history of outlawry and exile, see
Liebermann, Friedlosigkeit (Brunner-Festschrift), and Gesetze, II. 413 ; A. Reville,
Abjuratio regni^ Revue Hist.y vol. 50 (1892). Harcourt {Steward, 221) charac-
terises *^ destruatur^^ as a "colloquial expression" covering even amercements,
if of excessive amounts.
^ See Second Institute, p. 46. John Reeves, History of English Law, I. 249
(third ed.), while condemning Coke, gives an even more strained interpretation of
his own. Lingard, History of England, III. c. I, deserves praise as the first
commentator who took the correct view.
2B
386 MAGNA CARTA
not merely one particular form of legal process, which John
in these words renounced.
III. What Classes enjoyed the Protection of Judicium
Parium ? No " f reernan " was to be molested in any of the
ways specified; but how far in the social scale did this
description descend? Coke claims villeins as free for
purposes of this chapter and of chapter i, while rejecting:
them for the purposes of chapter 20.^ Their right to the
status of freeman has already been disallowed, and any
possible ambiguity as to the present chapter is removed by
the words of the revised version of 12 17. Chapter 35 of
that reissue, with the object of making its meaning clearer,
inserts after " disseisiatur " the words (already discussed)
" de lihero tenemento suo vel lihertatihus vel liheris con-
suetudinibus siiis," Mr. Prothero suggests that this addi-
tion implies an advance on the privileges secured in 1215 :
— " It is worth while to notice that the words in which
these liberties are stated in § 35 of the Charter of 12 17 are
considerably fuller and clearer than the corresponding
declaration in the Charter of 12 15." ^ It is safer to infer that
no change was here intended, but merely the removal of
ambiguity. If there is a change, it is rather a contraction
than an extension, making it clear that only " free " tene-
ments are protected, and excluding the property of villeins
and even villenagium belonging to freemen.^ It was made
plain beyond reasonable doubt that no villein should have
lot or part in rights hailed by generations of commentators
as the national heritage of all Englishmen.^
IV. Reactionary Side of these Provisions. To insist
that in all cases a judgment of feudal peers, either in King's
Court or in Court Baron, should take the place of a judg-
ment by the King's professional judges, was to reverse one
of the outstanding features of the policy of Henry II. In
^ Second Institutey pp. 4, 27, and 45.
"^ Simon de Montfoi't, 17 n. Cf. Blackstone, Great Charter, xxxvii,, **the more
ample provision against unlawful disseisins."
3 Cf. Pollock and Maitland, I. 340 n.
*Cf. supra, p. 118. Other verbal changes in the charter of 1217 show the same
care to exclude the villeins. E.g. c. 16 leaves the King's demesne villeins strictly
" in his mercy," that is, liable to amercement without any reservation.
CHAPTER THIRTY-NINE 387
this respect, the present chapter may be read in connection
with chapter 34. The barons, indeed, were not strict
logicians, and probably thought it prudent to claim more
than they intended to enforce. Yet, a danger lurked in
these provisions ; the clause was a reactionary one, tending-as-*
to restore feudal privileges and feudal usage, inimical
alike to the Crown and to the growth of popular liberties. ^
J John promised that feudal justice should be dispensed in ^xs^
'his feudal court; and, if this promise had been kept, the
result would have been to check the development of the
small committees destined to become at no distant date
the Courts of King's Bench and Common Pleas, and to
revive the fast-waning jurisdictions of the manorial courts
on the one hand and of the commune concilium on the
other.2
V. Genesis of this Chapter, The interpretation here
given is emphasized by comparison with certain earlier
documents and events. The reigns of Richard and John
furnish abundant examples of the abuses complained of.
In 1 191, Prince John, as leader of the opposition against
his brother's Chancellor, William Longchamp, concluded
a treaty that protected himself and his allies from the very
evils which John subsequently committed against his own
barons. Longchamp conceded in Richard's name that
bishops and abbots, earls, barons, " vavassors " and free-
tenants, should not be disseised of lands and chattels at
the will of the King's justices or ministers, but only by
judgment of the King's court according to the lawful
customs and assizes, or by the King's command.^
^ Mr. G. H. Blakesley, Law Quarterly Review^ V. 125, perhaps goes too far :
" It may reasonably be suspected that cap. 39 also was directed merely to maintain
the lord's court against Crown encroachments."
2 Mr. Pike, House of Lords ^ i70-4» shares this view of the reactionary nature of
the clause, although he considers that the claim \.o judicium paHuin by a Crown
tenant might be satisfied by the presence of one or more barons among the judges
of the *' Benches," and did not necessarily involve a full commune concilium.
Ibid.^ p. 204. If the ''judgment " of the full court was requisite (and, in spite of
the high authority of Mr. Pike, there is much to be said for that contention), then
the reactionary feudal tendency is even more prominent.
3 See R. Hoveden, III. 136.
388 MAGNA CARTA
Now, the main subject of the arbitration, ending in this
treaty, was the custody of certain castles and estates.
After the right to occupy each separate castle in dispute
had been carefully determined, provision was then made,
in the general words cited above, against this arrangement
being disturbed without a judgment of the curia regis.
Disseisin, and particularly disseisin of castles, was thus
in 1 191, as in 1215, a topic of special prominence.
Early in 12 13, the King had attempted to take vengeance
upon his opponents in a manner they are not likely to have
forgotten, two years later at Runnymede. John, resenting
the attitude of the northern barons who had refused alike
to accompany him to Poitou and to pay scutage, deter-
mined to take the law into his own hands. Without
summoning his opponents before a commune co^icilium,
without even a trial and sentence by one of his Benches,
he set out with an army to punish them. He had gone
as far north as Northampton when, on 28th August, 12 13,
Stephen Langton persuaded him to defer forcible proceed-
ings until he had obtained a legal sentence in a formal
Curia,^ That John again threatened recourse to violent
methods may be inferred from the letter patent issued in
May, 12 15, when both sides were armed for war. He
proposed arbitration, and promised a truce until the arbitra-
tors had given their award. The words of this promise
are notable ; since, not only do they illustrate the procedure
of August, 1 2 13, but they agree closely with the clause of
Magna Carta under discussion. The words are : — " Know
that we have conceded to our harons who are against us,
that we shall not take or disseise them or their men, nor
shall we go against them per vim y el per arma, unless by
the law of our kingdom, or by the judgment of their peers
in curia nostra."^ Magna Carta repeats this concession
in more general terms, substituting " freemen " for the
" barons " of the writ — an alteration which necessitated
the omission from the Charter of the concluding words of
the writ, " in curia nostra " ; because the peers of ordinary
^ Cf. supra^ p. 29.
2 The writ is dated loth May, 12 15, and appears in New Rymevt I. 128.
CHAPTER THIRTY-NINE 389
freemen would be found among the freeholders in the Court
Baron. ^
VI. Later History of ''Judgment of Peers." The claim
made by the barons at Runnymede was re-asserted on
subsequent occasions. The phrase " judicium parium "
which, probably in consequence of its use in Magna Carta,
sprang into " sudden and extraordinary prominence " ^ was
destined to have a long and distinguished career. Mr.
Harcourt ^ thinks that " it was the obscurity of the chapter
when reissued, the fact that it might mean so many things,
which supplied the congenial soil wherein the principle
of trial of peers was able to expand and grow to maturity,"
when " the Charter as a whole became the Bible of the
constitution."
(i) The baronial contention. The earls and barons,
throughout the reign of John's unhappy son, attempted
to place a broad interpretation on the privilege secured to
them by this chapter — claiming that all pleas, civil and
criminal (such at least as were raised against them at the
instance of the Crown) should be tried by their fellow earls
and barons, and not by professional judges of lower rank.
William de Braose in 1208 had declared himself ready to
satisfy John " secundum judicium curiae suae et baronum
parium meorum," ^
(2) The royal contention. The Crown, on the other
hand, while not openly infringing the Charter, tried to
narrow its scope. Judges appointed to determine pleas
coram rege, no matter what their original status might be,
became (so the Crown argued) by such appointment, the
peers of any baron or earl. This doctrine was enunciated
in 1233 when Peter des Roches denounced Richard, Earl
Marshal, as a traitor, in a meeting (colloquium) of crown-
tenants held at Gloucester on 14th August of that year.
Thereafter, " absque judicio curiae suae et parium suorumy"
as Matthew Paris carefully relates,^ Henry treated Earl
Richard and his friends as outlaws, and bestowed their
1 Magna Carta also omits "/^r vim et anna.''''
^Cf. Harcourt, ibid., 235. ^ Ibid., 236.
* M. Paris, II. 524. 5/^^^,^111.247-8.
390 MAGNA CARTA
lands on his own Poitevin favourites. An attempt was
made, at a subsequent meeting held on 9th October, to
have these proceedings reversed on the ground, already
stated, that they had taken place absque judicio parium
suorum.
The sequel makes clear a point left vague in Matthew's
narrative : there had been a judgment previous to the
seizure, but only a judgment of Crown officials coram rege,
not of earls and barons in commune concilium. The
justiciar defended the action of the government by a
striking argument : " there were no peers in England, such
as were in the kingdom of France," and, therefore, John
might employ his justices to condemn all ranks of traitors.^
Bishop Peter was here seeking to evade the provisions of
Magna Carta without openly defying them, and his line
of argument was that the King's professional judges, how-
ever lowly born, were the peers of an English earl or baron .^
Neither the royal view nor the baronial view entirely pre-
vailed. A distinction, however, must be drawn between
criminal and civil pleas.
(3) Criminal pleas. Offenders of the rank of barons
partially made good their claim to a trial by equals ; while
ordinary freemen failed. A further distinction is thus
necessary, (a) Crown tenants. The conflicting views held
by King and baronage here resulted in a compromise. In
criminal pleas, the Crown was obliged to recede from the
high ground taken by Peter des Roches in 1233. Un-
willingly, and with an attempt to disguise the fact of
surrender by confusing the issue, Bracton in theory and
Henry III. in practice admitted part of the barons' demand,
namely, " that in cases of alleged treason and felony, when
^M. Paris, C/iron. Mag.^ III. 251-2.
-Pollock and Maitland, I. 393, hesitate to condemn this argument. ** The very
title of the 'barons ' of the Exchequer forbids us to treat this as mere insolence."
Dr. Stubbs has no such scruples : " The Bishop replied contemptuously, and with
a perverse misrepresentation of the English law" {Const. Hist.^ II. 49). Else-
where he makes him, not so much contemptuous, as ill-informed of the law —
''ignorant blunder as it was" (II. 191). Yet Bishop Peter had presumably an
intimate knowledge of the law he administered as justiciar in 1233. In the matter
of amercements, at least, barons of exchequer acted as peers of earls and barons.
CHAPTER THIRTY-NINE 391
forfeiture or escheat was involved, they should be judged
only by earls and barons.^ Bracton does not admit that
the King's justices were not " peers " of barons ; but
deduces their disability from the narrower consideration
that the King, through his officials, ought not to be judge
in his own behalf, since his interests in escheats might bias
his judgment. This explains why " privilege of peers "
has never extended to misdemeanours, since these involved
no forfeiture to the Crown.
The judicium pariuvi was secured to earls and barons
in later reigns by bringing the case before the entire body
of earls and barons in commune concilium. What the
barons got at first was " judgment " by peers. The actual
" trial " was the " battle," the fellow-peers acting as umpires
and enforcing fair play.^ Although new modes of proce-
dure came to prevail, the Court of Peers continued its
control, and the judgment of peers gradually passed into
the modern trial by peers.^ The subject has been further
complicated by the growth of the modern conception of a
"peerage," embracing various grades of "nobles." In
essentials, however, the rights of a baron accused of crime
have remained unchanged from the days of Henry III. to
our own. The privilege of " trial by peers " still extends
to treason and felony, and is still excluded from mis-
demeanours. When competent, it still takes place before
a " Court of Peers " — namely, the House of Lords, if Parlia-
ment is in session, and the Court of the Lord High Steward,
if not. Under these limitations the privilege of a peer has
been for centuries a reality in England for earls and barons,
and also for members of those other ranks of the modern
"peerage" unknown in 1215 — dukes, marquesses, and
viscounts.^
1 Pike, House of Lords, 173. See also Bracton, f. 119 ; Pollock and Maitland,
I. 393.
2 "The trial, therefore— the ascertaining of the fact — was, though under the
direction and control of the Court of Peers, by battle ; but the judgment on the
trir.l by battle was to be given by the peers." Pike, Ilotise of Lords, 174.
^Pike, ibid., 1 74-9.
* The privilege was extended to peeresses by 20 Henry VI. c. 9.
392 MAGNA CARTA
(b) For tenants of a mesne lord no similar privilege has
been established, even in a restricted form. In charges of
felony, as in those of misdemeanour, all freemen outside
the peerage are tried, and have been tried for many
centuries past, in the ordinary courts of law. There is
no privileged treatment for knight or landed gentleman :
private feudal courts never recovered from the wounds
inflicted by Henry II. The clauses of Magna Carta which
sought to revive them were rendered nugatory by legal
fictions or simply by neglect.
(4) Civil pleas. Various attempts were made by the
barons to make good a claim to judicium parium in civil
cases. ^ The chief anxiety, perhaps, of the men of 12 15
was to save their estates and castles from disseisin conse-
quent on such pleas. Yet the barons' efforts in this direc-
tion were unsuccessful. The House of Lords (except in
cases involving the dignity or status of a peer) has never
claimed to act as a court of first instance in civil cases to
which a peer was a party. Noble and commoner here are
on a level. No " peer of the realm " has, for many centuries,
asked to plead before a special court of peers in any ordinary
non-criminal litigation, whether affecting real or personal
estate.
VII. Erroneous Interpretations, The tendency to vague-
ness and exaggeration has already been discussed. Two
mistakes of unusual persistence require detailed notice.
(i) The identification of judicium parium with trial by
jury. The words of the present chapter form the main, if
not the sole, ground on which this traditional error has
been based.^ The mistake probably owes its origin to a
tendency of later generations to explain what was unfamiliar
in the Great Charter by what was familiar in their own
experience. They found nothing in their own day to
correspond with the judicium parium of 1215 ; and nothing
in Magna Carta (unless it were this clause) to correspond
^The Earl of Chester claimed it in 1236-7, and the Earl of Gloucester (as a lord
marcher) in 1281. See Pollock and Maitland, I. 393 n. See, however, Harcourt,
Steward, 291.
2Cf. supra, pp. 134-5.
CHAPTER THIRTY-NINE 393
with trial by jury : therefore they identified the two.^ Mr.
Reeves, Dr. Gneist, and other writers long ago exposed
this error, but the most conclusive refutations are those
given by Prof. Maitland and Mr. Pike. The arguments
of these writers are of a somewhat technical nature ; ^ but
their importance is far-reaching. They seem to be mainly
three : —
(a) The criminal petty jury cannot be intended in this
chapter, since it had not been invented in 1215:^ to intro-
duce trial by jury into John's Great Charter is an unpardon-
able anachronism, (h) The barons would have repudiated
trial by jury if they had known it. They desired (here as
in chapter 21) that questions affecting them should be
" judged " before fellow barons, and in the normal case, by
the duellum. They would have scorned to submit to the
verdict of " twelve good men " of their own locality. Their
inferiors must have no voice in determining their guilt or
innocence. This sentiment was shared by the tenants of
mesne lords, (c) Judgment and verdict were essentially
different. The function of a petty jury (after it had been
invented) was to answer a specific question. The insurgent
barons demanded more than this : they asked a decision
on the whole case.^ The " peers " who judged presided
over the proceedings from beginning to end, appointing
the proof they deemed appropriate, sitting as umpires while
its fulfilment was essayed, and giving a final decision as to
success or failure therein.
^ The erroneous identification of judgment of peers with trial by jury can be
found far back in legal history. Pollock and Maitland, II. 622-3 i^m trace it to
within a century of Magna Carta. "This mistake is being made already in
Edward I.'s day; Y. B. 30-1 Edward I., p. 531." In spite of modern research
the error dies hard. It appears, e.^^., in Thomson, Magna Charta, 223 ; Taswell-
Langmead, Const. Hist., no; Goldwin Smith, " The United Kingdom,'''' I. 127.
* Pollock and Maitland, I. 152 n., and Pike, Hotise of Lords, 169.
^Cf. supra^ p. 134.
*Cf. Pike, ibid., 169. "From the time when trial by jury first commenced,
either in civil or in criminal cases, to this present end of the nineteenth century,
no jury ever did or could give judgment on any matter whatsoever." The differ-
ence between the ancient and modern conceptions of judgment, however, must
not be lost sight of.
394 MAGNA CARTA
(2) Magna Carta and arbitrary commitment. A second
erroneous theory has still to be discussed. The Petition of
Right, as already stated, treats Magna Carta as prohibiting
the Crown from making arrests without a warrant showing
the cause of detention ; and the earlier commentators further
interpreted it as making all acts of arbitrary imprisonment
by the Crown absolutely illegal. Hallam, for example,
declares that " It cannot be too frequently repeated that no
power of arbitrary detention has ever been known to our
constitution since the charter obtained at Runnymede." ^
Yet every King of England from John Lackland to Charles
Stewart claimed and exercised the prerogative of sum-
marily committing to gaol any man suspected of evil
designs against Crown or Commonwealth. Even the
famous protest of the judges of Queen Elizabeth, asserting
the existence of legal limits to the royal prerogative of
commitment, proves the lawfulness of the general practice
to which it makes exceptions. Such rights inherent in the
Crown were never seriously challenged until the struggle
between Charles I. and his parliaments had fairly begun.
Then only was it suggested that Magna Carta was intended
to prohibit arbitrary commitments at the command of the
Crown. Such was the argument deliberately put forth in
1627 during the proceedings known sometimes as Darnell's
case and someti«ies -as -the case of the Five Knights.
Heath, the Attorney-General, easily repelled this conten-
tion : " the law hath ever allowed this latitude to the King,
or his privy council, which are his representative body, in
€xtraordinary cases to restrain the persons of such freemen
as for reasons of state they find necessary for a time,
without for this present expressing the causes thereof." 2
The parliamentary leaders, however, too grimly in earnest
to be deterred by logic, were far from abandoning their
error because Heath had exposed it. They embodied it,
on the contrary, in the Petition of Right, which condemned
the Crown's practice of imprisoning political offenders
"" without any cause showed " (other than per speciale man-
1 Cons^. Hist.^ I. 234.
=^See State Trials, III. p i, and S. R. Gardiner, History, VI. 214.
CHAPTER THIRTY-NINE 395
datum regis), as contrary to the tenor of Magna Carta —
an effective contention as a political expedient, but unsound
in law.
CHAPTER FORTY.
NuLLi vendemus, nulli negabimus, aut differemus, rectum
aut justiciam.
To no one will we sell, to no one will we refuse or delay, right
or justice.
This chapter, like the preceding, has had much read into
it that would have astonished its framers : application of
modern standards to ancient practice has resulted in com-
plete misapprehension. The sums customarily received by
John, as by his predecessors, at every stage of legal proce-
dure, were not always the wages of deliberate injustice.
Many such payments were not bribes to an unjust judge,
but merely expedients for hastening the law's delays, or* to
ensure a fair hearing for a good plea, or to obtain some
unusual but not unfair expedient, such as a peculiarly potent
writ or the hearing of a case in the exchequer, which would
ordinarily have been tried elsewhere. If the royal courts
charged higher rates than the feudal courts, they supplied
a better article. When Henry of Anjou threw open the
doors of his court to all freemen who chose to pay for writs,
he found a ready market. These writs differed widely in
price. Some from an early date were issued whenever
applied for (writs de cursu) and at a fixed sum : others
were granted only as marks of favour or after a bargain
had been struck. Specially quick or cogent procedure had
to be specially paid for.
It would thus appear that the system of John was not
open to the unqualified and violent condemnation which it
usually receives. Hallam's language is too sweeping when
he says : " A law which enacts that justice shall neither
be sold, denied, nor delayed, stamps with infamy that
government under which it had become necessary." ^ In
'^Middle Ages, II. 451.
396 MAGNA CARTA
the twentieth century, as in the thirteenth, justice cannot
be had for nothing ; and tlie would-be litigant with a good
claim but a slender purse will be well advised to acquiesce
in a small loss rather than incur certainty of losing as much
again in extra-judicial outlays, and risk of losing many
times more in the judicial expenses of a protracted litiga-
tion. The lack of " free justice " is a reproach which the
men of to-day cannot with good grace fling at the adminis-
tration of John.
As the evils complained of are often exaggerated, so also
are the reforms promised by this chapter of Magna Carta.
John is usually held to have agreed to the abolition of
payments of every sort for judicial writs and other fees
of court. Justice, unlike other valuable commodities, was,
it would appear, to be obtained for nothing — an ideal never
yet attained in any civilized community.
Those who framed this chapter desired to secure a more
reasonable measure of reform : abuses of the system were
to be redressed.^ Unfortunately, it was not easy to define
abuses — to determine where legitimate payments stopped
and illegitimate ones began. Prohibitive prices ought not
to be charged for writs de cursu; but was the Crown to
have no right to issue writs of grace on its own terms?
Plaintiffs who had any special reason for haste frequently
paid to have their suits heard quickly : was that an abuse ? ^
Whatever the intention may have been, the practical
effect of the clause was not to secure the abolition of the
^ Cf. Madox, I. 455 : '* By Jtullz vendemus were excluded the excessively high
fines : by nulli negabimus, the stopping of suits or proceedings, and the denial of
writs : by nulli differe^nus^ such delays as were before wont to be occasioned by
the counterfines of defendants (who sometimes would outbid the plaintiffs) or by
the prince's will."
2 Fines for this purpose were frequent under Henry II. and his sons. Madox,
I. 447, cites many examples. Thus in ii66 Ralph Fitz Simon paid two marks
"for speeding his right." The practice continued under Henry III. in spite of
Magna Carta. Bracton's Note-book cites a hard case (No. 743) : Henry III. was
claiming prerogative wardship where it was illegal under c. 37 of Magna Carta
{q.v.). The court might have delayed hearing the mesne lord's plea until the
wardship was ended ; but he paid five mzxV^ pi-o festinando judicio suo. The fine
was said to be given ** willingly " {sponte). Did the use of this word make possible
an evasion of c. 40 of the Charter ?
CHAPTER FORTY 397
sale of writs. The practice under Henry III. has been
described by our highest authority :— " Apparently there
were some writs which could be had for nothing ; for others
a mark or a half-mark would be charged, while, at least
during Henry's early years, there were others which were
only to be had at high prices. We may find creditors
promising the King a quarter or a third of the debts that
they hope to recover. Some distinction seems to have been
taken between necessaries and luxuries. A royal writ was
a necessary for one who was claiming freehold; it was a
luxury for the creditor exacting a debt, for the local courts
were open to him and he could proceed there without writ.
Elaborate glosses overlaid the King's promise that he
would sell justice to none, for a line between the price of
justice and those mere court fees, which are demanded even
in our own day, is not easily drawn. That the poor should
have their writs for nothing, was an accepted maxim." ^
Probably the practice before and after 12 15 showed 'few
material differences. Some of the more glaring abuses
were checked : that was all.^ Parliament in subsequent
reigns had frequently to petition against the sale of justice
in alleged breach of Magna Carta.^ The King usually
returned a politic answer, but never surrendered his right
to exact large sums for writs of grace. Richard H., for
example, replied : " Our lord the King does not intend to
divest himself of so great an advantage, which has been
continually in use in Chancery as well before as after the
making of the said charter, in the time of all his noble
progenitors who have been kings of England."^
It is evident that Magna Carta did not put down the
practice of charging heavy fees for writs. Yet this chapter,
* Pollock and Maitland, I. 174. Cf. ibid.^ II. 204, and authorities cited.
^Madox, I. 455, says: ** And this clause in the great Charters seems to have
had its effect. For ... the fines which were paid for writs and process of law
were more moderate after the making of those great Charters than they used to be
before."
3 Instances are collected by Sir T. D. Hardy in Rot. de oblatis, p. xxi. See also
Stubbs, Const. Hist., II. 636-7.
^ Rot. Parl.i III. 116, cited Stubbs, Const. Hist.^ II. 637.
398 MAGNA CARTA
although so frequently misunderstood and exaggerated, is
still of considerable importance. It marks, for one thing,
a stage in the process by which the King's courts out-
distanced all rivals. In certain provinces, at least, royal
justice was left in undisputed possession. In these the
grievance was not that there was too much royal justice,
but that it was sometimes delayed or denied. Here, then,
even in the moment of John's bitter humiliation we find
evidence of the triumph of the policy inaugurated by his
father.
It is not to such considerations, however, that this chapter
owes the prominence usually given to it in legal treatises;
but rather to the fact that it has been interpreted as a
universal guarantee of impartial justice to high and low ; and
because, when so interpreted, it has become in the hands
of patriots in many ages a powerful weapon in the cause
of constitutional freedom. Viewing it in this light. Coke
throws aside his crabbed learning and concludes with what
is rather a rhapsody than a lawyer's commentary : " as the
gold-finer will not out of the dust, threads, or shreds of
gold, let pass the least crumb, in respect of the excellency
of the metal; so ought not the learned reader to pass any
syllable of this law, in respect of the excellency of the
matter." ^
CHAPTER FORTY-ONE.
Omnes mercatores habeant salvum et securum exire de
Anglia, et venire in Angliam, et morari et ire per Angliam,
tam per terram quam per aquam, ad emendum et ven-
dendum, sine omnibus malis toltis, per antiquas et rectas
consuetudines, preterquam in tempore gwerre, et si sint
de terra contra nos gwerrina; et si tales inveniantur in
terra nostra in principio gwerre, attachientur sine dampno
corporum et rerum, donee sciatur a nobis vel capitali
justiciario nostro quomodo mercatores terre nostre tracten-
^ Second Institute J $6.
CHAPTER FORTY-ONE 399
tur, qui tunc invenientur in terra contra nos gwerrina; et
si nostri salvi sint ibi, alii salvi sint in terra nostra.
All merchants shall have safe and secure exit from England,
and entry to England, with the right to tarry there and to move
about as well by land as by water, for buying and selling by the
ancient and right customs, quit from all evil tolls, except (in
time of war) such merchants as are of the land at war with us.
And if such are found in our land at the beginning of the war,
they shall be detained, without injury to their bodies or goods,
until information be received by us, or by our chief justiciar,
how the merchants of our land found in the land at war with us
are treated ; and if our men are safe there, the others shall be
safe in our land.
Merchants and merchandise had suffered from John's
greed. The control of commerce was reserved for the
King's personal supervision : no binding rule of law or
traditional usage trammelled him in his dealings with
foreign merchants, who were dependent on royal favour,
not on the law of the land, for the privilege of trading and
even for personal safety. No alien could enter England
or leave it, nor take up his abode in any town, nor move
from place to place, nor buy and sell, without paying heavy
tolls to the King. This royal prerogative proved a profit-
able one.^
John increased the frequency and amount of such exac-
tions, to the detriment alike of foreign traders and their
customers. Magna Carta, therefore, sought to restrain this
branch of prerogative, forbidding him to exact excessive
tolls for removing obstacles of his own creating. This
benefited merchants by securing to them certain privileges,
which may perhaps be analysed into three : safe-conduct,
^ So far all authorities are agreed, though a difference of opinion exists as to the
source of these prerogatives. Thus (a) Stephen Dowell, History of Taxation and
Taxes in England, I. 75, considers that the duties on imports and exports were in
their origin of the nature of voluntary dues paid by foreign merchants in return for
freedom of trade and royal protection ; {b) Hubert Hall, Customs Revenue oj
England, I. 58-62, justly reckons this prerogative as merely one aspect of purvey-
ance, that is, of the King's right to take what he needed for himself and household.
Under an autocrat, however, facts count for more than theories. The prerogative
was measured by brute force : Kings took what they could with no jealous regard
for the exact letter of the law, and left future ages to invent theories to justify or
explain their conduct.
400 MAGNA CARTA
that is protection of their persons and goods from violence ;
liberty to buy and sell in time of peace ; and a confirmation
of the ancient stereotyped rates of " customs."
So far, the general purport of the enactment is un-
doubted ; but discussions have arisen on several important
points, such as the nationality of the traders in whose
favour it was conceived ; the exact nature of the " evil tolls "
abolished; the motives for the rules enforced; and the
relations between denizens and foreign traders.
I. Magna Carta favours alien Merchants. The better
opinion would seem to be that this chapter applied to
foreign traders from friendly states. Attempts have been
made, indeed, to argue that denizens were to benefit equally
with strangers : such was the purport of a learned discourse
delivered in the House of Commons by William Hakewill,
Barrister of Lincoln's Inn, in 1610, during the debate on
John Bate's case.^ His main argument was that certain
statutes of Edward HI. ,2 in seeking to confirm and expand
the provisions of Magna Carta, did clearly embrace denizens
as well as aliens. Yet the framers of an Act in the four-
teenth century may well have misunderstood the tenor of
John's Charter, or may have deliberately altered it.
Intrinsic and extrinsic evidences combine to create a
strong presumption that here Magna Carta referred chiefly,
perhaps exclusively, to merchants of foreign lands.^
Denizens trading in England did not require those " safe
conducts " which form the chief concession in this chapter.
Their rights of buying and selling were already protected
in another way; for independent traders were unknown,
all merchants being banded into guilds in the various
towns whose privileges (" omnes libertates et liberas consue-
tudines ") were guaranteed in a previous part of the Great
Charter.^ Alien merchants, however, required protection,
^ See Sta^e Trials, II. 407-475, and especially 455-6.
"^E.g. 2 Edward III. c. 9 and 14 Edward III., stat. i, c. 21.
^ Two-thirds of the chapter is occupied in explaining that merchant strangers of
unfriendly States are not to benefit from it. Mr. Hakewill was aware of this, but
•sought to evade the natural inference by subtleties which are not convincing.
* See supra, under c. 13.
CHAPTER FORTY-ONE 401
since they had, strictly speaking, no status in the eye of
the law, and held their privileges from the King.^ The
policy of Henry II. and his sons was to favour merchant
strangers, but to exact in return the highest dues possible,
restrained only by an enlightened self-interest whiqh
stopped short at the point where trade would languish by
becoming unprofitable. The exchequer and patent rolls
afford illustrations of how individual traders or families
made private bargains with the Crown for trading privi-
leges. In 1 181, Henry obtained two falcons for granting
leave to export corn to Norway. In 1197, a certain Hugo
Oisel owed 400 marks for licence to trade in England and
Richard's other lands, in time of war as well as peace.^
At the commencement of John's reign, traders resident
in England collectively obtained confirmation of their
privileges. That King issued letters patent to the Mayor
of London, to the magistrates of many smaller towns, and
to the sheriffs of the southern counties of England, directing
them, in terms closely resembling those of Magna Carta,
to allow to all merchants, of whatsoever land, safe coming
and going, with their wares.^
These arrangements were temporary. John did not
intend that any general grant should prevent him from
exacting further payments from individuals as occasion
offered. For example, Nicolas the Dane promised a hawk
each time he entered England, that he might come and go
and trade " free of all customs which pertain to the King." ^
Such customary dues, at the usual rates, were not abolished
by the^Charter, but only the arbitrary additional payments
for which there was no warrant.
On this point, then, Magna Carta contained no innova-
tions, and the same is true of its provision for reprisals
against traders from lands where English merchants were
ill-treated. On the outbreak of war, the Charter directs
*For the legal position of aliens, see Pollock and Maitland, I. 441-450.
2 See Pipe Rolls^ 27 Henry II. and 8 Richard I., cited Madox, I. 467-8.
3 See Rot. Chart. ^ 60 (5tli April, 1200).
* See Pipe RoU^ 6 John, cited Madox, I. 469, where other illustrations will be
found. Cf. also TP^?/. Pat.^ 170, 170b, 171, 172b.
2C
402 MAGNA CARTA
that merchants of the enemy's nation should be detained
until the King ascertained how his own subjects were treated
in the enemy's territory. This is declaratory of previous
practice, of which an illustration may be found in the terms
of a writ of August, 12 14, which directed the bailiffs of
Southampton to detain all Flemings and their goods pend-
ing further instructions.^ There were thus precedents for
those rules for foreign traders, which have aroused the
admiration of Montesquieu.^
II. Customs and Tolls. " Consuetudines " is in this
passage used in its narrower, financial sense, relating to
those duties on imports and exports still called " customs "
at the present day, and to various local dues as well..
" Tolls," when not stigmatized as " evil tolls " would seem
to be practically synonymous with these customs. The
Crown had at first taken whatever it thought fit. Practice
soon established rules as to the normal rates considered
fair in various circumstances. When a ship-load of
foreign wine arrived, the normal toll was " one cask
from a cargo of ten up to twenty casks, and two casks
from a cargo of twenty or more."^ From other merchan-
dise a share was claimed of a fifteenth or sometimes a tenth
of the whole. Such tolls, if originally a species of ransom,
had in John's day come to be regarded as a legitimate
branch of royal revenue. Any arbitrary increase, however,
was condemned by public opinion, and ultimately by
Magna Carta as a " mala tolta."
The King was not the only one who exacted tolls. Every
town in England, and many feudal magnates, by prescrip-
tive usage or royal grant, levied payments on goods bought
1 In the same writ John bade them allow to depart freely all vessels of the land
of the Emperor or of the King of Scotland after taking security that they would
sail straight to their own countries, with none but their own crews. See jRo^.
Claus.f I. 211, and cf. series of writs in I. 210.
2 See Be V Esprit des Lois, II. 12 (ed. of 1750, Edinburgh), *' La grande chartre
des Anglois difend de saisir et de confisquer en cas de guerre les marchandises des
n^gociants itrayigers, a vioins que ce ne soit par reprisailles. LI est beau que la
nation Angloise ait fait de cela un des articles de sa liberty l'^
3 S. Dowell, IList. of Taxation, I. 83, citing Madox, I. 525-9 {2nd ed. I. 765-
770), and Liber Albus, I. 247-8.
CHAPTER FORTY-ONE 403
or sold at fairs and markets, or that entered the city gates,
or were unloaded at river wharves, or traversed certain
roads. The ambition of every borough was to increase its
own franchises at the expense of its neighbours. The free
customs of Bristol, for example, meant not only that the
men of that city should have freedom from tolls inflicted by
others, but that they should have the right to inflict tolls
upon those others. A whole network of such customs and
restrictions impeded the free exchange of commodities in
every part of England. Magna Carta had no intention of
sweeping these away, so far as they were " just and
ancient " ; and it is probable that the prohibition against
arbitrary increase of tolls was directed only against the
Crown.
III. The Motives prompting these Provisions. It has
been not unusual to credit the framers of Magna Carta
with a policy of quite a modern flavour; they are made
free-traders and credited with a knowledge of economic
principles far in advance of their contemporaries. This is
a misconception : Englishmen in the thirteenth century
had formulated no far-reaching theories of the rights of
the consumer, or the policy of the open door. The home
traders were not consenting parties to this chapter, and
would have bitterly resented any attempt to place foreigners
on an equal footing with the protected guilds of the English
boroughs. The barons acted on their own initiative and
from purely selfish motives. Rich nobles, lay and ecclesi-
astic, desired that nothing should prevent the foreign
merchants from importing wines and rich apparel that
England could not produce. John, indeed, as a consumer
of continental luxuries, partially shared their views, but
his selfish policy threatened to strangle foreign trade by
increasing the burdens attached to it, until it ceased to be
remunerative. The barons, therefore, in their own interests,
not in those of foreign merchants, still less in those of
native traders, demanded that the customs duties should
remain at their old fixed rates. In adopting this attitude,
they showed their selfish indifference to the equally selfish
claims of English traders, who desired a monopoly for
404 MAGNA CARTA
themselves. Every favour shown to foreign merchants
was an injury done to the guilds of the chartered boroughs.
This chapter thus shows a lack of gratitude on the barons'
part for the great service rendered by their allies, the
citizens of London. John, on the other hand, would have
little reluctance in punishing the men of his capital who,
with the ink scarce dry on their new municipal charter, had
not scrupled to desert his cause. ^ It must have been with
grim pleasure that, on 21st July, 1215, in strict conformity
with the tenor of Magna Carta, he addressed a writ to
King Philip inviting reprisals upon London merchants in
France in certain contingencies.^
In the reissue of 12 16 the privileges conferred on merchant
strangers were confined to such as had not been " publicly
prohibited beforehand." This was a material alteration,
the effect of which was to restore to the King full discre-
tionary authority over foreign trade, since he had only
to issue a general proclamation, and then to accept fines
for granting exemption from its operation.
IV. English Boroughs and Merchant Strangers. The
quarrel between home and alien traders underwent many
vicissitudes during succeeding centuries, the Crown taking
now one side, and now the other, as its pecuniary interests
happened to dictate for the moment. No glimmerings of
the doctrine of free trade can be traced : the merchants of
each town, banded in their guilds, directed their endeavours
towards securing rights of exclusive trading for themselves.
It is true that the men of London were scarcely more jealous
of the citizens of Rouen or Paris than of those of York or
Lincoln ; their ambition was to inflict restrictions upon all
rivals alike.
English traders were not yet merchant shippers and
therefore did not prevent foreigners from undertaking the
1 See supra, 34-35.
* See New Rymer, I. 135 : ** Know that we have ordered the mayor and sheriffs
of London to allow merchants of your land to remove their goods and chattels from
London, without hindrance to doing thence their will ; and that if they do not, you
may, if it please you, grieve and molest the men of that town {J.llius villae) in your
power, without our reckoning it a breach of truce on your part."
CHAPTER FORTY-ONE 405
carrying trade between England and the Continent.
Flanders bought English wool and sent back woven fabrics
to rival which English looms could not aspire. Londoners,
however, resold these goods at a profit and resented any
attempt of aliens to encroach on their retail monopoly by
coming into touch w4th English magnates or other con-
sumers. Foreigners must be kept " at the wharf-head."
The Liber Custumarum, a compilation of the early
thirteenth century, lays down minute rules for the regula-
tion of foreign traders in London. The merchant stranger
had to take up his abode in the house of a citizen. He
was prohibited from purchasing articles in process of
manufacture. He could buy only from those who had
the freedom of the city, and could not re-sell within the
borough walls. He was allowed to sell only to burgesses
of London, except on three specified days of the week.
Such were a few of the rules which the Londoners enforced
on all traders within their gates. The King, however,
intermittently encouraged foreigners. Under the fostering
protection of Henry HL, Lombards and Proven9als settled
in considerable numbers in the capital ; and, with conniv-
ance of the King, infringed these rules. When the
Londoners complained, Henry refused relief. Their loyalty
thus shaken, they sided with the King's opponents in the
Barons' War, and when the royalist cause triumphed at
Evesham, the Capital shared in the punishment meted out
to the Crown's opponents. Prince Edward in 1266 was
nominated protector of foreign merchants. At the acces-
sion of that Prince, London bought itself back into favour,
and an attempt was made to define what tolls might be
taken by the Crown. In 1275, in Edward's first parlia-
ment, a tariff was fixed by " the prelates, magnates, and
communities at the request of the merchants " on most of
what then formed the staple exports of England : half a
mark on every sack of wool, half a mark on every three
hundred wool-fells (that is untanned skins with the fleeces
on), and one mark on every load of leather.
These were subsequently called magna et antiqua
custuma. The settlement of 1275 was by no means final.
4o6 MAGNA CARTA
New disputes arose; and in 1285 Edward I. confiscated
the liberties of London, suppressed what he characterized
as abuses, and favoured the aliens. In 1298 the franchises
of the capital were restored, and very soon the abuses com-
plained of began anew. Edward retorted in 1303 by a
special ordinance known as the Carta Mercatoria in favour
of their foreign rivals, by the terms of w^hich the provisions
of the present chapter of Magna Carta became at last a
reality. This new charter, which was the result of a
bargain struck between the Crown and the alien traders,
conferred various privileges and exemptions in return for
an increase of fifty per cent, of duty, known henceforth as
parva et nova custuma. Edward I. made several attempts
to exact the higher rates from denizens as well as strangers ;
but in this he failed. In 1309 a Petition of Parliament
was presented against the exaction of the " new customs,"
declaring them to be in contravention of Magna Carta.
In 131 1 a temporary community of economic and political
interests resulted in an alliance between the English mer-
chants and the English baronage, whose combined efforts
forced the "Ordinances" upon Edward II., compelling
him for a time to reverse his father's policy of favouring
foreigners at the expense of native merchants. It is un-
necessary to follow the checkered fortunes of these Ordin-
ances, frequently enforced and as frequently abolished,
according as the fortunes of the barons or of Edward II.
were for the moment in the ascendant. During the reign
of Edward III. the deep-rooted quarrel between home and
alien merchants continued; and many changes of policy
were adopted by the Crown. The statute of 1328, which
abolished the "staples beyond the sea and on this side,"
provided " that all merchant strangers and privy may go
and come with their merchandises into England, after the
tenor of the Great Charter." ^ Seven years later, this was
confirmed by an act which placed strangers and denizens
on an exact equality in all branches of trade, both wholesale
and retail, under the express declaration that no privileged
rights of chartered boroughs should be allowed to interfere
^ 2 Edward III. c. 9.
CHAPTER FORTY-ONE 407
with its enforcement. 1 While this statute merely repeated
and applied the general doctrine of the present chapter
of Magna Carta, it directly infringed the provisions of
chapter 13.2 Such sweeping regulations were in advance
of their age and could not be carried out without revolu-
tionizing the medieval scheme of trade and commerce,
which depended on merchant guilds, town charters and
local monopolies. The influence of the English boroughs
and their political allies was strong enough to make the
strict enforcement of such legislation impossible ; and later
statutes, bowing to the inevitable, restored the privileges
of the boroughs, while continuing to enunciate an empty
general doctrine of free trade to foreigners.^ The English
boroughs, to which Parliament in the reign of Richard II.
thus restored their franchises and monopolies, were able
effectually to exclude foreign competition, in certain trades
at least, from within their walls, for four centuries, until
the Statute of 1835 ushered in the modern era of free
trade.4
CHAPTER FORTY-TWO.
LiCEAT unicuique de cetero exire de regno nostro, et redire,
salvo et secure, per terram et per aquam, salva fide nostra,
nisi tempore gwerre per aliquod breve tempus, propter
communem utilitatem regni, exceptis imprisonatis et utla-
gatis secundum legem regni, et gente de terra contra nos
gwerrina, et mercatoribus de quibus fiat sicut predictum
est.
It shall be lawful in future for any one (excepting always
those imprisoned or outlawed in accordance with the law of
the kingdom, and natives of any country at war with us, and
^See 9 Edward III. c. i, and cf. 25 Edward III., stat. 4, c. 7.
^Cf. nipra^ pp. 247-8, where the inconsistency between the two parts of the
Great Charter is pointed out. See also supruy p. 117.
3 See 2 Richard II., stat. i, c. i, and 11 Richard II. c. 7.
* See 5 and 6 William IV. c. 76, s. 14.
408 MAGNA CARTA
merchants, who shall be treated as is above provided) to leave
our kingdom and to return, safe and secure by land and water,
except for a short period in time of war, on grounds of public
policy — reserving always the allegiance due to us.
The terms of this permission for free intercourse between
England and foreign lands are peculiarly wide, the excep-
tions being reasonable and necessary. Prisoners obviously
could not leave our shores, nor outlaws return to them :
the case of merchants from hostile states had already been
provided for in a liberal spirit ; while the temporary restric-
tion of intercourse with the enemy on the outbreak of
hostilities was eminently reasonable.
Although the provision is thus general in its scope, it
was peculiarly welcome to the clergy, as enabling them
without a royal permit to proceed to Rome, there to prose-
cute their appeals or press their claims for preferment.
Thus considered, it contains a virtual repeal of article 4
of the Constitutions of Clarendon of 1166, which forbade
archbishops, bishops, and parsons (personcB) of the king-
dom to leave England without the King's licence. The
grant of freedom of intercourse in 12 15 opened a door
for the Church to encroach on the royal prerogative; and
for that reason it was omitted from the reissue of 12 16,
never to be replaced. A boon was thus withdrawn from
all classes from fear that it might be abused by the ecclesi-:
astics. Henry III. took advantage of the omission in order
to restrain the movements of clergy and laity alike. Those
who left the country without licence had frequently to pay
fines.^
The stringency with which the prerogative was at first
enforced tended afterwards to relax. The King preserved
the right, but only exercised it by means of proclamations
over particular classes or on special occasions, the inference
^ £.£'. Coke {Third Instittitey p. 179) cites from Rot.finitim of 6 Henry III. and
Rot. Claus. of 7 Henry IH. the following case: " Willielmus Marmion clericus
projectus est ad regem Franciae sine liceiitia domini regis ^ et proptereajinem fecit. ^* i
The practice had apparently been much the same prior to Magna Carta. E.g.
Madox (I. 3) cites from Pipe Roll of 29 Henry H. how " Randulfus filins Waltert'
reddit compotum de XX marcis, quia exivit de terra Domini Regis." See also
Makower, Const. Hist, of Eng. Church, 239-240 and notes.
CHAPTER FORTY-TWO 409
being that all not actually prohibited were free to come and
go as they pleased. Thus, in 1352 Edward III. had it pro-
claimed throughout every county of England that no earl,
baron, knight, man of religion, archer, or labourer, should
depart the realm under pain of arrest and imprisonment.^
The fact that Edward found it necessary to issue such an
ordinance, autocratic and abhorrent to modern ideals as its
terms now appear, points to a decrease of royal power, as
compared with that exercised by Henry II., John, or Henry
III. A further curtailment of prerogative may be inferred
from the terms of a Statute of Richard II. ,2 which, in con-
firming the King's power to prohibit free egress from
England, does so, subject to wide exceptions. Under its
provisions the Crown might prohibit the embarkation of all
manner of people, as well clerks as others, under pain of
forfeiture of all their goods, " except only the lords and
other great men of the realm, and true and notable mer-
chants, and the King's soldiers," who were apparently in
1381 free to leave without the King's licence, although earls
and barons had been prohibited in 1352. Even if this
statute confers on magnates, merchants, and soldiers, free-
dom to go abroad without royal licence (which is doubtful),
the powers of veto reserved to the Crown were still, to
modern ideas, excessive. The Act remained in force until
1606, when it was repealed under somewhat peculiar cir-
cumstances. After the union of the Crowns, King James,
anxious to draw the bond closer, persuaded his first English
parliament to abrogate a number of old laws inimical to
Scottish interests. It was in this connection that the Act of
Richard II. was declared (in words, however, not limited
to Scotland) to be "from henceforth utterly repealed."^
Coke stoutly maintains that this repeal left intact the
Crown's ancient prerogative, not founded upon statute but
on the common law, of which power the already-cited
Proclamation of Edward III. had been merely an emana-
tion. He seems almost, therefore, to argue that the King
in the seventeenth century retained authority which
, ^See Coke, ibid., citing the Close Roll of 25 Edward III.
25 Richard II., stat. I, c. 2. ^^ j^mes I. c. I, s. 22.
410 MAGNA CARTA
extended precisely over those classes mentioned in the
ordinance of 1352.
In any view, this prerogative has never been completely
abolished : yet the onus has been shifted. While, under
John or Henry III., the subject required, before embarking,
to obtain a licence from the Crown, under later Kings he
was free to leave until actually prohibited by a royal writ.
Coke ^ speaks of the form originally used for this purpose,
a form so ancient in his day as to be already obsolete, known
as Breve de securitate invenienda quod se non divertet ad
partes cxternas sine licentia regis. This was superseded by
the simpler writ Ne exeat regno which is still in use.^ The
sphere of this writ was restricted and altered : it ceased to
be an engine of royal tyranny and was never issued except
as part of the process of a litigation pending in the Court of
Chancery. Regarded with suspicion by the courts of com-
mon law, it was for centuries the special instrument which
prevented parties to a suit in equity from withdrawing to
foreign lands. Some uncertainty exists as to the proper
province of these writs since the Judicature Acts have
merged the Court of Chancery in the High Court of
Justice.^ The perfect freedom to leave the shores of Eng-
land and return at pleasure, accorded by John's Magna
Carta, but immediately withdrawn as impracticable for that
age, has thus in the course of centuries been fully realized.^
Two phrases, occurring in this chapter, call for comment :
(i) Salva fide nostra. This short-lived clause of Magna
Carta very properly provided that mere absence from Eng-
land should absolve no one from allegiance to his King.
The old doctrine of nationality was stringent : nemo potest
exuere patriam. Everyone born in the land owed allegi-
ance to its King — and this tie continued unbroken until
severed by death. A breach of allegiance, which was
1 Third Institute, p. 178.
'■^ Its origin is obscure. See Beames, Brief view of the writ of Ne Exeat,
fassim.
' See Encyclopaedia of Laws of England, IX. 79.
*On the whole subject of these writs, see Stephen, Commentaries, II, 439-40
(ed. of 1899), and authorities tliere cited.
CHAPTER FORTY-TWO 411
consequent thus on the mere accident of birth, might expose
the offender to the inhuman horrors inflicted upon traitors.
A series of statutes, culminating in the NaturaHzation
Act of 1870, have entirely abrogated this ancient doctrine.
A native of Great Britain is now free to become the subject
of any foreign state; and the mere fact of his doing so,
deliberately and with all necessary formalities, denudes him
of his British nationality, severs the tie of allegiance, and
frees him from the operation of the law of treason. The
words " salva fide nostra " no longer apply.
(2) Propter communem utilitatem regni. The Charter, in
placing restriction on the right of free egress in time of war,
declared that such restriction was to be imposed for the
common good of the kingdom, thereby enunciating what is
regarded as a modern doctrine : John was to take action,
not for his own selfish ends, but only pro bono publico.
CHAPTER FORTY-THREE.
Si quis tenuerit de aliqua eskaeta, sicut de honore Walling-
fordie, Notingeham, Bolonie, Lancastrie vel de aliis
eskaetis, que sunt in manu nostra, et sunt baronie, et
obierit, heres ejus non det aliud relevium, nee faciat nobis
aliud servicium quam faceret baroni si baronia ilia esset in
manu baronis ; et nos eodem modo eam tenebimus quo baro
eam tenuit.
If anyone holding of some escheat (such as the honour of
Wallingford, Nottingham, Boulogne, Lancaster, or of other
escheats which are in our hands and are baronies) shall die,
his heir shall give no other relief, and perform no other service
to us than he would have done to the baron, if that barony had
been in the baron's hand; and we shall hold it in the same
manner in which the baron held it.
This chapter reaffirms a distinction recognized by Henry
II. but ignored by John. Crown-tenants were divided into
two classes, according as their holdings had been originally
granted by the Crown, or by some mesne lord whose barony
had subsequently escheated. The latter class received pre-
412 MAGNA CARTA
ferential treatment from Henry II. for reasons to be
immediately explained. A mesne lord had no right to
appropriate the holdings of sub-tenants of a tenant who had
incurred escheat ; but the Crown did not submit to this just
restriction. The King treated all sub-tenancies as wiped
out by the mere fact that their lord's fief had escheated to
the Crown.
Henry II. mitigated in practice the full severity of this
theory, confirming as of grace, or from motives of policy,
or in return for money, claims which he refused to admit
as matter of right. The tenants of escheated baronies were
accepted as tenants in capite of the Crown. ^ Not only so;
but Henry did not allow them to be prejudicially affected
by the change. The King would only take from them
those services and feudal dues which they had been wont
to render to the lord of the barony previous to its escheat.
This just and lenient policy explains the origin of the
division of royal tenants into two classes; tenants who
held of Henry ut de corona, and tenants who held of him
ut de escaeta, ut de honore, or ut de haronia (phrases used
synonymously).^ In respect of such obligations as were
heavier for ordinary Crown tenants than for tenants of
mesne lords, holders of Crown fiefs ut de escaeta were
placed on the more favoured footing. Two illustrations
may be given. While tenants ut de corona under Henry
had to pay large and arbitrary reliefs, those ut de escaeta
paid no more than iocs, per knight's fee.^ Nor was their
^ Royal clemency in this respect could not be relied on by the sub-tenants of
sma/l escheated fiefs (not reckoned as honours or baronies). This seems to be the
opinion of Madox, Ba^-onia Anglica, 199: " If a fee holden of the Crown in
capite escheated to the King and was not an Honour or Barony, then such fee did
not (that is to say, I think it did not) vest in the Crown in the same plight in which
it was vested in the said tenant in capite.'''' Cf. also ibid., 203.
2 See Madox, Baronia An£-/ica,i6g-i7i ; also Pollock and Maitland, I. 261,
and authorities there cited.
3 See Dialogus^ II. x. F, and ibid., II. xxiv. The same rule applied to sub-
tenants of baronies in wardship (which was analogous to temporary escheat) : when
the see of Lincoln was vacant in 1168, the heirs of sub-tenants paid to Henry only
what they would have paid to the bishop ; one giving ;^30 for six fees, and another
30 marks for four. See Pipe Roll, 14 Henry II., and cf. supra, c. 2. In the
CHAPTER FORTY-THREE 413
obligation of " suit " to be increased : " the tenants of any
honour or manor which had come by escheat to the Crown,
were not suitors of the Curia Regis, but of the court of
the honour or manor which had so escheated." ^
John ignored this distinction, extending to tenants ut de
escaeta the more stringent rules applicable to tenants ut
de corona. Magna Carta reaffirmed the distinction; and,
not content with enunciating a general principle, made two
particular applications of it : neither reliefs nor services
of former tenants of baronies were to be augmented by
reason of the fact that such baronies had escheated to the
Crown. 2 Henry III.'s Charter of 1217 emphasized a third
application of the general rule, declaring that he would not,
by reason of an escheated barony, claim escheat or custody
over the sub-tenants of that barony.^ To understand this
concession, it must be remembered that under Henry HI.
sub-tenants of baronies were still liable to have their titles
reduced through the escheat of their lord; while sub-
tenants of those who were themselves sub-tenants were not
exposed to a similar mischance. Here also, the position of
matter of scutage, also, a distinction was recognized : while tenants ut de corona
might be compelled to serve in person without an option, Crown-tenants ut de
honore (and, a fortiori^ sub-tenants also) might claim exemption on tendering scutage.
See case of Thomas of Inglethorpe in 12 Edward II., cited by Madox, Baronia
Anglica^ 169- 1 71.
^ Report on the Dignity of a Peer, I. 60.
^ The need for this reference to relief is not, at first sight, obvious, since c. 2 of
Magna Carta, by forbidding John to exact from Crown-tenants of either class the
arbitrary sums taken by his father, would seem to have already secured them from
abuse. Probably, however, c. 43 sought to prevent John from treating each tenant
of the escheated barony as holder of a new barony of his own, and therefore liable
to a baron's relief of ;i^ioo instead of the £2^ he ought to pay for his five fees, or
;^50 for his ten fees, or as the case might be. The case of William Pantol (see
Pipe Roll, 9 Henry III., cited Madox, I. 318) seems to illustrate this. He was
debited with ^^loo of relief, but protested that he held nothing of the Crown save
five knights' fees of the land which was of Robert of Belesme. This plea was
upheld, and £'j^ of the amount debited was written off.
^See c. 38 of 1217, and cf. the gloss given by Bracton (II. folio 87b) which
makes the meaning somewhat less obscure. The Charter of 12 17 contained a
saving clause: "unless the holder of the escheated barony held directly of us
elsewhere." Bracton added a second proviso, namely, unless the said sub-tenants
(now Crown-tenants tit de escaeta) had been enfeoffed by the King himself.
414 MAGNA CARTA
Crown fiefs tU de escaeta was to be assimilated to that of
fiefs of mesne lords, and differentiated from that of Crown
fiefs ut de corona. Sub-tenancies of escheated baronies
were not to be wiped out, but to subsist, and the Crown
(or its grantee) would take the escheat, subject to all
liabilities to, and rights of, sub-tenants.
The Crown seems not to have strictly observed this
rule in practice. Article 12 of the Petition of the Barons
in 1258 ^ complained that Henry had granted charters
conferring rights not his to give (aliena jura), but which
he claimed as escheats. An act of the first year of
Edward III. narrated how the Crown had confiscated, from
purchasers, tenements held of the Crown " as of honours,"
thus treating them " as though they had been holden in
chief of the King, as of the Crown." Redress was pro-
mised by the statute i^ but irregularities continued through-
out the earlier Tudor reigns; and the first Parliament of
Edward VI. passed an act to protect purchasers of lands
appertaining to honours escheated to the Crown. ^
CHAPTER FORTY-FOUR.
Homines qui manent extra forestam non veniant de cetero
coram justiciariis nostris de foresta per communes sum-
moniciones, nisi sint in placito, vel plegii alicujus vel
aliquorum, qui attachiati sint pro foresta.
Men who dwell without the forest need not henceforth come
before our justiciars of the forest upon a general summons,
except those who are impleaded, or who have become sureties
for any person or persons attached for forest offences.
These provisions were intended to redress one of many
abuses connected with the oppressive forest laws.
I. The Royal Forests, The word " forest " had acquired
an exact technical meaning, and was applied_ip certain
^ See Sel. Charters^ 384 ; but see Adams, Origin^ 344 n.
2 See I Edward III., stat. 2, c. 13, Statutes of Realm, I. 256.
'See I Edward VI., c. 4, Statutes 0/ Realm, III. 9.
CHAPTER FORTY-FOUR 415
wide districts, scattered irregularly throughout England,
reserved to the Crown for^ purposes of sport. Here the
wild boar "and deer of various species found shelter, in
which they were protected by the severe regulations of the
" Forest Law." It was the prevalence of this code which
marked off the districts known as ro^^aHor^sts from all that
lay extra forestam; and this made an accurate definition
possible. A " forest " was a district where this law pre-
vailed to the exc|usion_Qf_th^common law which ruled
outside. The forests with their inhabitants had been
omitted from the process b^y wETch ttte rest of England
had been assimilated under a uniform lex terrae : this was
the root from which rnany evils grew.
From this definition of a forest as a legal, not a physical,
entity, it follows that the word is far from synonymous
with terms such as " wood " or " covert, " implying merely
natural characteristics. A forest was not necessarily
covered with trees throughout the whole or even the
greater part of its extent. Miles of moorland and heath
and undulating downs might be included, and even fertile
valleys, with ploughed fields and villages nestling among
them. The same forest, indeed, might contain many
woods, some of them on royal demesne and some the
property of private owners. Within the imaginary line
the King's power was supreme, and he used it frankly for
the preservation of beasts of the chase. The men who
happened to dwell there were subject to a law, in the
expressive words of Dr. Stubbs, " cruel to man and beast."
If accused of forest offences, they had no protection from
the common law of England any more than Trom the law
of a foreign land. It was something, however, that even
in these high places of prerogative, customary rules grew
up, obtained authoritative recognition, and hardened into
laws which set some limits to royal caprice. Before John's
time the forest code, as set forth in the Assize of Wood-
stock, had taken its place as a definite system of law% distinct
from common law and canon law alike. ^
*A convenient, short account of the forests, with their special laws, special
officials, and special courts, will be found in W. S. Holdsworth's Hi stoty of English
Lmv, I. 340-352. For fuller information see Dialogtis de Scaccario, I. xii. ; John
4i6 MAGNA CARTA
II. Origin of the Forests. Before the Norman Conquest
the Kings of England do not seem to have laid claim to
any exclusive prerogative in this respect. The only ordin-
ance of Canute on the subject, admitted to be authentic,
enacted merely that every man should have his own
hunting, while the King should have his.^ The rights of
the Crown, however, were strengthened by the events of
1066, and by the hardening of feudal theory which followed.
All unoccupied waste lands became royal property; and
these were the natural resorts of the larger sorts of game.
The King established a claim to an exclusive right to hunt
the more important species of animals ferae naturae, known
as " beasts of the forest " — embracing the red deer (harts
and hinds), the fallow deer (bucks and does), the roe deer
of both sexes, and the wild boar, w4th, exceptionally in
one forest, the ordinary hare.^ Henry I. formulated the
forest law, and it was probably due to him that " forest "
acquired its technical meaning. With the special meaning
came the express claim to a monopoly of hunting, together
with supreme and exclusive jurisdiction. The disorders
of Stephen's reign lowered the Crown's authority, and
Henry II. found the forests much curtailed. He had no
intention to acquiesce in this, but it was not till 1184 that
he attempted, by the Assize of Woodstock, to formulate
the rules of the forest law. In this sphere, as in so many
others, Henry II. built on foundations laid by his grand-
father. John's attitude to the forest laws was not consistent.
The monk of Barnwall relates how, in 12 12, John allowed
some relaxation in the severity of the forest code.^ More
characteristic of his normal attitude was the order issued
on 28th June, 1209, that hedges should be burned and
ditches levelled, so that, w^hile men starved, the beasts
might fatten upon the crops and fruits.^
Man wood, Booi of the Forests (1598) ; Coke, Fourth Institute^ 289-317 ; Lieber-
mann, Constitutiones de Foresta (1894) ; G. J. Turner, Preface to Select Pleas of
the Forest (1901) ; and an article in the Edinburgh Review for April, 1902.
1 Select Charters, 156. "Select Pleas of the Forest, xiii.
3 See W. Coventry, II. 207, and Stubbs' Preface, Ixxxvii.
• R. Wendover, III. 227. This, however, is clearly a hostile account of the King's
resumption of forest tracts illegally put under cultivation by way of purpresture.
CHAPTER FORTY-FOUR 417
III. Forest officials. The local magistrates who admin-
istered the rest of England were excluded from the forests
by a separate set of officials. At the head of this special
organization was placed, in early times, the Forest Justiciar
(called the chief forester in chapter 16 of the Carta de
Foresta), whose duties were divided in the year 1238, after
which there were two provinces separated by the river
Trent.^ His appointment was permanent, and his duties,
which continued between the eyres, were administrative
rather than judicial. He had discretionary authority to
release trespassers imprisoned for offences against the
forest law.2 Under his general supervision each forest, or
group of forests, was governed by a separate warden, aided
by a number of petty officials known as foresters, whose
duties were analogous to those of a modern gamekeeper,
but with magisterial powers in addition. Wardens were
of two classes — " the one appointed by letters patent under
the great seal, holding office during the King's pleasure;
the other hereditary wardens." ^ There was situated in or
near each forest of any extent a royal residence which, in
the Middle Ages, naturally took the form of a stronghold.
It was convenient that the office of warden should be com-
bined with that of constable of this neighbouring castle.^
" The wardens were the executive officers of the King in
his forests. Writs relating to the administration of forest
business, as well as to the delivery of presents of venison
and wood, were in general addressed to them." ^
The office was one of authority and profit, usually paid
in kind rather than by a salary. The warden often held
^ See Select Pleas of the Forest ^ xiv. The permanent routine work performed by
this functionary must not be confused with the intermittent duties of the Justices
of Forest Eyres, although he was usually a member of the commission who went
on circuit : e.g. chapter 1 6 of the Forest Charter speaks of the Chief Forester
holding pleas of the forest.
"^ Select Pleas, xv. ' Turner, in Select Pleas, xvii.
* Engelard de Cigogne, for example, whose name appears in chapter 50, occupied
this double position. Chapter 16 of Carta de Foresta forbids castellans to deter-
mine pleas of the forests, thus strengthening the presumption that wardens were
usually constables.
^ Select Pleas, xix.
2D
4i8 MAGNA CARTA
a fief by a tenure connected with the service, and enjoyed
rights and perquisites always of a valuable nature, though
varying with each forest. These were sufficient to provide
him with an income adequate to his position, and to allow
him to find the wages of his under-keepers, who ought
thus to have been paid officials. Such was the theory;
as matter of fact, the foresters, instead of receiving wages,
paid large sums to the warden, and recouped themselves
by extortions from the dwellers in their bailiwicks.^ These
unpaid foresters were expressively said to " live upon the
country." They may be classified in various ways, as,
into riding and walking foresters, or into foresters
nominated by the wardens, and foresters in fee. These
last had vested interests which the Forest Charter was
careful to respect; as, where chapter 14 reserved to them
the right to take "chiminage," or way-leave, denied to
other types of foresters. They might still enjoy, but not
abuse, the " vested rights " reserved to them. 2
With these professional gamekeepers there co-operated,
in later times at least, several groups of unpaid magistrates
appointed from the knights and freeholders of the district.
Of these honorary officials, whose original function was to
supply supplementary machinery for protecting the rights
of the Crown, but whose position as county gentleman, with
a stake in the district, led them also to act to some extent
as arbitrators between the King and outside parties, there
were three recognized kinds, (a) Towards the close of the
twelfth century officers known as verderers (usually four for
each forest) become prominent. They appear in the Carta
de Foresta of 1217, but had not been mentioned in the
Assize of Woodstock of 1 184. It is probable that the office
was devised in the interval as a check on the warden *s
power; just as the office of coroner had been instituted in
the reign of Richard as a drag on the sheriff. In other
important respects the duties of the verderers within the
forests resembled those of coroners within the rest of the
county. They were not royal employees, but local land-
owners whose unpaid magisterial services were required
* Se/ecf Pleas y xxi. ' The same chapter, however, fixed the rates of * ' chiminage. "
CHAPTER FORTY-FOUR 419
only on special occasions. They were responsible directly
to the King, not to the warden ; and were appointed in the
county court, their " election " taking place in accordance
with the terms of the writ " de viredario eligendo." They
attended the forest courts and swanimotes, and it appears
from chapter 16 of Henry's forest charter that it was their
duty to bring before the Justices in Eyre lists of all
offenders indicted in the lower courts. These " rolls of
attachment" were certified by their seals.^ (b) The
Regarders were twelve knights appointed in each forest
county to make tours of inspection every third year, finding
answers to a series of questions known as the " Chapters of
the Regard." In this way they reviewed the Crown's
interests alike in " the venison and the vert " (the technical
names for game and growing timber respectively), and
reported upon all encroachments : upon hawks and falcons,
bows and arrows, greyhounds and mastiffs (with special
reference to " expeditation " or cutting of their claws), ^ and
generally upon everything owned by private individuals
likely to harm the beasts of the forest.^ (c) The Agistors
are mentioned in the same clause of the Assize of Wood-
stock which mentions the Regarders. Four knights were
appointed to protect the King's interests in all matters
connected with the pasturing of swine or cattle within
the royal woods. For thirty days at Michaelmas, pigs were
turned loose to feed on acorns and beech mast, on payment
by their owners of a small fixed sum per head. The four
knights were required to take note of sums thus due, known
as "pannage," and to collect them at Martinmas.^
1 For the earliest notice of verderers see Se/ecf Pleas of the Forest^ xix. n. Their
appointment in county court may indicate that they acted in some measure as a
check on the professional foresters in the interests of the people generally, as well
as a check on the warden in the interests of the King. Within the forest the
warden, with the verderers and foresters, offered an exact parallel to the sheriff
with the coroners and bailiffs (or Serjeants) in other parts of a county.
^ See Carta de Foresta, c. 6.
' After 1217, if not before, it was their duty to fix the number of foresters required,
so that the inhabitants need not groan under a heavier burden than necessary.
*In one document they were styled agistatores precii {Select Pleas ^ p. 1.), which
suggests that fixing the rate was their chief duty. " Agist " was a general term ;
it was apparently correct to speak of "agisting a wood," of "agisting cattle," and
of ** agisting the money due."
420 MAGNA CARTA
Mention ought, perhaps, to be made of the private
foresters also, whom owners of woods within the forests
were obHged to appoint. These "wood wards," as they
were sometimes called, while paid for by the owner of the
wood, were expected to protect the King's interests. In
particular, they must prevent trees from being destroyed
or wasted : these formed shelter for the game.
IV. Forest Courts. The judicial side of the forest
system was developed in a manner equally elaborate.
Three sets of tribunals must be distinguished: (i) The
Court of Attachvients (or " view of attachments ") was a
petty tribunal, the chief duty of which was the taking of
evidence to be laid in due course before a higher court.
Exceptionally, however, it had power to inflict fines for
small trespasses against the "vert" — namely, for acts of
waste not exceeding the value of fourpence. It met once
in every forty days,^ which seems in practice to have been
interpreted as once every six weeks, the meetings being
always held on the same day of the week. 2 (2) Courts of
Inquisitions, When a serious trespass was discovered, a
special court was, in early days, immediately summoned.
The foresters and verderers conducted the inquiry, but it
was their right and duty to assemble the men of the neigh-
bouring townships to help them. In strictness, all inhabi-
tants might be compelled to attend. In practice, it was
sufficient if four men and the reeve represented each of the
four adjoining villages. Whenever a " beast " was found
dead in the forest, twenty men had thus to assemble, to
the neglect of their own affairs. In one district at least
(Somerton) the definition of beasts of the chase extended
to the ordinary hare ; and we read ^ how four townships
sat in solemn judgment, and found " that the said hare
died of murrain, and that they know of nothing else except
misadventure," and how, this verdict not giving satis-
faction, the townships were fined on the pretext that they
were not fully represented. The real offence was their
failure to disclose the culprit. Some alleviation of the
' Car^a de Foresta, c. 8. ^Select Pleas of the Forest ^ xxx.
^Select Pleas of the Forest, p. 42.
CHAPTER FORTY-FOUR 421
burden was effected when, at some date posterior to 12 15,
special inquisitions were superseded by one general inquisi-
tion, held at regular intervals (usually every six weeks),
to cover all trespasses committed during the interval.
These courts of inquiry (whether special or general) only
" kept " pleas without " trying " them — that is to say, they
received and recorded accusations, while judgment was
reserved for the justices. (3) Courts of the forest justices
in eyre. As the smaller courts, in the normal case, received
verdicts and reports, without punishing the offences
reported, it is evident that the whole system ultimately
depended on the justices. Their eyres, however, were
held at wide intervals — apparently once every seven years
during the reign of Henry III. A full attendance of forest
officials and of the public was summoned to meet them.
The evidence, stored up as a result of the work of the
smaller courts, supplemented by the Rolls of the Regard,
was laid before the justices, who summarily judged " pleas
of the vert," and "of the venison." These eyres came to
be known as "Courts of Justice Seat," but not until long
after the reign of John. No juries were present; the
justices punished offenders already convicted by juries at
a lower court.
In addition, there should be mentioned two other kinds
of assemblies which performed duties administrative rather
than judicial, as these terms are now understood. (4) The
regard, held once every three years — not by Crown officials,
but by what was practically a jury of local knights — has
already been referred to. These tours of inspection, some-
time known as visitationes nemorum,^ and sometimes even
as " views of expeditation," were of great practical import-
ance. The resulting report was placed before the justices
of eyre as evidence of forest trespasses. (5) Three times
every year, meetings, known from an early date as " Swani-
motes," were held to regulate the pasturing of swine and
cattle within the royal woods. A fortnight before Michael-
mas, the agistors met the foresters and verderers to provide
for the agisting of the King's woods, a process that lasted
'^ DialoguSy I. xi, E.
422 MAGNA CARTA
for thirty days — fifteen before and fifteen after Michaelmas.
At Martinmas the agistors collected the pannage in pres-
ence of the same officials. A third meeting was held in
June to make arrangements for excluding cattle from the
King's woods when the deer were fawning, but at this the
presence of the agistors were not required.^
The Carta de Foresta applies to these assemblies, and
to none other, the name " Swanimotes " — a word whose
correct use has been the subject of much discussion. Its
authoritative appearance in 12 17 affords strong evidence of
the original sense which it bore. In later days, however,
it was more loosely used, being applied to inquisitions and
also to courts of attachment. This has led to much con-
fusion, while its derivation has also been the subject of
discussion. Bishop Stubbs derived it from "swain," on
the supposition that courts so called were resorted to by
swains or country people. As matter of fact (whatever
doctrine may be correct philologically), these assemblies
were connected, not with "swains," but with "swine."
The peasantry were specially exempted; whereas all three
meetings sought to regulate the entry or exclusion of pigs
from the woods.
V. Chases, Parks, and Warrens. Forests were neces-
sarily royal monopolies and must on this and other grounds
be distinguished from three things with which they are apt
to be confused, (i) A " chase " was a district, once a royal
forest, which had, without any formal act of disafforesta-
tion, been granted by the King to a private individual.
The result was to transfer the monopoly of hunting to the
grantee, while modifying the nature of the rights trans-
ferred. The full force of the forest laws was abated,
although the extent and direction of this diminution was
nowhere strictly defined, but varied from chase to chase.
Such provisions of the forest law as continued to be binding
were no longer enforced by royal officials and royal courts,
but by those of the magnate, who obtained a franchise over
^ It is stated in Carta de Foresta (1217) that only verderers and foresters need be
present at the June moot, and the same officers, with the agistors, at the two
others. The public were exempted.
CHAPTER FORTY-FOUR 4^3
the chase and the royal beasts it contained. ^ (2) A " park "
was any piece of ground enclosed with a paling, or hedge,
whether with the object of protecting wild beasts or other-
wise, and the right to effect this was quite independent of
royal grant. If the owner of a manor in the near neigh-
bourhood of a royal forest wished to keep deer of his own,
which he might kill at pleasure, whether for sport or for
food, without infringing the forest laws, he had to stock
an enclosure with beasts legally his own, and to keep them
under conditions which made confusion with the King's
deer impossible.^ In 1234 the barons asserted their right
to keep private gaols for poachers taken in their parks
{in parcis et vivariis suis), but the King refused to allow
this.^ (3) A "warren," which might belong either to the
King or to any private owner, carried with it exclusive
rights of hunting within its bounds all wild animals,
except those technically defined as "beasts of the forest."*
In practice it chiefly embraced hares and foxes.^ Neither
parks nor warrens were protected by the forest law, but
by that part of the common law which related to theft and
trespass. This was, however, vigorously administered,
passing gradually into the modern Game Laws.^ Dr.
Stubbs held, apparently, too narrow a conception of warren,
when he read it in its modern sense of " a rabbit warren." ''
It was a tract of land wherein exclusive rights of hunting
lesser game (together with rabbits and other vermin) were
preserved to its owner. The King might, and did, have
his warrens and warreners, just as any subject might ; and
these royal warreners might inflict cruel injustice on the
common people ; ^ but their power was less than that of
foresters, as they were dependent on the common law.
The forest code did not apply even to royal warrens.^
^Select Pleas of the Forest ^ cix. et seq. ^Ibid.^ cxvii.
^ Statute of Merton, c. Ii. * Select Pleas of the Forest , cxxiii.
^ Ibid. , cxxviii. -cxxix. Wild cats should perhaps be added.
'^ See W. S. Holds worth, History of English Law, I. 346.
■^ See Select Charters, 552.
* Some of these Magna Carta sought to guard against. See c. 48.
'Rights of hunting were conferred on subjects over territory not their own.
Richard I. granted permission to Alan Basset to hunt foxes, hares, and wild cats
throughout the realm. Round, Ancient Charters, No. 18.
424 MAGNA CARTA
VI. Forest Rights and Forest Grievances. It is not
difficult to understand the store which the Kings of Eng-
land set upon their forests. They prized them not merely
as a pleasure ground, but also as a source of revenue.
Fines and amercements, individually small, but amounting
to a large sum in the aggregate, flowed into the Exchequer.
Great as were the pleasure and the profit to the King, the
burden and loss inflicted upon the people were greater out
of all proportion. Not only were the interests of forest-
dwellers sacrificed to the royal hunting, not only were legal
fines rendered trebly burdensome by the galling and waste-
ful manner of their collection ; but the men who paid them
were victims of illegal exactions in addition. These griev-
ances may be considered under seven heads : — (i) The
extent of the forests. The Crown constantly strove to
extend the boundaries; the people to contract them. The
Conqueror and Ruf us each " afforested " wide tracts of
land, of which the New Forest is only one example. In
the Charter of iioo, Henry bluntly declared: — "I retain
in my hand, by the common consent of my barons, my
forests as my father had them." This consent of the
magnates would suggest that the barons were allowed some
share in royal rights of hunting, which led them here to
make common cause with the Crown. Henry, as matter
of fact, retained not only the forests of his father but those
of Ruf us, and created new ones of his own.^ Stephen,
while retaining the forests of the two Williams, renounced
those added by Henry I. Under Henry 11., afforestation
began anew.^ The words of the Great Charter leave no
doubt that Henry of Anjou had extended the boundaries
of Stephen's forests; and that both Richard and John
* This is implied in the terms of Stephen's Oxford Charter. An example of an
act of afforestation by Henry is given in Select Pleas, 45, which shows how ** a
district could be afforested in a moment by the mere word of the monarch ; it took
centuries to free it from the royal dominion. " See Edinburgh Review, vol. cxcv.
(1902), p. 459. Even the Forest Charter (cc. i and 3) admitted the Crown's right
to afforest woods on its own demesne — reserving, indeed, common of pasture to
those with legal rights thereto.
^The policy of Henry I., Stephen, and Henry H. respectively" is well illustrated
by the case of Waltham forest ir. Essex. See Round, Geoffrey de Martdeville, 377-8.
CHAPTER FORTY-FOUR 425
carried the process further, bringing within the circle of
the cruel law, not only waste and moor, but also " woods "
belonging to private owners. These royal encroachments
were the more oppressive, occurring in an age when popula-
tion was increasing and seeking outlet in the reclamation
of waste places on the debateable land that surrounded the
forests. The vagueness of the frontier aggravated this
grievance, as it was often difficult for the honest reclaimer
of barren land to know when he was committing a trepass
for which he might be punished by a crushing fine.^
(2) The monopoly of hunting. The Crown also made
the law more stringent. The Crown's insistence on a
strict monopoly may not seem an important grievance, but
it was one likely to exasperate the sport-loving nobles.
John, in 1207, admitted that his barons still retained some
share in the hunting of royal beasts.^ These rights were
formally recognized and defined in 12 17. Chapter 11 of
the Carta de foresta allowed each magnate when passing
through a forest to take one or two beasts at sight of the
foresters, or, if these officials could not be found, then after
blowing a horn to show that nothing underhand was being
done.
(3) Interference with rights of property. Freeholders
whose lands lay in districts which the King was successful
in afforesting, retained their freeholds, but their proprietary
rights lost half their value. They could not root out trees,
to clear their own lands for cultivation ; for that was to
commit an assart. They could not plough up waste land
or pasture (even outside the covert) and turn it into arable,
nor build a mill, nor take marl or lime from pits, nor make
fishponds, nor enclose any space with hedge or paling ; for
these acts of ownership were purprestures or encroachments
on the King's rights. They could not destroy a tree or
^ This group of grievances was partly remedied by chapters 47 and 53 of Magna
Carta. The former provided for the summary disafforestation of all districts made
forests by -Richard and John, while the latter showed a more judicial spirit in the
undoing of the similar work effected by their father. The Carta de Foresta of 12 17
contained clauses which took the place of these somewhat crude provisions.
2 See Rot. Claus., I. 85 (dated nth June, 1207).
426 MAGNA CARTA
lop off branches (except under stringent conditions), with-
out being guilty of waste.'^ They could not agist their
woods until a fortnight after Michaelmas, when the agisting
of the King's demesnes was over (thus reserving for him
the best market and " pannage dues ").^ Heavy tolls were,
under the name of "chiminage," taken from carts and
sumpter-horses passing through the woods. The Great
Charter endeavoured to strike at the abuse of these Crown
rights by providing machinery for the abolition of " evil
custorns." The Carta de foresta entered more into detail.
Not only were past trespasses of all three kinds — wastes,
purprestures, and assarts — to be condoned, but the law
was altered for the future. The long list of purprestures
was curtailed : it was made lawful for a man to make (on
his own freehold in the forest) mills, ponds, lime pits,
ditches, and arable lands, provided these were not placed
within the covert and did not infringe on any neighbour's
rights.^ He might also keep eyries for breeding falcons
and other birds of prey, and take honey found on his own
ground — rights previously denied.^
(4) Interference with the pursuits of the poor. If the rich
suffered injury in their property, the poor suffered in a
more pungent way : stern laws prevented them from
supplying three of their primary needs ; food, firewood, and
building materials. On no account could they kill deer;
while difficulties surrounded the taking of timber from the
woods.^ It is true that even the Assize of Woodstock
allowed them the privilege of " estovers " (that is of cutting
firewood), but only under stringent rules. All waste was
prohibited ; and " waste " was a wide word covering, not
merely wanton destruction, but all sales or gifts of logs;
while nothing could be taken except at sight of the forester,
^For detailed information as to wastes, purprestures, and assarts with their
ascending scale of penalties, see Seka P/eas, Ixxxii.
"See Assize of Woodstock, article 7. *See Car^a de Foresta, c. 12.
^Ibid.^ c. 13; another clause (c. 14) forbade ordinary foresters to exact
chiminage, and fixed the rates payable to those with vested rights at two pennies
for each cart per half-year, and one half-penny for each sumpter horse.
^ See Assize of Woodstock, article 3.
CHAPTER FORTY-FOUR 427
whose consent would not be procured for nothing. This
may be illustrated from a period sixty years later than
John's reign : Hugh of Stratford, who paid two and a half
marks of yearly rent to the Warden for his post, recouped
himself by taking " from the township of Denshanger for
every virgate of land one quarter of wheat in return for their
having paling for their corn and for collecting dead wood
for their fuel in the demesne wood of the lord king; and
from the same town he took from every house a goose and
a hen in every year." ^ A sum might be taken for every
load of sticks; the men of Somerset complained that
" from the poor they take, from every man who carries
wood upon his back, sixpence." ^ Dwellers within or near
the forests were prohibited from keeping dogs, unless
their value for other pursuits, as well as for hunting, was
destroyed by the removal of three claws of the forefoot.^
Nor could they keep bows or arrows, so necessary for their
protection amid the dangers that beset the inhabitants of
lonely districts throughout the Middle Ages.^ No tanner
or bleacher of hides could reside in forest districts, unless
within a borough.^
(5) Attendance at forest courts. At every inquisition,
representatives from neighbouring townships must be
present, while the entire population were compelled to
meet the justices on their forest eyres. Henry H. enforced
this duty upon those outside the boundaries as well as on
those within. The Assize of Woodstock admits no exemp-
tion for earl or baron, for knight or freeholder, nor even
(according to one version) for archbishop or bishop. The
double duty of doing suit at county courts and forest courts
meant double loss of time, and double risk of amercement.
This nth Article of the Assize was repealed by chapter 44
of Magna Carta, which restricted the obligation to denizens
of the forests, a concession confirmed in 1217.^
1 See Seka Pleas, 123 (6 Edward I.).
"^ Select Pleas ^ (127 (1278-9). This was a heavy rate, the more remarkable in
face of the provisions against "chiminage" in Carta de Foresta, c. 14.
^ Assize of Woodstock, article 14. Cf. Carta de Foresta, c. 6.
^ Ibid,^ article 2. ^ Ibid., article 15. *See Carta de Foresta, c. 2.
428 MAGNA CARTA
(6) Fines and punishments. Frequent exactions ground
down the dwellers in royal forests to abject poverty. If
they failed to attend one of the numerous inquisitions or
to disclose the guilty poacher, they paid a fine. If they
gave false information; sold or gave away timber; kept
grey hounds or mastiffs, which had not been " lawed, "
they paid a fine.^ If a bow or arrow were found in their
keeping ; if they committed any one of the numerous forms
of waste or trespass, they paid a fine.
The Northampton Eyre Roll of 1209 illustrates how a
township might suffer severely for no fault of their own.
" The head of a hart recently dead was found in the wood
of Henry Dawney at Maidford by the King's foresters.
And the forester of the aforesaid Henry is dead. And
because nothing can be ascertained of that hart, it is
ordered that the whole of the aforesaid town of Maidford
be seized into the King's hand, on the ground that the
said Henry can certify nothing of that hart." ^ There was
a strong inducement to find someone guilty.
In certain cases Henry II. would not accept a fine, but
inflicted mutilation upon violators of the King's monopoly.
It was often better to kill a fellow-man than a boar or stag.
Article i of the Assize of Woodstock announced that the
full rigour of the laws would be enforced, as under Henry I.,
while article 12 laid down more definitely that sureties
would only be accepted twice. For the third offence
nothing would suffice save the body of the offender. John's
Magna Carta made no regulation on this head; but
chapter 10 of the Carta de foresta in 12 17 conceded that
no one should henceforth lose life or limb for such offences.
The culprit should lie in prison for year and day, and
thereafter find sureties for his good behaviour, or be
banished the realm.
(7) Arbitrary government and illegal exactions. If the
laws of Henry's code were stringent and the legal pay-
^ It had been the practice to exact an ox in reparation of such transgression, thus
leaving the peasant without means of tilling his land. The Forest Charter (c. 6)
limited the fine to 3s.
2 See Select Forest Pleas y p. 4.
CHAPTER FORTY-FOUR 429
ments onerous, it was a worse evil that the law could be
defied by Crown officials, and that payments of a perfectly
illegal nature might be freely exacted. Within the forest
bounds, the peasantry lived in daily fear of the discretionary
authority of officials, whose most unreasonable wishes they
dared not oppose. Sometimes a local tyrant established a
veritable reign of terror. This happened in the forest of
Riddlington under Peter de Neville, as the records of the
Rutland Eyre, held in 1269, disclose. One item, taken
almost at random from the long list of his evil deeds, will
suffice : " The same Peter imprisoned Peter, the son of
Constantine of Liddington, for two days and two nights
at Allexton, and bound him with iron chains on suspicion
of having taken a certain rabbit in Eastwood; and the same
Peter the son of Constantine, gave two pence to the men of
the aforesaid Peter of Neville, who had charge of him, to
permit him to sit upon a certain bench in the gaol of the
same Peter, which is full of water at the bottom." ^ Other
examples are only too abundant. In 1225, Norman Samson,
a petty official of the forest of Huntingdon, put men to
the torture without cause, and only released them from
their torments in return for heavy bribes. If such things
could happen after the Charters of 12 15 and 1217, it is not
likely that foresters were more merciful before. John was
always too indifferent or too busy to redress such wrongs.
The only guarantee against their recurrence was that honest
officials sTiould be selected. Chapter 45 of Magna Carta,
which tried to effect this, was withdrawn in 12 16.
Some good must have resulted from chapter 16 of the
Forest Charter, which forbade wardens to hold pleas of
the forest. This prevented wardens from being judges in
their own cause; but their arbitrary acts continued to be
plentiful under Henry III., as has been already shown.
Sixty years after Magna Carta, the men of Somerset com-
plained that "foresters come with horses at harvest time
and collect every kind of corn in sheaves within the bounds
of the forest and outside near the forest, and then they
make their ale from that collection, and those who do not
^ Select Pleas ^ 50.
430 MAGNA CARTA
come there to drink and do not give money at their will,
are sorely punished at their pleas for dead wood, although
the King has no demesne; nor does anyone dare to brew
when the foresters brew, nor to sell ale so long as the
foresters have any kind of ale to sell; and this every
forester does year by year to the great grievance of the
country." ^ Each one of these abuses had been forbidden
by chapter 7 of the Carta de foresta, which had prohibited
the making of "scotale" and the collection of corn, lambs,
and pigs. Such rules were easier to enunciate than
enforce.
VII. Later History of Forests and Forest Laws. The
Forest Charter signally failed to secure a pure administra-
tion of the law; but two ameliorating processes were at
work. The long struggle to define the boundaries ended,
in the reign of Edward II., in the defeat of the King, who
consented to the frontier being drawn to suit the barons.^
Within these restricted limits, time and the progress of
civilization softened the severity of the forest code, many
customs becoming obsolete.^ Charles I. made an ill-judged
attempt to revive some of the Crown's long-forgotten rights.
Justice-seats were held by the Earl of Holland, accompanied
by amercements and attempts to extend the forest bounds.'*
The result was a drastic act of the Long Parliament, limiting
them to their old extents.^ This statute, however, abolished
neither the forests, the forest laws, nor the forest courts.
After the Restoration a Justice-seat actually took place
pro forma before the Earl of Oxford. Blackstone declares
this to be the last ever held,^ although the offices of justice
and warden of the forests were not abolished till 181 7."^
The forests, much curtailed in extent, are still Crown pro-
perty, now administered in the interests of the public by
^Select Pleas ^ 126. -See infra, under c. 47.
^"Assz'sa et consuetudines forestae,^^ issued by Edward I. in 1 278, although
declaratory, may have done something towards curtailing discretionary authority.
Statutes of Realniy I. 243 ; Bemont, Charles^ Ixv.
*See S. R. Gardiner, Hist. Engl., VII. 363, and VIII. 282.
* 16 Charles I. c. 16. ' CompientaHeSy III. 72.
' By 57 George HI. c. 61.
CHAPTER FORTY-FOUR 431
Commissioners of Woods and Forests.^ The operation of
the common law is, of course, no longer excluded from
their confines, the old antithesis between forest law and
the law of England being a thing of the past.^
CHAPTER FORTY-FIVE.
Nos non faciemus justiciarios, constabularios, vicecomites
vel ballivos, nisi de talibus qui sciant legem regni et earn
bene velint observare.
We will appoint as justices, constables, sheriffs, or bailiffs
only such as know the law of the realm and mean to observe it
well.
The object of this plainly worded clause was to prevent
the appointment of unsuitable men to responsible posts
under the Crown. The list of officers is a comprehensive
one — justices, sheriffs, constables and bailiffs — embracing
all royal ministers and agents, both of the central and of
the local government, from the chief justiciar down to the
humblest serjeant.^ This clause was directed in particular
against John's foreign favourites such as the Poitevin
Bishop of Winchester, Peter des Roches, who had wielded
the authority of chief justiciar in 12 14 when the King was
abroad,* or such as Engelard de Cigogne, stigmatized by
name in a later part of Magna Carta. ^ Such men had no
interests at stake in England, and little love for its
customs and free traditions. In future John must choose
a different type of servants, avoiding all such unscrupulous
men, whether Englishmen or foreigners, as were ready to
break the law in their master's interests or their own.
But what class were to fill their places ?
Bishop Stubbs credits the framers of the Charter
^ In virtue of a series of Acts of which 14-15 Victoria c. 42 is the latest.
^ See Stephen, Commentaries, II. 465-6.
' Constable and bailifif are discussed sjipra, c. 24, and shown to include forest
magistrates, supi-a^ c. 44.
* See supra, p. 30, and cf. Blackstone, Great Charter, viii.
^ See c. 50.
432 MAGNA CARTA
with an intention to secure the appointment of men well
versed in legal science : " on this principle the steward of
a court-leet must be a learned steward."^ The clause of
Magna Carta, however, refers to royal nominees, not to the
officers appointed by mesne lords to preside over their
feudal courts. The barons appointed their own stewards
and bailiffs, and had no wish to hamper their own freedom
of choice; but only that of the King. Further, the barons
did not desire that John should employ men steeped in legal
lore, but plain Englishmen with a rough-and-ready know-
ledge of insular usage, who would avoid arbitrary acts con-
demned by the law. The barons at Runnymede desired
precisely what the council of St. Albans had desired on
4th August, 1 2 13, when it issued formal writs to sheriffs
and foresters to observe the laws of Henry I. and abstain
from unjust exactions ; ^ and these laws of Henry werei
but the laws of Edward Confessor (or, in reality, of Canute)]
slightly amended.
The attitude of John's barons was the same as that ofj
Henry's barons, when the latter declared, in 1234, inj
emphatic terms, that they did not wish the laws of Eng-J
land to be changed.^ They were far from desiring to be!
governed by ministers deeply versed in the science andl
literature of jurisprudence, since these would necessarilyj
have been churchmen and civilians.
This well-meaning provision of Magna Carta disappeare<
in 1 2 16 (without any comment in the so-called "respiting
clause "). Even if it had remained intact, it would not hav<
effected much, in the absence of adequate machinery t(
ensure its enforcement. In promising the appointment oi
such ministers as knew the law and meant to keep it, John]
remained sole judge of the men appointed and their inten-
tions. The clause indicated no standard of fitness, no
neutral arbitrator to decide between fit and unfit,^ and no
^ Const. Hist., I. 578 n. «Cf. supra, p. 28.
' '^Nolunt leges Anglie mutare que usitate sunt et approbate.^' See Statute of
Merton, c. 9.
^ It would have been a notable anticipation of modern constitutional theory if the
barons in 121 5 had referred such questions to the decision of the Commune
Concilium summoned as in c. 14 (g.v.).
CHAPTER FORTY-FIVE 433
sanction to enforce compliance on an unwilling King.
Half a century later, the Provisions of Oxford gave proof
of some advance in political theory. They contained an
expedient, crude enough it is true, for constraining royal
officials to keep the law : forms of an oath of office to be
taken by castellans and ministers of all grades were care-
fully provided.^ Even this was only a first step towards
settling a problem not completely solved until the
modern doctrine of ministerial responsibility was firmly
established.^
CHAPTER FORTY-SIX.
Omnes barones qui fundaverunt abbatias, unde habent
cartas regum Anglie, vel antiquam tenuram, habeant earum
custodiam cum vacaverint, sicut habere debent.
All barons who have founded abbeys, concerning which they
hold charters from the kings of England, or of which they have
long-continued possession, shall have the wardship of them,
when vacant, as they ought to have.
Religious houses of various orders (abbeys, priories, and
convents), which had increased rapidly in number since
the reign of Henry I., fell naturally into two classes,
according as they had been founded by the King or by
private individuals. The King or the great baron, in
bestowing lands on a religious foundation, reserved, either
expressly or by implication, valuable rights of property :
of these the control over the election of the abbot or prior,
together with the wardship of the fief during vacancies,
were the most important. King John, while by his charter
to the clergy he renounced control over election of bishops,
reserved his rights of wardship; and the barons insisted
that the proprietary rights of mesne lords who had founded
religious houses, should also be respected. John, however,
^ See Select Charters^ 388-391, and Madox, II. 149, with authorities there cited.
^ Prof. Adams seems to make too much of this chapter {Origin, 259-260). It is
only a vague promise to employ honest officials : it confers no constitutional veto
upon anyone. Had the function of defining fit ministers been conferred on the
Common Council, it would have been a notable innovation.
2 E
434 MAGNA CARTA
wherever he had any plausible pretext, usurped the ward-
ship over private foundations. It would appear, from the
terms of a later chapter,^ that in 12 15 the Crown actually
held in ward certain abbeys founded by mesne lords. The
present chapter looks to the future, forbidding new usurpa-
tions of this nature.
In reissues of the Charter verbal changes occur, but it is
not clear that they imply changes of substance. In 12 16
the words " and as it has been above declared " were added,
implying that the rights of mesne lords were to be restricted
by the rules previously laid down in chapter 5, as to ward-
ship— rules particularly applied to the lands of bishoprics
and religious houses in 12 16 by a clause which had no
parallel in John's Charter.^ In 12 17 three other small
changes tend to define and perhaps to widen the scope of
the clause. The " barons who have founded abbeys "
become " the patrons of abbeys " ; royal " charters "
become more explicitly " charters of advowson " ; " ancient
tenure " is expanded into " ancient tenure or possession." ^
These alterations seem to indicate an effort towards greater
verbal accuracy, and do not involve any change of principle.
It should, perhaps, be noted, however, that the words
" patroni " and " de advocatione," occurring in 12 17, contain
a tacit assertion of lay patronage of which there was no hint
in 1215 ; but it would not be safe to conclude from this alone
that there had been any change of attitude on the question
of canonical election.
The object of this chapter was to define the relations
between the King and the barons as to wardship, not those
between the lay and ecclesiastical authorities as to rights of
appointment. It seems to have made little difference, if
any, in practice: Henry III. never observed in its fullness
the doctrine here enunciated, but claimed wardship over
abbeys and priories founded by earls and barons on their
own fiefs.** On the closely allied question of lay patronage,
*See m/ra, c. 53. 2Qf^ sn/>ra, p. 212.
'See Appendix for final form in charter of 1225.
*See Petition of Barons, c. 11 {Sel. Chart., 384) ; Maitland, Sel. Pleas Man.
Courts, Ixxvii. For the practice in Normandy, see authorities cited by Adams,
Origin, 246 n.
CHAPTER FORTY-SIX 435
not directly raised in any version of Magna Carta, Henry's
practice seems not to have differed from his father's. John
interfered freely between abbeys and their founders. On
i6th August, 1 200, he granted to William Marshal the
privilege of bestowing the pastoral staff of Nuthlegh Abbey,
which lay within that earl's fief ; this shows that he forbade
appointments without royal licence. ^ The barons in 1258
protested against similar conduct on the part of Henry HI.^
CHAPTER FORTY-SEVEN.
Omnes foreste que afforestate sunt tempore nostro, statim
deafforestentur; et ita fiat de ripariis que per nos tempore
nostro posite sunt in defenso.
All forests that have been made such in our time shall forth-
with be disafforested; and a similar course shall be followed
with regard to river-banks that have been placed " in defence "
by us in our time.
An analogy may be traced between the prerogatives of
hunting and of falconry here brought together. William
the Conqueror claimed wide and ill-defined rights to
" afforest " whole districts at his discretion ; and for pro-
tecting his preferential rights of fowling, whole rivers might
be placed " in defence." The parallel must not be pushed
too far. River-banks were preserved only for such limited
period as was covered by the King's express command;
and, although wardens were appointed to guard them,^ the
Crown never established such absolute control over the
banks of rivers as it did within districts declared
" afforested."
The provision of the present chapter, defining what river-
banks might be " defended," disappeared, together with the
^See New Rymei\ I. 8i. John had also interfered "in the time of the
interdict " with what Robert fitz Walter considered his rights of patronage over
Binham Priory (a cell of St. Albans). See J. H. Round, Eng. Hist. Rev., XIX.
710-11.
2 Petition of Barons, c. ii {Sel, Chart,, 384).
* Mention of these officers is made in c. 48. The phrase *' in defence" is
explained supra, pp. 301-3.
436 MAGNA CARTA
relative clause of chapter 48 {'' ripariis et earum custodi-
hus")y from the reissue of 12 16; but, in the "respiting"
clause there was promised further deliberation, which
resulted in its replacement in chapter 20 of the final version
of Magna Carta.^
More attention is usually paid to the bearing of the
present chapter upon the limits of the forests. John, if he
had created no new forests, had extended the boundaries of
the old ones. All such encroachments are to be immedi-
ately given up. This summary redress should be contrasted
with the more judicial procedure appointed by chapter 53
for determining encroachments made by Henry II. and
Richard. A somewhat similar distinction is also to be
found in the corresponding provisions of the Forest Charter
of 12 17 (chapters i and 3); but the line is there differently
drawn. Chapter i of the Carta de foresta extends the
summary methods of redress to the disafforesting of all
forests created by Richard as well as those created by John.
The terms of the later document are also more detailed.
Both seem to be directed against encroachments on the
rights of landowners, affording no protection to the poor.
While they deny the Crown's right to afforest private woods
" to the damage of any one " (that is, of barons or free-
holders owning them), they admit the legality of past acts,
whether of Henry, of Richard, or of John, in afforesting
Crown lands, subject always to a saving clause in favour of
freeholders in right of common of pasturage. ^
Even if Henry III. had cordially co-operated with his
barons to disafforest all tracts of ground afforested by Henry
II. and his sons, difficulties of definition would still have
made the task tedious. As it was, struggles to settle boun-
daries embittered the relations between Crown and Parlia-
ment, until the very close of Edward Plantagenet's reign.
Only the leading steps in the slow process by which the
opposition triumphed need here be mentioned.
^ Cf. supra, p. 147.
*G. J' Turner, Select Pleas of Forest^ xciii., points out that although forests
included open country as well as woods, yet Carta de foresta spoke only of
** woods" in this connection.
I
CHAPTER FORTY-SEVEN 437
After the issue of Carta de fores ta on 6th Nov ember y 1217,^
machinery was set in motion, in obedience to its terms,
to ascertain the old boundaries and disafforest recent
additions. The work of redress continued for some years,
suffering no interruption from the issue of the new royal
seal at Michaelmas, 1218.^ In face of many difficulties,
only slow progress was possible. More strenuous efforts
followed the reissue of the Charters on nth February,
1225 ; 3 for, five days later, justices were appointed to make
new perambulations, which resulted in the disafforestation
of wide tracts. Henry considered himself, and with some
reason, unjustly treated by these justices, or by the local
juries on whose verdicts they had relied. After he had
proclaimed himself of age in January, 1227, he challenged
their findings; and this has been misinterpreted as an
attempt to annul the Forest Charter.*
Some of the knights who had perambulated the forests
were persuaded or coerced into acknowledging that they
had made mistakes; and, after further inquiry, Henry
restored the wider bounds. His reactionary measures went
on for two years ; but thereafter the frontiers were fixed, in
spite of many complaints, until strong pressure compelled
Edward I. to reopen the whole question. Perambulations
in 1277 and 1279 produced apparently no results. Renewed
complaints were followed by new perambulations in 1299-
1300, the reports of which were laid before a Parliament at
Lincoln on 25th January, 1301. The King on 14th Febru-
ary confirmed the Forest Charter, and agreed to the reduced
boundaries as defined by the most recent inquests. Edward
had acted under constraint : on this plea he subsequently
obtained from Pope Clement V. a bull, dated 29th Decem-
ber, 1305, revoking all concessions made at Lincoln.^ The
Crown seemed thus to triumph once more ; but the barons
refused to accept defeat, forcing upon Edward H. the
acceptance of the narrower bounds as defined at his father's
^Cf. supra, p. 146. ^Cf. sup7'a, p. 153, and see Selecf Fleas, xcv.
^Cf. supra, p. 154. *Cf. Select Pleas, xcix. ; and see also supra, p. 156.
^ See Select Pleas, cv. Mr. Turner's account of Edward's conduct may be com-
pared with the estimate of M. Bemont, Chart es, xlviii.
438 MAGNA CARTA
Parliament in 1301. This settlement was confirmed by
statute in the first year of Edward III.^ and that King
failed in all attempts to escape from its provisions. Thus
the authoritative pronouncement made in 1301 by the
Parliament of Lincoln furnished the basis on which the
protracted controversy was finally determined.^
The further history of the forest boundaries may be told
in a few sentences. No changes were made until the six-
teenth century. When Henry VIII. afforested the districts
surrounding Hampton Court in 1540, he did so by consent
of Parliament, and on condition of compensating all who
suffered damage. The same course was followed by
Charles I. in creating the Forest of Richmond in 1634.
Finally, as a result of attempts of the Stewarts to revive
obsolete rights, a statute of the Long Parliament, reciting
the Act of 1327, " ordained that the old perambulation of
the forest in the time of King Edward the First should be
thenceforth holden in like form as it was then ridden and
bounded." ^
CHAPTER FORTY-EIGHT.
Omnes male consuetudines de forestis et warennis, et de
forestariis et warennariis, vicecomitibus et eorum ministris,
ripariis et earum custodibus, statim inquirantur in quolibet
comitatu per duodecim milites juratos de eodem comitatu,
qui debent eligi per probos homines ejusdem comitatus, et
infra quadraginta dies post inquisicionem factam, penitus,
ita quod numquam revocentur, deleantur per eosdem, ita
quod nos hoc sciamus prius, vel justiciarius noster, si in
Anglia non fuerimus.'*
All evil customs connected with forests and warrens, foresters
and warreners, sheriffs and their officers, river-banks and their
1 1 Edward III., stat. 2, c. i.
^ See Select Pleas ^ cvi. There was one exception. On 26th December, 1327,
Edward III. had to submit to further disafforestations in Surrey.
3 16 Charles I. c. 16.
^ The last sixteen words, inclusive of ^* per eosdem" appear at the foot of both of
the Cottonian versions of Magna Carta. Cf. supra, p. 166.
CHAPTER FORTY-EIGHT 439
wardens, shall immediately be inquired into in each county by
twelve sworn knights of the same county chosen by the honest
men of the same county, and shall, within forty days of the said
inquest, be utterly abolished, so as never to be restored, provided
always that we previously have intimation thereof, or our justiciar,
if we should not be in England.
This chapter is mainly, though not exclusively, a forest
one. It provides in a sweeping and drastic manner for
the abolition of " evil customs," three groups of which are
specially emphasized : (a) those connected with forests and
warrens (presumably royal warrens only), or with their
officials ; (b) those connected with sheriffs and their subor-
dinates ; and (c) those connected with river-banks and their
guardians. The word " customs " is obviously here used
in its wider sense, embracing all usages and procedure,
whether specially connected with pecuniary exactions or
not.^ The word " evil " is not defined, but machinery is
provided for arriving at a definition. In each county a
local jury of twelve knights was to be immediately chosen
by " the good people " of that county, and these twelve
received a mandate to hold a comprehensive inquest into
" evil customs " : practices condemned by them were to be
abolished within forty days of the inquiry, " so that they
shall never be restored."
At the end of the chapter appears a proviso that, before
actual abolition, notice must be sent to the King, or, in
his absence, to his justiciar. Although such intimation
was necessary, both on grounds of policy and of ordinary
courtesy, this clause is written (apparently as an after-
thought) at the foot of two of the copies of the Great Charter.
John lost no time in instituting machinery for effecting
this part of the reforms. On the very day on which terms
of peace were concluded at Runnymede, namely, on 19th
June, 1 2 15, he began the issue of writs to sheriffs, warreners,
and river bailiffs. Within a few days every one of these
had been certified of the settlement arrived at, and had
received commands to have twelve knights chosen in the
first county court to make sworn inquest into evil customs.^
^ Contrast the more restricted meaning of the same word in c. 41.
^See I^ot, PaL, I. 180, cited also Select Charters^ 306-7. Cf. supra, p. 42.
440
MAGNA CARTA
The knights appointed seem to have taken a liberal view
of their functions, claiming to share with the sheriffs the
exercise of the whole executive authority of the county.
Some warrant for these pretensions may be found in the
terms of a second series of writs issued in the King's name
on 27th June and following days. These were addressed
to the sheriff and the twelve knights jointly, commanding
them to make instant seizure of all who refused to take, as
required in the previous writs, the oath of obedience to the
twenty-five executors of the Charter.^ The revolutionary
committee of the central government had thus, in each
county, local agents in the twelve knights whose original
duties had been to see evil customs abolished.
The hatred to the forest laws is well illustrated by the
iconoclastic spirit in which these knights set about the
remedy of abuses. Moderate-minded men began to fear
that sweeping changes would abolish the royal forests.
Accordingly, the leading prelates issued a written protest
that this chapter must be understood by both parties " as
limited," and " that all those customs shall remain, without
which the forests cannot be preserved." ^ What effect, if
any, this protest had, is not known. The country was soon
plunged in civil war, during the continuance of which
neither side had leisure for the reform of abuses. In 12 16
the subject was " respited " for future consideration, and in
1 217 an attempt was made to specify in detail the evil cus-
toms to be abolished. The dangerous experiment of leaving
the definition to local juries in each district was not
repeated.
^Cf. infra^ c. 6i.
*Cf. supra, p. 43. The text is given Rot. Claus., I'j John, m. 27 d. and N'ezv
Rymer, I. 134. It runs in name of the archbishops of Canterbury and Dublin,
and the bishops of London, Winchester, Bath, Lincoln, Worcester and Cov^entry,
comprising (with one exception) those mentioned in the preamble to Magna
Carta. For text, see Appendix.
CHAPTER FORTY-NINE 441
CHx\PTER FORTY-NINE.
Omnes obsides et cartas statim reddemus que liberate
fuerunt nobis ab Anglicis in securitatem pacis vel fidelis
servicii.
We will immediately restore all hostages and charters delivered
to us by Englishmen, as sureties of the peace or of faithful service.
A feature of John's system of government was the
constant demand for hostages as guarantees of his subjects'
loyalty. Such an expedient was, indeed, naturally resorted
to in the Middle Ages upon special occasions, as, for
example, to secure the observance of a recent treaty, or
where the leaders of a rebellion, newly suppressed, had been
spared on condition of future good behaviour. Thus the
Conqueror, in 1067, during a forced absence from England,
took with him Edgar Atheling and the Earls Morkere and
Edwin. Such cases were, however, exceptional, until John
resorted to such a policy, not merely in face of danger, but
as a constant and normal practice in times of peace.
John lived in his native England like a conqueror in the
midst of a hostile race, keepmg sons and daughters in his
clutches to answer for their parents' attempts at revolt.
This ingenious but unfair practice accords well with what
we know of John's character and general policy. It was a
measure of almost devilish cunning for obtaining his
immediate ends, but likely to recoil on himself whenever a
critical state of his fortunes arrived. Its efficacy lay in this,
that it forced the hand of discontented magnates, compelling
them to decide, upon the instant, between the desperate
expedient of open rebellion and delivery of their children
to an unscrupulous enemy, thus renouncing, perhaps for
ever, the possibility of resistance or revenge, thereafter to be
purchased at too dear a price — the life of the hostage. By
thus paralyzing his enemies one by one, John hoped to
render disaffection innocuous.^
^The only magnates not exposed to this dilemma were the prelates, whose
celibacy cut them adrift from acknowledged family ties. They had no hostages
442 MAGNA CARTA
The history of the reign shows of what excessive practical
importance this question of hostages had become. Thus,
in 1201, John seized the castles of certain of his barons ; and
one of them, William of Albini, only saved his stronghold
of Belvoir by handing over his son as a hostage.^ In the
same year, the men of York offended the King by omitting
to meet him in procession when he visited their city, and by
their failure to provide for the billeting of his archers.
John, as usual, demanded hostages, but ultimately allowed
the citizens to escape on payment of ;^ioo, to buy good-
will.2
Hardly a year passed without similar instances; but,
apparently, it was not until 1208 that the practice was
enforced wholesale. In that year, the King's abject fear
of the effects of the Pope's absolution of his barons from
their allegiance, led to his demand that every leading man
in England should hand over his sons, nephews, or other
blood relations to the King's messengers.^
The danger of failure to comply with such demands is
illustrated by the fate of Maud of Saint- Valery, wife of
William de Braose, who refused point-blank to hand over
her grandchildren to a King who, she was unwise enough
to say, "had murdered his captive nephew." ^ Two years
later John, after failing to extort enormous sums in name of
fines, caused her, with her eldest son, to be starved to death,
a fate to which her own imprudence had doubtless contri-
buted.^ John's drastic methods of treating his hostages
may also be illustrated from the chronicles of his reign, for
example, from the fate of the youths he brought from Wales
in June, 121 1. When he heard of the Welsh rebellion of
the following year, he ordered his levies to meet him at
Nottingham. At the muster, early in September, John
found awaiting him a great concourse, who were treated to
to give, and were, further, in the normal case, exempt from fear of personal
violence.
^See R. Hoveden, IV. 161. 'See Rotuli de Finibnsy p. 119.
2 See R. Wendover, III. 224-5, ^^d M. Paris, II. 523.
* R. Wendover and Matthew Paris, ibid.
•* See authorities cited by Miss Norgate, John Lacklandy p. 288.
CHAPTER FORTY-NINE 443
an object lesson which long might haunt their dreams.
His passion at white heat, John incontinently hanged eight-
and-twenty defenceless boys of the noblest blood of Wales. ^
This ghastly spectacle could not have been forgotten,
when later in the same month the King, in the throes of
sudden panic, fled to London ; and, secure in the fast-
nesses of the Tower, demanded hostages w^holesale from all
the nobles whose fidelity he doubted. Eustace de Vesci and
Robert fitz Walter preferred to seek safety in flight.^ The
others, with the Nottingham horror fresh in their memories,
were constrained to hand over sons and daughters to the
tender mercies of John, cunning and cruel by nature, and
rendered doubly treacherous by suspicion intensified by
fear.
The defects of this policy, in the long run, may be read
in the events which preceded Magna Carta. When John's
hold on the hostages was relaxed, because of the campaign
of 1 2 14, ending as it did in discomfiture, the disaffected
were afforded their long-desired opportunity, and were
stimulated to rapid action by the thought that such a chance
might never occur again. John, on his return, held com-
paratively few hostages, and the northern barons saw that
they must act, if at all, before their children were once more
in the tyrant's clutches.
Even in June, 12 15, however, John had still a few
hostages, and this chapter demands the immediate restora-
tion of those of English birth (the Welsh receiving separate
treatment), together with the charters which John held as
additional security. This provision of Magna Carta was
immediately carried out. Letters were dispatched to the
■custodians of royal hostages, ordering an immediate
release.^ The practice of taking hostages, however, by
no means ended with the granting of the Great Charter.
Before a year had run, some of the insurgent nobles,
repenting of their boldness, succeeded in making terms
with John by the payment of large sums of money and the
delivery of their sons and daughters in security for their
^ Cf. supra^ p. 25. 2 Cf. supra^ p. 25.
^ See letter of 23rd June to Stephen Harengod in Appendix.
444 MAGNA CARTA
future loyalty. Simon fitz Walter, for example, thus gave
up his daughter Matilda.^
CHAPTER FIFTY.
Nos amovebimus penitus de balliis parentes Gerardi de
Athyes, quod de cetero nullam habeant balliam in Anglia ;
Engelardum de Cygony, Petrum et Gionem et Andream,
de Cancellis, Gionem de Cygony, Galfridum de Martinny
et fratres ejus, Philippum Marci et fratres ejus, et Galfridum
nepotem ejus, et totam sequelam eorundem.
We will entirely remove from their bailiwicks, the relations of
Gerard of Athee (so that in future they shall have no bailiwick
in England); namely, Engelard of Cigogne, Peter, Guy, and
Andrew of Chanceaux, Guy of Cigogne, Geoffrey of Martigny
with his brothers, Philip Mark with his brothers and his nephew
Geoffrey, and the whole brood of the same.
Chapter 45 sought to secure the appointment of suitable
men to posts of trust under the Crown ; the present chapter
definitely excludes from bailiwicks (a comprehensive term
embracing all grades of local magistracies) one particular
group of royal favourites. This clause was omitted from
future reissues, along with chapter 45.
The Charter does not explain the reasons that had
rendered these men obnoxious; but the testimony of con-
temporary Plea Rolls and Pipe Rolls amply supplies the
omission. Each one of them can be shown to have held
places of profit under the Crown as sheriffs of counties,
forest wardens, and commanders of royal garrisons. They
formed a group of kinsmen who, after John had lost his
French dominions, preferred to follow their royal master to
England. The three villages of Athee, Cigogne, and
Chanceaux lie close together in Touraine, in the modern
department of Indre-et-Loire, not far from the cities of
Tours and Loches. The group of men here named all came
^See Rotiili de Finibus, 571. The custody of hostages might be a desirable
office ; in 1199 Alan, the earl's son, offered three greyhounds for the custody of a
hostage of Brittany, Rotuli de Finibus, p. 29.
CHAPTER FIFTY 445
from this district. " They were neither courtiers nor poli-
ticians, but soldiers of experience, whom the barons feared
with good cause." ^
The career of Engelard de Cigogne may be taken as
typical of the rest. He was a nephew of Gerard of Athee,
whom he succeeded, in 1209, as sheriff of Gloucester and
Hereford, an office he held until about the time of Magna
Carta. The Plea Roll of the Gloucestershire Eyre of 1221
covers the period of his shrievalty, and contains a striking
and detailed picture of his misdeeds and extortions.^ He
accounted for the firma burgi of Bristol,^ which seems to
imply interference with its chartered liberties. He also
held pleas of the Crown for Gloucestershire,^ in violation of
the ordinance of 11 94 forbidding any sheriff to act as
justiciar in his own county.^ Several entries tell of barrels
of wine which he took as " prise " from ships entering the
port of Bristol, and thereafter sold to the King. For
example, the exchequer officials allowed him to deduct from
the firma, the sum of 60s., in respect of four tuns of red
wine, as certified by the King's writ.^ Engelard guarded a
rich treasure for the King at Bristol, probably as constable
of the castle there, sums being paid to him ad ponefidum in
thesauro regis J On one occasion he was entrusted with
more than 10,000 marks of the King's money .^ Hostages,
as well as bullion, were placed under his care; a writ dated
1 8th December, 12 14, directed him to liberate three noble
Welshmen whom it mentioned by name.^
In the civil war to which the treaty of peace sealed at
Runnymede was a prelude, Engelard, then constable of
Windsor Castle and warden of the adjacent forest of
Odiham, proved active in John's service. He successfully
defended Windsor from the French faction, making vigor-
^G. J. Turner, Trans. R.H.S., XVIII. p. 254.
^ See Gloucester Pleas^ edited by Maitland, passim.
^ Pipe RoU^ 12 John, cited Madox, I. 333.
^ Pipe Roily 12 John, cited Madox, II. 146. ^ Gloticcster Pleas, xiii. ff.
« Pipe Roll, 12 John, cited Madox, I. 766.
''Ibid., I. 606. ^Ibid., I. 384.
^ Rot. Pat., 16 John, m. 9 (I. 125), and New Rymer, I. 126.
446 ' MAGNA CARTA
ous sorties until relieved by the King.^ He requisitioned
supplies to meet the royal needs ; and a plea was brought
against him so long afterwards as 1232, in connection with
twelve hogsheads of wine thus taken. 2 He acted as sheriff
of Surrey under William Marshal, but was suspended from
this office in 12 18, in consequence of a dispute with Earl
Warenne.^ He remained warden of the castle and forests
for twenty years after the accession of Henry IH.,* and his
long services were rewarded with grants of land : in the
county of Oxford he held the manor of Benzinton, with
four hundreds and a half, during the King's good
pleasure ; ^ while his son Oliver received the lucrative post
of guardian over the lands and heirs of Henry de Berkley.^
In 1 22 1, however, acting in consort with Falkes de
Breaute, Philip Mark, and other castellans, Engelard sup-
ported earl William of Aumale in his resistance to the
demands of Henry's ministers, that all royal castles should
be restored to the King. Notwithstanding the secrecy with
which he sent men to the earl at Biham castle,^ he fell under
suspicion of treason, and found hostages that he would hold
the castle of Windsor for the King.^ In 1236, he was
relieved of some of his offices, but not of all, for in 1254 he
was two years in arrears with the firma of the manor of
Odiham.^ In that year, apparently, he died; for the patent
roll contains a writ granting him permission to make his
will, and an entry in 1255 relates how "for good service
done to the King by Engelard of Cigogne in his lifetime,
the King granted to his executors that they should be quit
of all accounts to be rendered by them at the exchequer,
and of all averages of accounts, and of all debts and
imposts." 10 Engelard thus died, as he had lived, the
1 See M. Paris, II. 665, who calls him *' Ingelardus de Athie " and describes him
as vir in opere martis prohatissimits. Cf. AW. Pat., g Henry III. m. 9.
2 See Bracton's Note-booh, No. 684. 3 gee Rot. Pat., 2 Henry III. m. 7.
^Ibid., 19 Henry III. ^ggg 7^^^^^ ^^ Neville, p. 18, and ibid., p. 120.
• Rot. Pat., 9 Henry III. m. 6. ' R. Wendover, IV. 66.
^ Annals of Dunstable, III. 68.
^Mem. Roll, 28 Henry III., cited Madox, II. 201.
^^ Mick. Cominunia, 29 Henry III., cited Madox, II. 229.
CHAPTER FIFTY 447
trusted servant and favourite of kings. His career illus-
trates how the very same men who had incurred odium as
partisans of John became, when the civil war was over,
instruments of his son's misgovernment.^
CHAPTER FIFTY-ONE.
Et statim post pacis reformacionem amovebimus de regno
omnes alienigenas milites, balistarios, servientes, stipen-
diarios, qui venerint cum equis et armis ad nocumentum
regni.
As soon as peace is restored, we will banish from the kingdom
all foreign-born knights, cross-bowmen, Serjeants, and mercenary
soldiers, who have come with horses and arms to the kingdom's
hurt.
John here binds himself to disband his foreign troops,
the agents of his tyrannies. These men, who had garri-
soned royal castles, are to be banished " as soon as peace
is restored," an indication that a state of virtual war was
recognized. This promise was partially fulfilled : on 23rd
June writs were issued for disbandment of the mercenaries. ^
The renewal of the civil war, however, was followed by
enrolment of new bands of foreigners, whose presence was
one of the main causes of the rebellion of 1224, after the
suppression of which most of them were again banished
with their ringleader, Falkes de Breaute.
The words used to describe these soldiers are compre-
hensive. Stipendiarii embraced mercenaries of every kind :
halistarii were cross-bowmen. This weapon, imported into
England as a result of the crusades, quickly superseded the
^Some particulars respecting the other individuals named will be found in
Thomson, Alagjta Charta, 244-5. Philip Mark was Constable of Nottingham
(R. Wendover, III. 237), and Sheriff both before and after 1215 (see, e.g., Rot.
Claus.j I. 412), while Guy de Chanceaux in 12 14 accounted for scutage of the
honour of Gloucester (Madox, I. 639), and for the rent of the barony of William
of Beauchamp {ibid., I. 717). See also Petit-Dutaillis, Lo7iis VIII., p. ii6;
Gloucester PleaSy passim ; Turner, op. cit. passim.
2 See Rot. Pat., 17 John, m. 23 (A>w Rymer, I. 134).
448 MAGNA CARTA
earlier short bow, but had, in turn, to succumb to the
long bow, which was apparently derived from Wales by-
Edward I., who gained by means of it many battles against
the Scotch and Welsh, and made possible the later triumphs
of the Black Prince and Henry V.
CHAPTER FIFTY-TWO.
Si quis fuerit disseisitus vel elongatus per nos sine legali
judicio parium suorum, de terris, castellis, libertatibus, vel
jure suo, statim ea ei restituemus; et si contentio super
hoc orta fuerit, tunc inde fiat per judicium viginti quinque
baronum, de quibus fit mencio inferius in securitate pacis :
de omnibus autem illis de quibus aliquis disseisitus fuerit
vel elongatus sine legali judicio parium suorum, per Hen-
ricum regem patrem nostrum vel per Ricardum regem
fratrem nostrum, que in manu nostra habemus, vel que
alii tenent que nos oporteat warantizare, respectum
habebimus usque ad communem terminum crucesigna-
torum ; exceptis illis de quibus placitum motum fuit vel
inquisicio facta per preceptum nostrum, ante suscepcionem
crucis nostre : cum autem redierimus de peregrinacione
nostra, vel si forte remanserimus a peregrinacione nostra,
statim inde plenam justiciam exhibebimus.
If any one has been dispossessed or removed ^ by us, without _
the legal judgment of his peers, from his lands, castles, franchises,
or from his right, we wiH immediately restore them to him ; and
if a dispute arise over this, then let it be decided by the five-
and-twenty barons of whom mention is made below in the clause
for securing the peace.^ Moreover, for all those possessions,
from which any one has, without the lawful judgment of his
peers, been disseised or removed, by our father, King Henry,
or by our brother, King Richard, and which we retain in our
hand (or which are possessed by others, to whom we are bound
to warrant them) we shall have respite until the usual term of
crusaders ; excepting those things about which a plea has been
' The elongatus of the Charter replaces the proloiigahis of the Articles.
2 The so-called "executive clause," the '* forma seairitatis adobservandtimpacem^*
of the Articles, which became chapter 6i.
CHAPTER FIFTY-TWO 449
raised, or an inquest made by our order, before our taking of
the cross ; but as soon as we return from our expedition (or if
perchance we desist from the expedition) we will immediately
grant full justice therein.
Chapter 39, in so far as it relates to illegal disseisins,
is here supplemented : remedy is provided for everyone
dispossessed by the Crown " sine legali judicio parium
suorum." Yet, a distinction is drawn between wrongs
inflicted by John himself (where summary methods are to
rule) and by his predecessors (where less precipitate proce-
dure must take its course).
The Articles of the Barons had recognized the same
distinction, while providing somewhat different treatment.
Those disseised by Henry or Richard were to get redress
" according to the judgment of their peers in the King's
court " ; those disseised by John, " according to the judg-
ment of the twenty-five barons." Both cases were, in the
Articles, qualified by a stipulation which calls for comment.
John had taken the crusader's vow a few months before,
and now claimed the usual three years' " respite " from all
legal proceedings. The barons, viewing John's vow as a
notorious perjury, rejected his claim. The Articles referred
the question to arbitration. The prelates, whose judicium
on this point was declared to be final (" appellatione
remota "), and who w^ere bound to give an early decision
(" ad certum diem "), might not unreasonably have been
suspected of partiality, since " taking the cross " was not
a step to be belittled by churchmen. Yet they seem to
have acted in a spirit of not unfair compromise, if the clause
as it finally appeared in John's Magna Carta may be taken
as giving the substance of their award.
In cases where John himself had been the disseisor, the
twenty-five executors might decide forthwith. Respite was
allowed, however, in respect of disseisins of Henry and
Richard (except where legal proceedings were already
pending).^ The Charter says nothing of the procedure at
the close of the three years; but there was probably no
^This "benefit of a crusader" was extended to John in three other sets of com-
plaints, specified in c. 53 {^.v.).
2 F
450 MAGNA CARTA
intention to depart from the terms of the Articles in this
respect, namely, "judgment of peers in the King's court."
John had good reason to consider as unfair the mode
here appointed for deciding disputes as to the other class
of disseisins, namely, those effected by him : many delicate
points would be referred to the summary decision of a
baronial committee, sure to be composed of his most bitter
enemies — the very men, perhaps, who claimed to have been
dispossessed. If the " judgment of the twenty-five " meant
for the barons "the judgment of peers," it meant for the
King the judgment of inferiors and enemies. ^
CHAPTER FIFTY-THREE.
EuNDEM autem respectum habebimus, et eodem modo de
justicia exhibenda de forestis deafforestandis vel reman-
suris forestis, quas Henricus pater noster vel Ricardus
frater noster afforestaverunt, et de custodiis terrarum que
sunt de alieno feodo, cujusmodi custodias hucusque habui-
mus occasione feodi quod aliquis de nobis tenuit per
servicium militare, et de abbaciis que fundate fuerint in
feodo alterius quam nostro, in quibus dominus feodi dixerit
se jus habere; et cum redierimus, vel si remanserimus a
peregrinacione nostra, super hiis conquerentibus plenam
justiciam statim exhibebimus.^
We shall have, moreover, the same respite and in the same
manner in rendering justice concerning the disafforestation or
retention of those forests which Henry our father and Richard
our brother afforested, and concerning the wardship of lands
which are of the fief of another (namely, such wardships as we
have hitherto had by reason of a fief which anyone held of us
by knight's service), and concerning abbeys founded on other
fiefs than our own, in which the lord of the fee claims to have
^ This chapter embraced not merely estates retained in John's possession, but also
those granted out anew. If the former owner recovered these, the Crown was
bound to make good the loss caused by the eviction. The case of Welshmen is
specially treated in c. 56 {^.z>.).
2 The words, *'-?/ eodem modo, de justicia exhibenda,''' and '■^ vel remansuris
forestis'''' are written at the foot of both the Cottonian versions. Cf. supra, 195 n.
They make clear, rather than add to, the meaning of the rest.
CHAPTER Fin Y-THREE 45i
right; and when we have returned, or if we desist from our
expedition, we will immediately grant full justice to all who
complain of such things.
This chapter makes an addition to the Articles of the
Barons, extending to three additional kinds of abuses, the
respite provided in chapter 52 for redressing acts of illegal
disseisin. The " close time " secured to John in virtue of
his crusader's vow is to cover (a) inquiries into boundaries
of forests alleged to have been extended by his father or
his brother; (h) wardships over lands usurped by illegal
extensions of prerogative wardship ; and (c) abbeys founded
by mesne lords but seized by John during vacancies.^
CHAPTER FIFTY-FOUR.
NuLLUS capiatur nee imprisonetur propter appellum femine
de morte alterius quam viri sui.
No one shall be arrested or imprisoned upon the appeal of a
woman, for the death of any other than her husband.
The object of this chapter was to find a remedy for
what the barons evidently considered an unfair advantage
enjoyed by women appellants, who were allowed to appoint
some champion to act for them in the duellum^ while the
accused man had to fight for himself. The connection
between appeal and battle, and the distinction between
battle following on appeal and battle on a writ of right,
have already been explained.^ In civil pleas, neither party
could fight in person : champions were essential, although
hired champions were condemned.-^ In criminal pleas, the
parties must fight in their own persons. This distinction
is not so illogical as it seems at first sight, for the appellant
himself, in the one case, and the champion who fought for
^ It thus supplements three previous chapters {a) c. 47 ; {b) c. yj ; and [c) c. 46
respectively.
^Ct stcpra, c. 36.
^ Bracton, folio 151b, cites the case of a champion sentenced to mutilation of a
foot because he confessed that he was paid to appear. Statute of Westminster, I.
{c. 41), enacted that champions need not swear to personal knowledge. Neilson,
Trial by Combat, 48-51.
452 MAGNA CARTA
him, in the other, were both supposed to be eye-witnesses of
the facts.i
In a case of homicide, no private accuser would be heard
unless he alleged that he had seen the accused actually
do the deed. The stringency of this rule was, however,
modified by legal fictions. The near relation, or the feudal
lord, of the slain man, was treated as constructively present
at his slaying. This, at least, is the most plausible inter-
pretation of Glanvill's words : " No one is admissible to
prove the accusation unless he be allied in blood to the
deceased or be connected with him by the tie of homage
or lordship, so that he can speak of the death upon testi-
mony of his own sight." ^
The rule which required an appellant to offer proof by
his own body was also relaxed in certain cases; women,
men over sixty, and those with broken bones or who had
lost a limb, an ear, a nose, or an eye, might fight by proxy .^
The privilege accorded to women was looked on with dis-
favour : accordingly, the man accused by a woman might,
in Glanvill's words, elect either " to abide by the woman's
proof or to purge himself by the ordeal.""* This option
was freely used; an appellee in 1201 was allowed to go to
the ordeal of water, ^ while two years later when the widow
of a murdered man offered to prove her accusation " as the
court shall consider," the accused "elected to bear the
iron."^ After the virtual abolition of ordeal in 12 15,
appeals by women were usually determined per patriam :
such is the doctrine of Bracton,^ whose authority is borne
out by recorded cases. Thus in 1221, a man accused by a
woman of her husband's murder offered fifteen marks for
a verdict of the jurors.^
^ The appellant " in all cases except murder, that is, secret homicide, made
oath as a witness that he had seen and heard the deed." Neilson, Tj-tal by
Combat^ 48.
2 Glanvill, XIV. c. 3.
3 See Bracton, II. ff., 142b, 145b; also Neilson, Trial by Combat 47, and
authorities there cited.
4 Glanvill, XIV. c. 3. ^ Sel. Pleas of the Crown, No. 1.
*Ibid., No. 68. Cf. No. 119. 'Bracton, folio 142b.
* Select Pleas of the Crown, No. 130.
CHAPTER FIFTY-FOUR 453
A woman's right of accusation (even when thus safe-
guarded from abuse) was restricted to two occasions, the
murder of her husband and the rape of her own person.
Magna Carta mentions only one of these two grounds of
appeal ; but silence on the subject of assault need not be
interpreted as indicating any intention to deprive women
of their rights in such cases. ^
The present chapter of the Great Charter confines itself
to appeals of murder, declaring that no woman has the
right to institute proceedings in this way for the death of
father, son, or friend, but only for that of her husband.
Hard as this rule may seem, the barons here made no
change on existing law. Glanvill does not recognize a
woman's appeal save for the death of her husband r^ —
" A woman is heard in this suit accusing anyone of her
husband's death, if she speak as being an eye-witness to
the fact, because husband and wife are one flesh " — another
example of constructive presence.^
There seems to be no authority for Coke's hasty infer-
ence, that previous to 12 15 a woman had an appeal for the
death of any of her " ancestors " :^ this chapter was purely
declaratory. Yet its provisions were by no means gallant.
The barons were more careful to guard themselves against
risk than to champion the cause of women. ^
^The Act 6 Richard II. c. 6, to prevent the woman's connivance, extended the
right of appeal in such cases to a woman's husband, father, or other near relative ;
but denied the appellee's right to the option of defending himself by battle — thus
proving no exception to the policy of discouraging the duellum wherever possible.
2 Glanvill, XIV. c. 3.
^Fleta I. c. 33 seems to indicate the same doctrine when he speaks '* de morte
viri sui inter brachia sua interfecti^* although laboured explanations are some-
times attempted, e.g. Coke, Second Institute^ 93. Pollock and Maitland (I. 468 n.)
dismiss the phrase inter brachia sua as '* only a picturesque common form."
* See Coke, Second Institute^ p. 68, and contrast Pollock and Maitland, I. 468.
John's justices rejected in 1202 a woman's claim to appeal for her father's deatli,
and some ten years later two claims for the death of sons. See Select Pleas of the
Crown, Nos. 32, 117, and 118; yet Gloucester Pleas (No. 482) records n 1221
a woman's appeal for a sister's death.
^ A peculiarity of wording should, perhaps, be noticed. It restricts explicitly
not appeals, but " arrest and imprisonment" following on appeal.
/
454 MAGNA CARTA
CHAPTER FIFTY-FIVE.
Omnes fines qui injuste et contra legem terre facti sunt
nobiscum, et omnia amerciamenta facta injuste et contra
legem terre, omnino condonentur, vel fiat inde per judicium
viginti quinque baronum de quibus fit mencio inferius in
securitate pacis, vel per judicium majoris partis eorundem,
una cum predicto Stephano Cantuariensi archiepiscopo,
si interesse poterit, et aliis quos secum ad hoc vocare
voluerit : et si interesse non poterit, nichilominus procedat
negocium sine eo, ita quod, si aliquis vel aliqui de predictis
viginti quinque baronibus fuerint in simili querela, amove-
antur quantum ad hoc judicium, et alii loco eorum per
residuos de eisdem viginti quinque, tantum ad hoc facien-
dum electi et jurati substituantur.
All fines made with us unjustly and against the law of the
land, and all amercements imposed unjustly and against the
law of the land, shall be entirely remitted, or else it shall be
done concerning them according to the decision of the five-and-
twenty barons of whom mention is made below in the clause for
securing the peace, or according to the judgment of the majority
of the same, along with the aforesaid Stephen, archbishop of
Canterbury, if he can be present, and such others as he may
wish to bring with him for this purpose, and if he cannot be
present the business shall nevertheless proceed without him,
provided always that if any one or more of the aforesaid five-
and-twenty barons are in a similar suit, they shall be removed
as far as concerns this particular judgment, others being sub-
stituted in their places after having been selected by the rest of
the same five-and-twenty for this purpose only, and after having
been sworn.
The thirty-seventh of the Articles, forming the draft of
this chapter, refers specially to fines exacted by John from
widows for the peaceful enjoyment of their own and their
husband's estates (" pro dotihus, maritagiis, et heredita-
tihus ") : it forms thus a natural supplement to chapter 7.
The earlier chapter had confirmed widows in their rights
for the future; this one remits fines unjustly taken in the
past. It is probable that the Articles of the Barons did
not intend to limit their own operation to this one group
CHAPTER FIFTY-FIVE 455
of unjust fines; and they mention amercements without
qualification. In any view, the terms of Magna Carta were
broadened out to embrace illegal fines and amercements of
every sort.^
The distinction between fines and amercements has been
explained in a former chapter. ^ The system of arbitrary
fines culminated in the reign of John, whose talents were
well suited to the development of its ingenious and mean
details. Dr. Stubbs describes the product of John's labours
as " the system of fines which was elaborated into that
minute and grotesque instrument of torture which all the
historians of the reign have dwelt on in great detail,"^ and
Hallam has a passage which has become classical : — " The
Bishop of Winchester paid a tun of good wine for not
reminding the King (John) to give a girdle to the countess
of Albemarle ; and Robert de Vaux five best palfreys, that
the same King might hold his peace about Henry Pinel's
wife. Another paid four marks for leave to eat (pro licentia
comedendi),"'^
Unique procedure was provided by the present chapter
for deciding disputes as to the legality of fines and
amercements. Authority to decide was vested in a board
of arbitrators to consist of thirteen or more of the twenty-
five executors, together with Stephen Langton and such
others as he chose to summon. No mention is made of
the maximum number whom the primate might nominate,
and there is no attempt to define their powers relative to
the other members, a somewhat unbusinesslike omission,
but one which testifies to the confidence placed in Langton
by those who approved its terms. Care is taken to prevent
members of the twenty-five from sitting in judgment on
suits arising from circumstances resembling their own.
^In its expanded form the clause becomes a supplement also to cc. 20, 21, and
22 (which defined procedure at amercements), and to cc. 36 and 40 (which con-
demned John's practice of refusing writs and justice until heavy fines were offered).
^ See supra, c. 20. ^ See Preface to W. Coventry, II. Ixix.
* Middle Ages, II. 438. Hallam's examples are all drawn from Madox, I. 507-9.
Other illustrations of fines and amercements may be found under several of the
foregoing chapters. Every man who began a plea and lost it, or abandoned it,
was amerced.
456 MAGNA CARTA
This chapter, like others addressed to special needs of
John's reign, found no echo in future charters.
CHAPTER FIFTY-SIX.
Si nos disseisivimus vel elongavimus Walenses de terris
vel libertatibus vel rebus aliis, sine legali judicio parium
suorum, in Anglia vel in Wallia,^ eis statim reddantur;
et si contencio super hoc orta fuerit, tunc inde fiat in
marchia per judicium parium suorum, de tenementis
Anglie secundum legem Anglie, de tenementis Wallie
secundum legem Wallie, de tenementis marchie secundum
legem marchie. Idem facient Walenses nobis et nostris.
If we have disseised or removed Welshmen from lands or
liberties, or other things, without the legal judgment of their
peers in England or in Wales, they shall be immediately restored
to them ; and if a dispute arise over this, then let it be decided
in the marches by the judgment of their peers ; for tenements
in England according to the law of England, for tenements in
Wales according to the law of Wales, and for tenements in the
marches according to the law of the marches. Welshmen shall
do the same to us and ours.
Three chapters, redressing wrongs suffered by Welsh-
men, testify to the importance attached by the barons to
the Welsh alliance. Restoration is to be made (a) of illegal
disseisins effected by John (chapter 56) ; (b) of those effected
by Henry II. and Richard I. (chapter 57); and (c) of host-
ages and charters delivered to John as pledges of peace
(chapter 58).
This chapter does for Welshmen dispossessed by John
what chapter 52 did for Englishmen, but substitutes "m
marchia per judicium parium suorum " for the " per judi-
cium mginti quinque haronum " of the earlier chapter.
The "venue" was thus fixed in the marchland for all
Welshmen's cases, although different kinds of law were to
be applied according to the situation of the property in
*The words *' in Anglia vel in Wallia^^ are written at the foot of one of the
Cottonian versions (cf. sttpi'a, i66 n.) ; but they appear in sitti in the Articles of
the Barons.
CHAPTER FIFTY-SIX 457
dispute. This indication of the existence of three distinct
bodies of law, one for England, another for Wales, and a
third for the marches, shows that the unifying tasli of the
common law had not yet been completed. Interesting
questions of a nature analogous to those treated by the
branch of modern jurisprudence known as International
Private Law must constantly have arisen.
All three classes of alleged disseisins (whatever the law
involved) were to be decided by a judicium parium; but
the " peers " of a Welshman were not defined — a vital
CHAPTER FIFTY-SEVEN.
De omnibus autem illis de quibus aliquis Walensium
disseisitus fuerit vel elongatus sine legali judicio parium
suorum per Henricum regem patrem nostrum vel Ricar-
dum regem fratrem nostrum, que nos in manu nostra
habemus, vel que alii tenent que nos oporteat warantiz-
are, respectum habebimus usque ad communem terminum
crucesignatorum, illis exceptis de quibus placitum motum
fuit vel inquisicio facta per preceptum nostrum ante
suscepcionem crucis nostre : cum autem redierimus, vel
si forte remanserimus a peregrinacione nostra, statim eis
inde plenam justiciam exhibebimus, secundum leges Wal-
ensium et partes predictas.
Further, for all those possessions from which any Welshman
has, without the lawful judgment of his peers, been disseised
or removed by King Henry our father, or King Richard our
brother, and which we retain in our hand (or which are possessed
by others, to whom we are bound to warrant them) we shall
have respite until the usual term of crusaders ; excepting those
things about which a plea has been raised or an inquest made
by our order before we took the cross ; but as soon as we return,
(or if perchance we desist from our expedition), we will im-
mediately grant full justice in accordance with the laws of the
Welsh and in relation to the foresaid regions.
The provisions for Welshmen unjustly dispossessed by
Henry or Richard are identical with those made in the
^Cf. Harcourt's comment, "A bad piece of work this" {Steward, 220).
458 MAGNA CARTA
latter part of chapter 52 for Englishmen, except for the last
words, " in accordance with the laws of the Welsh in rela-
tion to the foresaid districts " : no machinery is here
specified for declaring or applying these laws.
The Articles of the Barons had, however, mentioned the
procedure to be adopted; and a comparison of articles 25
and 44 with this chapter suggests the antithesis between
" per judicium parium suorum in curia regis " for English-
men, and " in marchia per judicium parium suorum " for
Welshmen.
CHAPTER FIFTY-EIGHT.
Nos reddemus filium Lewelini statim, et omnes obsides
de Wallia, et cartas que nobis liberate fuerunt in securi-
tatem pacis.
We will immediately give up the son of Llywelyn and all the
hostages of Wales, and the charters delivered to us as security
for the peace.
The treatment of hostages in general and Welsh host-
ages in particular has already been illustrated.^ The
patent and close rolls show a constant coming and going
of these living pledges of the peace. A writ of i8th
December, 12 14, for example, bade Engelard of Cigogne
restore three Welsh nobles to Llywelyn.^ Since then, new
hostages, including Llywelyn's son, had been handed over;
and charters also had been pledged.
The Articles of the Barons had treated this question as
an open one, referring it to the arbitration of Stephen
Langton and others he might nominate. The point had
apparently been decided in favour of the Welsh before the
Charter was engrossed in its final form.^ John is now
1 See suj>ra, p. 441. ^See supra, p. 445.
3 No. 45 of the Articles is connected by a rude bracket with No. 46 (relating to
Scotland) ; and a saving clause, thus made applicable to both, is added with some
appearance of haste : *' nisi aliter esse debeatper cartas quas rex habet^ per judicium
archiepiscopi et aliorum quos sectcm vocare vohierit.^^ Cf. supra^ p. 38. So far as
related to Scotch affairs, the King's caveat found its way, in an altered form, into
Magna Carta. See c. 59.
CHAPTER FIFTY-EIGHT 459
made to promise an immediate surrender of hostages and
charters.
The Welsh prince must have breathed more freely when
this was fulfilled. Soon, with a light heart, his son by
his side, he renewed hostilities. Gualo, on nth November,
1 2 16, laid interdict on the whole of Wales for holding with
the barons.^ By the treaty of Lambeth, Louis was to send
a copy of the peace to Llywelyn and the other Welsh
princes.2
CHAPTER FIFTY-NINE.
Nos faciemus Alexandro regi Scottorum de sororibus suis,
et obsidibus reddendis, et libertatibus suis, et jure suo,
secundum formam in qua faciemus aliis baronibus nostris
Anglie, nisi aliter esse debeat per cartas quas habemus
de Willelmo patre ipsius, quondam rege Scottorum ; et hoc
erit per judicium parium suorum in curia nostra.
We will do towards Alexander, King of Scots, concerning the
return of his sisters and his hostages, and concerning his
franchises, and his right, in the same manner as we shall do to-
wards our other barons of England, unless it ought to be other-
wise according to the charters which we hold from William his
father, formerly King of Scots ; and this shall be according to
the judgment of his peers in our court.
The barons w^elcomed allies whether from Whales or
Scotland; and this chapter was dictated by a desire to
conciliate Alexander. John was forced to promise to
restore to the King of Scots his sisters and other hostages,
together with his franchises and his "right."
Opinions have been, and still are, sharply divided as
to whether, or in what degree, Scotland was subject to
feudal overlordship. David I. and his successors. Kings
of Scotland, had been wont to do fealty and homage to the
Kings of England; but this fact has received different
interpretations. Such homage, it is argued, was performed
in respect of certain English baronies which happened to
belong by hereditary right to the Kings of Scotland,
* Annals of Waverley, sub anno 1216. "^ New Rymer^ I. 148.
46o MAGNA CARTA
namely, the earldom of Huntingdon, and the counties of
Northumberland, Cumberland, and Westmoreland. The
terms of homage did not indicate for what fiefs it was
sworn — whether for the English earldoms alone, or for the
country north of Tweed as well.
The position of the Kings of Scots remained ambiguous,
until William the Lion was placed at a terrible disadvantage
by his capture at Alnwick in 1174. To gain release, he
ratified the Treaty of Falaise on 8th December, of that year,
by which he agreed to hold his territories as fiefs of the
English Crown. All his tenants in Scotland were to take
oath to Henry ; while hostages were surrendered, along with
the castles of Berwick, Roxburgh, Jedburgh, Edinburgh,
and Stirling.!
Henry's diplomacy was undone by his successor.
Richard, preparing for his crusade of 1190, sold recklessly
every right that would fetch a price : William bought back
the independence of his kingdom ; but this restoration of
the relations that prevailed previous to 11 74, involved a
restoration of the old ambiguities. When Richard died,
William despatched ambassadors to England, pressing
claims upon the northern counties, promising to support
John's title in return for their admission, and adding
threats.2 John avoided committing himself until his posi-
tion in England was assured; thereafter he commanded
William to do homage unconditionally. The Scots King
disregarded the first summons, but yielded to a second,
yet " reserving always his own right." ^ The saving clause
left everything vague as before.
In April, 1209, the King of Scots incurred John's dis-
pleasure. William's only son, Alexander, was demanded
as a hostage, or alternatively three border castles must be
delivered up. After a refusal, the old King gave in on
^ See Ramsay, Angevin Empire^ 183-4. In the spring of 1 185 Henry confirmed
William's claim to Huntingdon, and the Scots King transferred it to his brother
David ; ibid., 226 n.
2 See Miss Norgate,y<?//« Lackland, 66.
^See Stubbs, Const. Hist., I. 596 n., and Norgate, yb/^« Lackland, 73, 78. Cf.
the words '■^ salvo jure suo" with the ^^etjure suo" of Magna Carta.
CHAPTER FIFTY-NINE 461
7th August, 1209.1 Alexander did homage on behalf of
his father " for the aforesaid castles and other lands which
he held," and found sureties for the payment of 15,000
marks. William's daughters, Margaret and Isabel, became
wards of John, who had the right to bestow them in
marriage.2 There seems to have been an understanding
that one of them should wed John's eldest son.^ Margaret
and Isabel, though virtually prisoners in Corfe Castle,
were honourably treated there. The Close Rolls contain
orders for supplying them with articles of comfort and
luxury. Thus on 6th July, 12 13, John instructed the Mayor
of Winchester to despatch in haste, for the use of his niece
Eleanor and of the two Scots princesses, robes of dark
green (tunics and super-tunics) with capes of cambric and
fur of miniver, together with twenty-three yards of good
linen cloth, with light shoes for summer wear, " and the
Mayor is to come himself with all the above articles to
Corfe, there to receive the money for the cost of the same." ^
Meanwhile, events in Scotland had favoured English
pretensions. In 1212, Cuthred, a claimant for the Scottish
throne, endeavoured to dethrone King William. English
succour was asked and paid for by a treaty sealed at Norham
on 7th February, 1212, by which William granted to John
the right to marry the young Alexander, then fourteen years
of age, *^ sicut hominem suuin ligiuniy^^ to whomsoever he
would, at any time within the next six years, but always
" without disparagement." ^ William pledged himself and
his son to keep faith and allegiance to John's son, Henry,
" as their liege lord " against all mortals.^ William had
saved his Crown, but Scotland was sinking into the position
^ New Rymer^ I. 103, where " Northampton " is apparently a mistake for
'* Norham." See Ramsay, Angevin Empire, 421 n.
"Ramsay, ibid.^ and authorities there cited.
•^Ramsay, Angevin Empire, 421, and authorities.
* Rot. Clatis., I. 144, and I. 157. This Eleanor was the sister of Prince Arthur.
The fortunes of war had in 1202 placed both of them in John's hands. Arthur dis-
appeared— murdered it was supposed ; Eleanor remained a prisoner for life ; the
Scots princesses were virtually her fellow-prisoners for a time in Corfe Castle.
^ See supra, c. 6.
^ Nezv Fymer, I. 104. See also W. Coventry, II. 206.
462 MAGNA CARTA
of a vassal state. On 28th October, 12 13, Innocent Ill-
ordered the King of Scotland and his son to show fealty
and devotion to King John.^
William the Lion died at Stirling on 4th December,
1 2 14, and Alexander's peaceful succession was facilitated
by the knowledge that he had the support of John. Such
was the position of affairs when John was brought to bay
at Runnymede. The barons were willing to bid for the
alliance of Alexander; yet it was unnecessary to bid high.
John was made to promise to restore Alexander's sisters
and other hostages unconditionally, but words were used
which committed him on none of the disputed points.^
Franchises and " right " were to be restored only in so far
as accorded with William's "charters," as interpreted by
the judgment of the English barons in the court of the
English King.
The allusion in the text to the Scottish King as one among
" our other barons of England " need not be pressed against
Alexander, any more than similar expressions should be
pressed against John, whose position as Duke of Normandy
and Aquitaine in no way made England a fief of the French
Crown or prevented him becoming a vassal of Rome. In
questions affecting his feudal position in France, John's
peers were the dukes and counts of that country; and
similarly those who had a right to sit in judgment as
Alexander's peers over his claims to English fiefs were the
English earls and barons. Such a tribunal was not likely
to give decisions favourable to Scots pretensions, at the
expense of England.^
Alexander, though no party to the treaty at Runnymede,
^See New Ry/ner, I. ii6.
2 Both ladies, however, remained prisoners after Henry III.'s accession. Peter
de Maulay, constable of Corfe Castle, was, in that King's fifth year, credited with
sums expended on their behalf. Rof. Claus., I. 466; see also I. 483. Both
found permanent homes in England — Margaret as wife of Hubert de Burgh, Isabel
as wife of Roger Bigod. See Ramsay, Angevin Empire, 421, and authorities
there cited.
^ No. 46 of the Articles referred the question of Alexander's "right" to the
judgment of Langton and his nominees, for which Magna Carta substituted
"judgment of his peers in our court."
CHAPTER FIFTY-NINE 463
was willing to profit by it : on 7th July, 12 15, he despatched
the Bishop of St. Andrews and five laymen to John " con-
cerning our business which we have against you to be
transacted in your court." ^ Nothing came of this; and
Alexander invaded England in order to push his claims.
John swore his usual oath, " by God's teeth," that he would
" chase the little red-haired fox-cub from his hiding-holes." ^
By the treaty of Lambeth (12th September, 12 17), Louis
and Henry were each to send a copy of the peace to Alex-
ander that he might be included in its terms on his restoring
castles, lands, and prisoners, taken by him in the war.^^
On 23rd September, they joined in urging him to restore
Carlisle, and Alexander, anxious to preserve his English
honour of Huntingdon, was constrained to yield.* The
deeper question at issue between England and Scotland was
still unsolved when the relations between the two countries
entered on a new phase, as a consequence of the attempts
at annexation made by Edward I., " the hammer of the
Scots."
CHAPTER SIXTY.
Omnes autem istas consuetudines predictas et libertates
quas nos concessimus in regno nostro tenendas quantum ad
nos pertinet erga nostros, omnes de regno nostro, tam clerici
quam laici, observent quantum ad se pertinet erga suos.
Moreover, all these aforesaid customs and liberties, the
observance of which we have granted in our kingdom as far
as pertains to us towards our men, shall be observed by all of
our kingdom, as well clergy as laymen, as far as pertains to them
towards their men.
It would have been as impolitic as it was obviously unfair
for the barons, in their capacity of mesne lords, to inflict
upon their own tenants those very exactions which they
compelled the King to abjure as against themselves.
Accordingly, the benefit of the " customs and liberties "
^ Neiv Rymevy I. 135. ^ M. Paris, II. 642.
^ New Rymer^ I. 148. ^ Rot. Pat., I. 93.
464 MAGNA CARTA
conceded by John to his feudal tenants was — in a somewhat
perfunctory manner, it is true — extended to the feudal
tenants of all other magnates, whether cleric or lay.
Although the reference to " customs and liberties " was
quite general in its terms, it seems natural to infer that
feudal grievances were chiefly meant, since the view of
society indicated is feudal rather than national.^
These considerations suggest that too liberal a view has
sometimes been taken of the scope of this chapter. Coke
treated it as affecting not merely freeholders, but the whole
mass of the people : — " This is the chief felicity of a king-
dom, when good laws are reciprocally of prince and people
(as is here undertaken) duly observed." ^ In this view, he
has had many followers; and the present chapter has
received undue emphasis as supporting a democratic inter-
pretation of Magna Carta.^ It has been referred to as
" the only clause which affects the whole body of the
people."^ The better view is that its provisions were con-
fined to feudal sub-tenants.
Even authors who interpret the chapter in this restricted
application are still prone to exaggerate its importance,
(i) The clause is sometimes regarded as springing directly
from the barons' own initiative : Dr. Stubbs, contrasting it
with Henry I.'s Charter of Liberties, holds that it was
"adopted by the lords themselves."^ Such praise is
unmerited ; the barons inserted it because they had need of
allies. (2) On the other hand, credit for the clause, equally
unwarranted, has been sometimes bestowed on John. Dr.
Robert Henry says that " this article, which was highly
reasonable, was probably inserted at the desire of the
King." 6
The substance of this chapter appears in the reissues of
1 Harcourt, Steward^ 221, treats this chapter as extending to manorial courts the
principles regulating Xht judicium parium and amercements.
^ Second Institute, 77. ^Qi. mpra, p. 113.
* Thomson, Magna Carta, 269, and authorities there cited.
^ Const. Hist., I. 570. Cf. supra, p. 117.
^History of Great Britain, VI. 74 (1823). See also Henshall, History 0/ South
Britain, cited Thomson, Magna Carta, 268-9.
CHAPTER SIXTY 465
12 17 and 1225; but its force there is possibly somewhat
impaired by the addition of a new clause reserving' to arch-
bishops, bishops, abbots, priors, templars, hospitallers,
earls, barons, and all other persons as well ecclesiastical as
secular, all the franchises and free customs they previously
had 1 — a" saving clause " that might be turned to various
uses.
CHAPTER SIXTY-ONE.
Cum autem pro Deo, et ad emendacionem regni nostri,
et ad melius sopiendam discordiam inter nos et barones
nostros ortam, hec omnia predicta concesserimus, volentes
ea Integra et firma stabilitate in perpetuum^ gaudere,
facimus et concedimus eis securitatem subscriptam ; vide-
licet quod barones eligant viginti quinque barones de regno
quos voluerint, qui debeant pro totis viribus suis observare,
tenere, et facere observari, pacem et libertates quas eis con-
cessimus, et hac presenti carta nostra confirmavimus, ita
scilicet quod, si nos, vel justiciarius noster, vel ballivi
nostri, vel aliquis de ministris nostris, in aliquo erga
aliquem deliquerimus, vel aliquem articulorum pacis aut
securitatis transgressi fuerimus, et delictum ostensum fuerit
quatuor baronibus de predictis viginti quinque baronibus,
illi quatuor barones accedant ad nos vel ad justiciarum
nostrum, si fuerimus extra regnum, proponentes nobis
excessum, petent ut excessum ilium sine dilacione faciamus
'Cmendari. Et si nos excessum non emendaverimus, vel,
si fuerimus extra regnum justiciarius noster non emen-
daverit, infra tempus quadraginta dierum computandum
a tempore quo monstratum fuerit nobis vel justiciario nostro
si extra regnum fuerimus, predicti quatuor barones referant
causam illam ad residuos de illis viginti quinque baronibus,
et illi viginti quinque barones cum communa tocius terre
distringent et gravabunt nos modis omnibus quibus pote-
^See c. 46 of 1217.
2 The words "m perpetiium'''' are written at the foot of one of the Cottonian
•versions. See supj'a, 1 66 n.
2G
466 MAGNA CARTA
runt, scilicet per capcionem castrorum, terrarum, posses-
sionum, et aliis modis quibus poterunt, donee fuerit
emendatum secundum arbitrium eorum, salva persona
nostra et regine nostre et liberorum nostrorum; et cum
fuerit emendatum intendent nobis sicut prius fecerunt.
Et quicumque voluerit de terra juret quod ad predicta
omnia exequenda parebit mandatis predictorum viginti
quinque baronum, et quod gravabit nos pro posse suo cum
ipsis, et nos publice et libere damus licenciam jurandi
cuilibet qui jurare voluerit, et nulli umquam jurare pro-
hibebimus. Omnes autem illos de terra qui per se et
sponte sua noluerint jurare viginti quinque baronibus, de
distringendo et gravando nos cum eis, faciemus jurare
eosdem de mandato nostro, sicut predictum est. Et si
aliquis de viginti quinque baronibus decesserit, vel a terra
recesserit, vel aliquo alio modo impeditus fuerit, quominus
ista predicta possent exequi, qui residui fuerint de pre-
dictis viginti quinque baronibus eligant alium loco ipsius,
pro arbitrio suo, qui simili modo erit juratus quo et ceteri.
In omnibus autem que istis viginti quinque baronibus
committuntur exequenda, si forte ipsi viginti quinque
presentes fuerint, et inter se super re aliqua discordaverint,
vel aliqui ex eis summoniti nolint vel nequeant interesse,
ratum habeatur et firmum quod major pars eorum qui
presentes fuerint providerit, vel preceperit, ac si omnes
viginti quinque in hoc consensissent ; et predict! viginti
quinque jurent quod omnia antedicta fideliter observabunt,
et pro toto posse suo facient observari. Et nos nichil
impetrabimus ab aliquo, per nos nee per alium, per quod
aliqua istarum concessionum et libertatum revocetur vel
minuatur; et, si aliquid tale impetratum fuerit, irritum sit
et inane et numquam eo utemur per nos nee per alium.
Since, moreover, for God and the amendment of our kingdom
and for the better allaying of the quarrel that has arisen between
us and our barons, we have granted all these concessions, de-
sirous that they should enjoy them in complete and firm endur-
ance for ever, we give and grant to them the under-written security,
namely, that the barons cJ^qqss five-and-twenty barons of the
kingdom, whomsoever they will, who shall be bound with all their
might, to observe and hold, and cause to be observed, the peace
CHAPTER SIXTY-ONE 467
and liberties we have granted and confirmed to them by this ou
present Charter, so that if we, or our justiciar, or our bailiffs or
any one of our officers, shall in anything'be at fault towaTd any-
one, or shall have^broken any one of the articles of the peace or
of this security, and the offence be notified to four barons of the
foresaid five-and-twenty, the said four barons shall repair to us (or
our justiciar, if we are out of the realm) and, laying the trans-
gression before us, petition to have that transgression redressed
without delay. And if we shall not have corrected the trans-
gression (or, in the event of our being out of the realm, if our
justiciar shall not have corrected it) within forty days, reckoning
from the time it has been intimated to us (or to our justiciar,
if we should be out of the realm), the four barons aforesaid
shall refer that matter to the rest of the five-and-twenty barons,
and those five-and-twenty barons shall, together with the com-
munity of the whole land, distrain and distress us in all pos-
sible ways, namely, by seizing our castles, lands, possessions,
and in any other way they can, until redress has been obtained
as they deem fit, saving harmless our own person, and the persons
of our queen and children ; and when redress has been obtained,
they shall resume their old relations towards us. And let whoever
in the country desires it, swear to obey the orders of the said five-
and-twenty barons for the execution of all the aforesaid matters,
and along with them, to molest us to the utmost of his power ; and
we publicly and freely grant leave to every one who wishes to swear,
and we shall never forbid anyone to swear. All those, moreover,
in the land who of themselves and of their own accord are un-
willing to swear to the twenty-five to help them in constraining and
molesting us, we shall by our command compel the same to swear
to the effect foresaid. And if any one of the five-and-twenty barons
shall have died or departed from the land, or be incapacitated in
any other manner which would prevent the foresaid provisions
being carried out, those of the said twenty-five barons who are left
shall choose another in his place according to their own judgment,
and he shall be sworn in the same way as the others. Further,
in all matters, the execution of which is intrusted to these twenty-
five barons, if perchance these twenty-five are present and dis-
agree about anything, or if some of them, after being summoned,
are unwilling or unable to be present, that which the majority of
those present ordain or command shall be held as fixed and
established, exactly as if the whole twenty-five had concurred in
this ; and the said twenty-five shall swear that they will faithfully
observe all that is aforesaid, and cause it to be observed with all
their might. And we shall procure nothing from anyone, directly
or indirectly, whereby any part of these concessions and liberties
might be revoked or diminished ; and if any such thing has been
procured, let it be void and null, and we shall never use it
personally or by another.
468 MAGNA CARTA
This important chapter stands by itself, providing
machinery for enforcing all that precedes it. It thus
forms what modern jurisprudence would describe as the
" sanction " of the whole, but what was known in the current
phrase of its own day as " the form of security " {forma
securitatis ad observandum pacem et lihertates).^ It con-
tains the only executive clause of the Charter, the sole
constitutional machinery .^
I. The ''Security'' or legal Sanction. The procedure
devised for enforcing the Charter was crude : John con-
ferred upon twenty-five of his enemies a legal right to
organize rebellion, whenever in their opinion he had broken
any one of the provisions of Magna Carta. Violence
might be legally used against him, until he redressed their
alleged grievances " to their own satisfaction " {secundum
arhitrium eoriirn). If it had been possible to put so violent
an expedient into practice, the "sovereignty," or supreme
power in England, would have been split into two. John
would have held the sceptre only until his opponents
declared that he had broken the Charter, when, by his
own previously-granted mandate, it would pass to the
twenty-five barons forming what has been variously styled
a " Committee of Remonstrance and Constraint " or a
*' Committee of Rebellion." ^
The procedure for redressing grievances is described in
some detail ; the wronged party must make known his case
to four barons of the twenty-five, who would then make it
known to the King, and ask redress. If John refused or
^ This phrase occurs in the 49th (and last) of the Articles, as the title of a clause
separated from the others by a blank of the width of several lines of writing :
'■^ Haec est forma securitatis^'''' etc. The words are not used as a heading in the
present chapter itself, but c. 52 refers to c. 61 as the clause **z*« securitate pads"
and c. 62 refers to it as ^* super securitate ista."
* Histoire des dues, 150, has a commentary on this chapter : ** Over and above
all this they desired that 25 barons should be chosen, and by the judgment of these
25 the King should govern them in all things, and through them redress all the
wrongs that he should do to them, and they also, on the other hand, would
through them redress all the wrongs that they should do to him. Also they further
desired, along with all this, that the King should never have power to appoint a
bailiff in his land except through the 25." Cf. supra, p. 123 and p. 177.
'Cf. S. R. Gardiner, Short History of England, 183 : "a permanent organiza-
tion for making war against the King."
CHAPTER SIXTY-ONE 469
unduly delayed, compulsion might be used. On the matter
of undue delay, the Articles of the Barons said " within a
reasonable time to be determined in the Charter." The
Charter did determine this, naming forty days. Compul-
sion might take any form, except violence against the
person of the King, or of his wife or children.
II. Minor details. Although the whole expedient seems
chimerical to the modern mind, the opposition leaders in
1 2 15 evidently thought they had devised a practicable
scheme of government. This is shown by the care with
which they elaborated the procedure.
(i) Appointment of the twenty-five executors. The
members of the committee were to be, in the first instance,
" elected " by the " barons." Vacancies were to be filled
by the method now known as " co-optation " : the com-
mittee, once appointed, would form a close corporation ;
no one uncongenial to the majority could gain admission
— an arrangement with a thoroughly oligarchic flavour.
The provision for supplying vacancies caused by death
proves that the scheme was not to be temporary.
Writs, issued to the sheriffs on 19th June, command the
enforcement of the oath to the twenty-five barons, but do
not mention them by name. Matthew Paris supplies the
omission, and though he does not disclose the source of
his information, it is unlikely that so comprehensive a list
could be entirely a work of the imagination. ^ They occur
in the following order, the earls of Hertford, Aumale,
Gloucester, Winchester, Hereford, Norfolk, and Oxford,
William Marshall the younger, Robert fitz Walter the
elder, Gilbert de Clare, Eustace de Vesci, Hugh Bigod,
William of Mowbray, William Hardell (Mayor of London),
William de Lanvalei, Robert de Ros, John de Lacy
(Constable of Chester), Richard de Perci, John fitz Robert,
William Mallet, GeoftVey de Say, Roger de Mumbezon,
William of Huntingfield, Richard de Muntfitchet, and
William of Albini.^ There are here no churchmen and
1 R. Wendover, from whom Paris borrows so freely, gives no list.
2 The list is from Matthew Paris, II. 604-5, as corrected by Blackstone, Great
Charter, p. xx, after collation with a marginal note on the Harleian MS. of the
470 MAGNA CARTA
no members of the moderate party whose names appear in
the preamble. All except two, or at most three, were
declared enemies of John.^ It was an oligarchy of dis-
affected Crown tenants, whose baronial homogeneity was
only broken by the presence of the Mayor of London.
Such a committee was not likely to use its powers to further
other interests than its own.
(2) A majority to form a quorum. Driven by necessity
the barons devised, or stumbled upon, a peculiarly modern
expedient. Unanimity would be difficult to obtain. It was
provided, accordingly, that the will of the majority of those
present should prevail. It would be inaccurate to say, in
modern phraseology, that thirteen formed a quorum, since
the quorum varied with the number of those present. No
provision was made for summoning or constituting this
committee, and room was thus left for packed meetings :
one faction, hurriedly convened, might usurp the rights of
the whole body. The precedent tentatively introduced, for
allowing a majority to act for the whole, was followed
only timidly and at intervals. Still, its appearance in
John's Charter marks a stage in the advance of the principle
of modern politics which substitutes the " counting of heads
for the breaking of them." 2
(3) The suh-committee of four. Four of the twenty-five
executors were to act as intermediaries between aggrieved
individuals and the King. Such a position involved dis-
cretionary powers; for, if the four refused to endorse the
justice of any complaint, John also would be in safety to
refuse. 3
charter (cf. supra, p. i68 n.). For biographical information, see Thomson, Magna
Carta, 270-312.
1 These three were Earl Aumale (a title sometimes exchanged for Earl of York,
see Round, Geoffrey de Mandeville, 157 n.), William of Albini, and Geoffrey de
Say (see Stubbs, Const. Hist., I. 583).
2 This is not the earliest reference in English law to the binding power of a
majority; Liebermann, Gesetze, II. 575, points to Leges Heririci, c. 5, s. 6 {.ibid.,
I. 549) as formulating the principle.
' An alternative explanation is possible, namely, that the function of intermediary
might be exercised by any four of the twenty-five. In that view, an aggrieved
individual might place pressure on the King if he persuaded any four to support his
claim.
CHAPTER SIXTY-ONE 47i
(4) Local agents of the twenty-five executors. In each
county the twelve knights, whose original function was to
preside at inquiries into "evil customs," came to act as
local representatives of the revolutionary committee, being
armed with power to constrain the sheriff to carry out the
provisions of Magna Carta, very much as the twenty-five
were authorized to constrain the King. In particular, these
knights were charged with enforcement of the oath of
obedience to the revolutionary committee, and with con-
fiscating the property of all who refused.^
(5) The part to he played by the public. John authorized
his subjects to side against him, if he should violate the
Charter : his general mandate was granted to the twenty-five
" cum communa totius terrce," while licence was " freely and
publicly " bestowed on everyone so disposed, to swear obedi-
ence to the executors. Two aspects of this provision require
attention : (a) Its relation to allegiance and treason. John
solemnly authorized his subjects, in certain circumstances,
to transfer their allegiance from himself to the committee of
his foes. If they refused, he agreed to their compulsion ;
and on 27th June, 1215, writs were actually issued instruct-
ing the seizure of the lands and goods of all who would not
swear to obey the twenty-five.^ (b) Communa totius terrce.
The " community of the whole land " was thus to afford
active help in subjecting the King to the reign of law ; and
the phrase has been pressed into the service of democracy
by enthusiasts, who seek to magnify modern conceptions by
finding their roots in the past. Few words of medieval
Latin offer a more tempting field to enquirers than this
communa, which, with its English and French equivalents,
holds the key to many problems of constitutional origins.
The appearance in Magna Carta of a body described as a
" commune," in conjunction with an oath of obedience to a
revolutionary committee, suggests comparison with the
form of civic constitution known in that age as " the sworn
commune."^ The "communa" referred to in chapter 61
1 Cf. supra, c. 48. 2 See Appendix-.
^ It was fourteen years since London had extorted its ** commune," in this sense,
from Prince John ; cf. supra, c. 13.
472 MAGNA CARTA
was something widely different : to the barons at Runny-
mede it may have meant either the entire body of feudal
tenants or only the magnates ; but medieval analogies make
it impossible that the word could embrace the free peasantry,
still less the villeins of England. The occurrence of such
a word is far from proving that the Charter rests on any
broad or popular basis.
III. Relations to Contemporary Theory. Clumsy and
impracticable as the whole scheme appears to modern eyes,
it was quite in accord with medieval theory. The concep-
tion of a relation founded upon contract between lord and
vassal lies at the root of feudalism. If either party glaringly
broke the terms of the compact, the other was justified in
repudiating the relationship, but he must observe due
formalities. Diffidatio, intimated to his lord, must precede
any attempt of the vassal to redress his wrongs by force.
The barons at Runnymede, having complied with this pre-
liminary, had for the moment ceased to owe fealty to John.
In reserving power to appoint an Executive Committee
(even if this be regarded as implying a right of legalized
rebellion), as a condition precedent to a renewal of allegi^
ance, they moved in the direction of legal restraint as
opposed to revolutionary violence. The right here recog-
nized by John, likely as it might be to lead to hostilities,
was in theory and intention an honest effort to obviate war
by recourse to the nearest approximation to constitutional
action then available. It was, further, an attempt to sub-
stitute united action of the body of feudal tenants (com-
muna totius terrce) for the individual vassal's right of
private judgment, claimed and sometimes exercised in that
age, on the European continent, and actually confirmed in
1222 by Andreas II. of Hungary by his Bulla Aurea.^
The expedient contained in this chapter is a logical deduc-
tion from the vassal's right of defiance as a prelude to
private war against a lord who has wronged him. It was
no innovation, but something found by the barons in feudal
law.2 Foreign parallels have been found for it, not only in
^Hantos, Magna Caria, 149, 198. Cf. Gneist, Eng. Const., 251.
2 Adams, Origin, 181 ff.
CHAPTER SIXTY-ONE 473
the more anarchic procedure of the Hungarian Bulla Aurea,
but also in the institutions of Aragon and elsewhere. ^
When the baronial leaders in 1263 performed diffidalio, they
echoed the words of this chapter, " salva persona regis,
reginae et liheroriim suoru7n."^
This chapter has been acclaimed as embodying for the
first time the idea that formed " the true corner stone of the
English Constitution,"^ namely, the right to compel an
erring King to bow to a body of law that lies outside his
will. There is much to be said for this view. It is quite
consistent, however, to combine an appreciation of the value
of this conception, with an admission of the defective and
clumsy nature of the machinery by which a first attempt was
made to realize it.^
IV. Modern Criticism. Until the last twenty-five years
or so, commentators were wont to credit the framers of
Magna Carta with anticipating most of the cardinal prin-
ciples of the modern Constitution. In combating such
exaggerations, it would not be unnatural to lay emphasis
on the extent to which the machinery of this chapter is
condemned by the standards of the nineteenth century.
Yet it is well to steer a middle course, neither praising the
men of 12 15, nor blaming them for failing to achieve the
impossible.
The faults of the scheme, whether viewed from the side of
modern theory or of modern practice, are obvious. It was
a violent measure, full of immediate dangers, and calculated
to exercise a baneful influence on constitutional development
in the future. The fact that Magna Carta provided no
better sanction for its own enforcement than the right of
legalized rebellion, has already been discussed as its cardinal
defect.^ It is instructive to note a few of its other defects in
detail.
(i) The scheme challenged hostility by its want of
^Hantos, op. cit., 150. Adams, Origin, 181 n., suggests a parallel from the
kingdom of Jerusalem. Dr. Riess, Historische Zeitschrift, 1906, p. 170, compares
also the Ephors of Sparta.
^ Liber de Antiquis Legibus, 53. ^ Adams, Pol. Hist. Eng., 11. 439.
*Cf. Adams, Origin, 276 n. ^See supra, p. 129.
474 MAGNA CARTA
moderation. On every vexed political question of the day,
John's authority would have been superseded by that of
twenty-five of the most hostile faction of the baronage. If
the King thought himself aggrieved in anything, he would
require to plead his cause before a tribunal in which his
opponents sat as judges.^ The scheme was thus repugnant
to loyal Englishmen, who cherished a respect for the mon-
archy. No King would submit tamely to remain a sove-
reign, whose " sovereignty " existed on sufferance of his
enemies. The powers thus conferred in 12 15 were more
sweeping than those conferred on a similar committee in
1258, and yet the Parliament which appointed the latter has
been branded as "the Mad Parliament," because of the
violence of its measures.
(2) Rebellion, even where morally justified, is necessarily
illegal ; to attempt to map out for it a legitimate sphere of
action is to attempt the logically impossible. The barons
had failed to rise to the true conception of a limited mon-
archy; their scheme recognized a King still absolute in
some matters, but in others powerless and abject. The
powers of the twenty-five, a body which received no proper
organization, were those of aggression rather than of
administration. Viewed in this modern light, the claims of
the barons to constructive statesmanship rank low.
(3) The powers of the revolutionary committee, excessive
though ill-defined, backed by the sworn obedience of all
classes of the nation, would tend completely to paralyze the
King. The nominal sovereign, nervous under this sword
of Damocles, would lose all power of initiative, while the
committee, powerful to reduce him to impotence, would be
powerless to goad him into action or to act in his stead.
The revolutionary committee had been planned as a drag
on a bad executive, not as a good executive to take its place.
(4) Even as a drag, the efficiency of the committee would
have been neutralized in either of two contingencies : if the
l)arons composing it disagreed among themselves, or, if the
^Dr. Riess, Historische Zeitschrift, 1906, p. 170, thinks this goes too far. Cf.,
however, Adams, Origin, 179: John "was reduced to the function of executing
the judgments of a court not his own."
CHAPTER SIXTY-ONE 475
King refused to surrender. Not a step to restrain the King
could legally be taken, until he had received formal intima-
tion followed by an interval of forty days, during which he
might complete his preparation for war without fear of
interruption.
(5) If the scheme of the barons seems ill-suited to the
needs of the hour of its conception, it was fraught with
even greater dangers to the future development of the
English constitution. The problem it sought to solve was
one of no transient or unimportant nature : the barons
sought the best method of turning royal promises into laws
which succeeding Kings must obey. In attempting this,
Magna Carta moved along lines that were radically wrong ; ^
which, if not departed from in time, would have rendered
any enduring progress impossible. The statesmanship
which, while leaving one King on the throne, subjected him
to the dictation of " five-and-twenty over-kings " was crude
and ill-advised. It is true that the party of reform, through-
out the long reign of Henry III., clung to the same
erroneous solution ; but they met with no success. After
half a century of unrest, a settlement seemed as far distant
as before. The dangers of schemes like those of 12 15, 1244,
and 1258 are clearly seen in contrast with the more tactful
efforts of Edward I. towards a true solution, along lines
leading in due time to complete success.
The true policy for the barons was to use the King's own
administrative machinery and the King's own servants to
control the King. The principle was slowly established
that the sovereign could perform no single act of preroga-
tive except through the agency of a particular officer or
organ of the royal household; while very gradually the
doctrine of ministerial responsibility grew up, compelling
each officer of the Crown to obey not only the law of the
land, but also the Commune Concilium.^ fast changing into
the modern Parliament. The credit of starting the consti-
tution on its right line of development is in great measure
due to Edward i.^
' Cf. Adams, Origin, 179 : " It was not finally to be the way of the constitution."
^Cf. supra^ pp. 159-164, for a sketch of Edward's policy.
476 MAGNA CARTA
V. Failure of the Scheme, Almost before John's Charter
had been engrossed and sealed, the futility of its " sanction "
was recognized. Each side grew suspicious and demanded
new " sanctions " not contained in the Charter.
(i) Quis custodiet ipsos custodes? Magna Carta, assum-
ing apparently that perfect trust could be placed in the
revolutionary committee, provided no machinery for con-
trolling them, no guarantee that they would observe the
Charter. The futility of this complacency was soon mani-
fest. One tyrant had brought distress on the whole nation ;
and now he was to be superseded by five-and-twenty. Who
was to restrain the new tyrants ? A second committee was
nominated, partly to assist and partly to control the twenty-
five. Matthew Paris ^ describes it as composed of thirty-
eight " Ohsecutores et Observatores," including the Earl
Marshal, Hubert de Burgh, the earls of Arundel and
Warenne, and other prominent members of the moderate
party, not unfriendly to the King. Dr. Stubbs dismisses
their relations to the executors with the remark that they
"swore to obey the orders of the twenty-five." ^ Miss
Norgate takes what seems to be a better view, in em-
phasizing, as the chief reason for their appointment, the
duty of compelling " both the King and the twenty-five
to deal justly with one another."^ The thirty-eight were
required to constrain the twenty-five, as the twenty-five
constrained the King.^
(2) Suspicions of the barons* good faith. There is
evidence that the King was distrustful of the barons' good
faith, and desired on his part some " sanction " that they
would not again renounce allegiance. The barons' pro-
mise to grant John security, and the written protest against
their breach of faith, made by Langton and other prelates
at John's request, have already been described.^
1 Chron. Maj., II. 605-6. - Const. Hist., I. 583 n. ^John Lackland, 236.
^ One version of the narrative of Matthew Paris is fuller than the other. " Isti
omnes Juraverunt quod ohsequerenttw mandato viginti qninqne baronuvi " of the
first becomes " Omnes isti juravertmt cogere si opus esset ipsos xxv. barones ut
redificarent regem. Et etiam cogere ipsnm si niutato ani mo forte 7'ecalcitraret''''
in the second, II. 606 n.
^ See supra, p. 43, and Proter-t in Appendix.
CHAPTER SIXTY-ONE 477
(3) Suspicions of John's good faith. The barons, on
their part, soon came to the conclusion that the Com-
mittee, in spite of all its powers, formed an inadequate
sanction against John. They demanded further " security."
The city of London was placed in their hands, and the
Tower of London in the neutral custody of the primate, as
pledges of John's good faith, until 15th August or longer
if need were. Those terms were reduced to writing in a
document entitled " Conventio facta inter Re gem Angliae
et harones ejusdem regni," which thus supplied a new
"form of security," supplementing, if not superseding,
that contained in chapter 61.^
(4) Precautions against papal intervention. The Articles
of the Barons afford evidence of the framers' suspicions
that John would apply to Rome for release from his bargain.
They demanded that the English prelates and the papal
legate should become the King's sureties, that he would
not invite the Pope to invalidate the Charter. If Pandulf,
as the Pope's accredited agent, had put seal to such a
document, he would have seriously embarrassed his august
master.
Two important alterations in the completed Charter were
effected, however, whether at John's instance, or at that of
Pandulf, or of the English prelates, is matter of conjecture.
All mention of Innocent by name was omitted, the clause
being made quite general in its terms : John promised to
procure a dispensation " from no one " ; while the question
of sureties was ignored. Innocent was left free to support
John's policy of repudiation. 2
^ See suj>ra, p. 43. The text is given in Appendix. Thirteen of the twenty-five
executors are mentioned by name as agreeing to this new treaty ; cf. Wendover,
III. 319. A third sanction appears in the garbled versions of the Charter given by
Wendover (III. 317) and M. Paris (II. 603) : the constables of the four royal
castles of Northampton, Kenilworth, Nottingham, and Scarborough were to swear
to hold these strongholds under orders of the twenty-five. This clause has not
been found in any known copy of any issue of Magna Carta : cf. Luard's Preface
to M. Paris, II. xxxiii to xxxvi.
^Cf. supra, P' 45*
478 MAGNA CARTA
CHAPTER SIXTY-TWO.
Et omnes malas voluntates, indignaciones, et rancores ortos
inter nos et homines nostros, clericos et laicos, a tempore
discordie, plene omnibus remisimus et condonavimus.
Preterea omnes transgressiones factas occasione ejusdem
discordie, a Pascha anno regni nostri sextodecimo usque
ad pacem reformatam, plene remisimus omnibus, clericis et
laicis, et quantum ad nos pertinet plene condonavimus. Et
insuper fecimus eis fieri litteras testimoniales patentes
domini Stephani Cantuariensis archiepiscopi, domini
Henrici Dublinensis archiepiscopi, et episcoporum predic-
torum, et magistri Pandulfi, super securitate ista et conces-
sionibus prefatis.
And all the ill-will, hatreds, and bitterness that have arisen
between us and our men, clergy and lay, from the date of the
quarrel, we have completely remitted and pardoned to everyone.
Moreover, all trespasses occasioned by the said quarrel, from
Easter in the sixteenth year of our reign till the restoration of
peace, we have fully remitted to all, both clergy and laymen, and
completely forgiven, as far as pertains to us. And, on this
head, we have caused to be made for them letters testimonial
patent of the lord Stephen, archbishop of Canterbury, of the
lord Henry, archbishop of Dublin, of the bishops aforesaid, and
of Master Pandulf as touching this security and the concessions
aforesaid.
The clauses that follow the forma securitatis are of a
formal nature. The present chapter, after making a well-
meant declaration that bygones should be bygones, so that
peace and goodwill should everywhere prevail — a pious
aspiration doomed to speedy disillusion — proceeds to
authorize the prelates to issue, under their seals, certified
copies of the Great Charter. Such letters were actually
issued, and their terms are preserved in the Red Book of
the Exchequer. 1
^See folio 234. Compare supra, p. 41 ; also R. L. Poole in Engl. Hist.
Rev., XXVIII. 448 fF. The text, as reproduced by Bemont, Chartes, 35, runs as
follows: "Omnibus Christi fidelibus ad quos presens scriptum pervenerit,
Stephanus Dei gratia Cantuariensis archiepiscopus, tocius Anglie primas et sancte
romane ecclesie cardinalis, Henricus, eadem gratia Dublinensis archiepiscopus,
CHAPTER SIXTY-THREE 479
CHAPTER SIXTY-THREE.
QuARE volumus et firmiter precipimus quod Anglicana
ecclesia libera sit et quod homines in regno nostro habeant
et teneant omnes prefatas libertates, jura, et concessiones,
bene et in pace, libere et quiete, plene et integre sibi et
heredibus suis, de nobis et heredibus nostris, in omnibus
rebus et locis, in perpetuum, sicut predictum est. Juratum
est autem tarn ex parte nostra quam ex parte baronum,
quod hec omnia supradicta bona fide et sine malo ingenio
observabuntur. Testibus supradictis et multis aliis. Data
per manum nostram in prato quod vocatur Ronimede, inter
Windlesoram et Stanes, quinto decimo die Junii, anno regni
nostri decimo septimo.
Wherefore it is our will, and we firmly enjoin, that the English
Church be free, and that the men in our kingdom have and
hold all the aforesaid liberties, rights, and concessions, well and
peaceably, freely and quietly, fully and wholly, for themselves
and their heirs, of us and our heirs, in all respects and in all
places for ever, as is aforesaid. An oath, moreover, has been
taken, as well on our part as on the part of the barons, that all
these conditions aforesaid shall be kept in good faith and without
evil intent. Given under our hand — the above-named and many
others being witnesses — in the meadow which is called Runny-
mede, between Windsor and Staines, on the fifteenth day of
June, in the seventeenth year of our reign.
This last of the sixty-three chapters into which Magna
Carta has been divided by modern commentators, contains
little that calls for remark. Beginning with a repetition of
the declarations made in chapter one that the English
Willelmus Londoniensis, Petrus Wintoniensis, Joscelinus Bathoniensis et Glas-
toniensis, Hugo Lincolniensis, Walterus Wigorniensis, Willelmus Coventriensis
et Benedictus Roffensis, divina miseracione episcopi, et magister Pandulfus domini
pape subdiaconus et familiaris, salutem in Domino. Sciatis nos inspexisse cartam
quam dominus noster Johannes illustris rex Anglie fecit comitibus, baronibus et
liberis hominibus suis Anglie de libertate sancte ecclesie et libertatibus et
liberis consuetudinibus suis eisdem ab eo concessis sub hac forma . . .
[Here follows the text of John's Magna Carta]
Et ne huic forme predicte aliquid possit addi vel ab eadem aliquid possit subtrahi
vel minui, huic scripto sigilla nostra apposuimus.'
J C i >1 < I ^ , \X^^^
48o MAGNA CARTA
church should be free (omitting, however, any second refer-
ence to canonical election) and that homines in regno nostro
should have and hold all of the aforesaid liberties, rights
and concessions, it records that both parties had taken oath
to observe its contents in good faith. ^ The magnates named
in the preamble were thereafter, along with many others
unnamed, referred to collectively as witnesses. The
Charter concludes with a declaration that it has been " given
by our hand," and the place and date are specified. The
actual " giving " by John's hand was effected by impress of
his great seal .2
^Cf. stepra, p. 40.
^ There are no signatures to the document. The frequent references to '* the
signing of the Great Charter" are thus inaccurate, if '"signing" is taken in its
modern sense of "subscribing," but may perhaps be justified by a reference to
signum in its original meaning of **a seal." To imprint a seal was, in a sense,
" to sign." Reasons have already been given for holding that Magna Carta, in
spite of its mention of its own date as 15th June, was actually sealed on the 19th.
See supra, pp. 48-9.
APPENDIX.
DOCUMENTS RELATIVE TO, OR ILLUSTRATIVE OF,
MAGNA CARTA.
I. THE CHARTER OF LIBERTIES OF HENRY I.i
(iioo.)
Henricus Del gratia rex Anglorum omnibus baronibus et
fidelibus suis tarn Francigenis quam Anglis salutem.
1. Sciatis me Dei misericordia et communi consilio baronum
regni Anglie ejusdem regni regem coronatum esse. Et quia
regnum oppressum erat injustis exactionibus, ego, respectu Dei
et amore quem erga vos omnes habeo, sanctam Dei ecclesiam
imprimis liberam facio : ita quod nee vendam nee ad firmam
ponam nee, mortuo archiepiscopo sive episcopo sive abbate,
aliquid accipiam de dominio ecclesie vel de hominibus ejus,
donee successor in eam ingrediatur. Et omnes malas con-
suetudines, quibus regnum Anglie injuste opprimebatur, inde
aufero ; quas malas consuetudines ex parte hie pono :
2. Si quis baronum meorum, comitum, sive aliorum qui de me
tenent, mortuus fuerit, heres suus non redimet terram suam
sicut faciebat tempore fratris mei, sed legitima et justa releva-
tione relevabit eam. Similiter et homines baronum meorum
legitima et justa relevatione relevabunt terras suas de dominis
suis.
3. Et si quis baronum vel aliorum hominum meorum filiam
suam nuptum tradere voluerit, sive sororem, sive neptem, sive
cognatam, mecum inde loquatur. Sed neque ego aliquid de
suo pro hac licentia accipiam, neque defendam ei quin eam
1 The text is taken from Liebermann, Gesetze, I. 521. Cf. Trans. li.H.S.
viii. 21 ff., for an exhaustive discussion of the various copies of the lost charter.
For commentary, cf. supra, pp. 96-101. Liebermann shows the striking variations
of the opening and ending clauses : the preamble varied with the persons to whom
each copy was addressed. Cf. R. L. Poole, E^ig. Hist. Rev,^ XXVIII. 444 ff.
2 H
482 APPENDIX
d€t, excepto si earn vellet jungere inimico meo. Et si, mortuo
barone vel alio homine meo, filia heres remanserit, illam dabo
consilio baronum meorum cum terra sua. Et si, mortuo marito,
uxor ejus remanserit et sine liberis fuerit, dotem suam et
maritationem habebit; et earn non dabo marito, nisi secundum
velle suum.
4. Si vero uxor cum liberis remanserit, dotem quidem et
maritationem suam habebit, dum corpus suum legitime serva-
verit ; et eam non dabo, nisi secundum velle suum. Et terre et
liberorum custos erit sive uxor sive alius propinquorum, qui
justius esse debebit. Et precipio ut barones mei similiter se
contineant erga filios et filias vel uxores hominum suorum.
5. Monetagium commune quod capiebatur per civitates et
comitatus, quod non fuit tempore regis Eadwardi, hoc ne amodo
sit, omnino defendo. Si quis captus fuerit, sive monetarius^
sive alius, cum falsa moneta, justicia recta inde fiat.
6. Omnia placita et omnia debita que fratri meo debebantur
condono exceptis rectis firmis meis et exceptis illis que pacta
erant pro aliorum hereditatibus, vel pro eis rebus que justius
aliis contingebant. Et si quis pro hereditate sua aliquid pepi-
gerat, illud condono, et omnes relevationes que pro rectis
hereditatibus pacte erant.
7. Et si quis baronum vel hominum meorum infirmabitur^
sicut ipse dabit vel dare disponet pecuniam suam, ita datam
esse concedo. Quodsi ipse, preventus vel armis vel infirmitate^
pecuniam suam non dederit nee dare disposuerit uxor sua sive
liberi aut parentes aut legitimi homines ejus eam pro anima ejus
dividant, sicut eis melius visum fuerit.
8. Si quis baronum vel hominum meorum forisfecerit, non
dabit vadium in misericordia pecunie sue, sicut faciebat tempore
patris mei vel fratris mei, sed secundum modum forisfacti, ita
emendabit, sicut emendasset retro a tempore patris mei, in
tempore aliorum antecessorum meorum. Quodsi perfidie vel
sceleris convictus fuerit, sicut justum fuerit, sic emendet.
9. Murdra etiam retro ab ilia die qua in regem coronatus fui
omnia condono ; et ea que amodo facta fuerint juste emendentur
secundum lagam regis Eadwardi.
10. Forestas communi consensu baronum meorum in manu
mea retinui, sicut pater meus eas habuit.
11. Militibus qui per loricas terras suas deserviunt terras
dominicarum suarum quietas ab omnibus geldis et ab omni
opere, proprio dono meo, concedo, ut, sicut tam magno grava-
CHARTER OF HENRY I. 4^3
mine alleviati sunt, ita equis et armis se bene instruant, ut apti
et parati sint ad servitium meum et ad defensionem regni mei.
12. Pacem firmam in toto regno meo pono et teneri* amodo
precipio.
13. Lagam regis Eadwardi vobis reddo cum illis emendationi-
bus quibus pater meus earn emendavit consilio baronum suorum.
14. Si quis aliquid de meo vel de rebus alicujus post obitum
regis Willelmi, fratis mei, cepit, totum cito reddatur absque
emendatione. Et si quis inde aliquid retinuerit, ille super quem
inventum fuerit graviter michi emendabit.
Testibus Mauricio Lundonie episcopo, et Gundulfo episcopo,
et Willelmo electo episcopo, et Henrico comite, et Simone
comite, et Waltero Giffardo, et Rodberto de Monfort, et Rogero
Bigoto, et Henrico de Portu, apud Westmonasterium, quando
coronatus fui. Valete !
H. THE SECOND OR OXFORD CHARTER OF STEPHEN.i
(1136.)
Ego Stephanos Dei gratia, assensu cleri et populi in regem
Anglie electus, et a Willelmo Cantuariensi archiepiscopo et
sancte Romane ecclesie legato consecratus, et ab Innocentio
sancte romane sedis pontifice postmodum confirmatus, respectu
et amore Dei sanctam ecclesiam liberam esse concedo, et debitam
reverentiam illi confirmo. Nichil me in ecclesia vel rebus
ecclesiasticis simoniace acturum vel permissurum esse promitto.
Ecclesiasticarum personarum et omnium clericorum et rerum
eorum justiciam et potestatem et distributionem bonorum
ecclesiasticorum in manu episcoporum esse perhibeo et con-
firmo. Dignitates ecclesiarum privilegiis earum confirmatas et
consuetudines earum antiquo tenore habitas inviolate manere
statuo et concedo. Omnes ecclesiarum possessiones et tenuras,
quas die ilia habuerunt qua Willelmus rex avus meus fuit vivus
et mortuus, sine omni calumpniantium reclamatione, eis liberas
et absolutas esse concedo. Si quid vero de habitis vel possessis
^The text is founded on that of the StaUites of the Realm, I. 3, which follows
the Exeter version. Cf. Bemont, Chartes, 8-10, who discusses the various editions.
Dr. R. L. Poole has noted the variants of an original of the Charter preserved in
the muniment room of Salisbury Cathedral ; see Report on Mamiscripts in Various
Collections, I. 384-5 (Historical Manuscripts Commission, 1901). Two of these
variants have been here adopted: (a) *■'■ regem Anglie'' for ^^ regem Anglortim'''
and (p) '■'■postmodum''' added after ^^ pontifice."
484 APPENDIX
ante mortem ejusdem regis quibus modo careat, ecclesia deinceps
repetierit, indulgentie et dispensationi mee vel restituendum
vel discutiendum reserve. Quecunque vero post mortem ipsius
regis liberalitate regum vel largitione principum, oblatione vel
comparatione, vel qualibet transmutatione fidelium eis coUata
sunt, confirmo. Pacem et justiciam me in omnibus facturum
et pro posse meo conservaturum eis promitto.
Forestas quas Willelmus avus meus et Willelmus avunculus
meus instituerunt et habuerunt mihi reservo. Ceteras omnes
quas rex Henricus superaddidit, ecclesiis et regno quietas reddo
et concedo.
Si quis episcopus vel abbas vel alia ecclesiastica persona ante
mortem suam rationabiliter sua distribuerit vel distribuenda
statuerit, firmum manere concedo. Si vero morte preoccupatus
fuerit, pro salute anime ejus, ecclesie consilio, eadem fiat distri-
butio. Dum vero sedes propriis pastoribus vacue fuerint, ipsas
et earum possessiones omnes in manu et custodia clericorum
vel proborum hominum ejusdem ecclesie committam, donee
pastor canonice substituatur.
Omnes exactiones et injusticias et mescheningas sive per vice-
comites vel per alios quoslibet male inductas funditus exstirpo.
Bonas leges et antiquas et justas consuetudines in murdris et
placitis et aliis causis observabo et observari precipio et con-
stituo. Hec omnia concedo et confirmo, salva regia et justa
dignitate mea.
Testibus Willelmo Cantuariensi archiepiscopo, et Hugone
Rothomagensi archiepiscopo, et Henrico Wintoniensi episcopo,
et Rogero Saresberiensi episcopo, et Alexandre Lincolniensi
episcopo, et Nigello Eliensi episcopo, et Evrardo Norwicensi
episcopo, et Simone Wigorniensi episcopo, et Bernardo episcopo
de S. Davide, et Audoeno Ebroicensi episcopo, et Ricardo
Abrincensi episcopo, et Roberto Herefordiensi episcopo, et
Johanne Rovecestriensi episcopo, et Athelulfo Carlolensi epis-
copo, et Rogero cancellario, et Henrico nepote Regis, et Roberto
comite Gloecestrie, et Willelmo comite de Warenna, et Ran-
nulf o comite Cestrie, et Rogero comite de Warewic. , et Roberto
de Ver., et Milone de Gloecestria, et Brientio filio Comitis, et
Roberto de Oilly conestabulis, et Willelmo Martello, et Hugone
Bigot, et Hunfredo de Buhun, et Simone de Belcamp dapiferis,
et Willelmo de Albiniaco, et Eudone Martello pincernis, et
Roberto de Ferreriis, et Willelmo Pevrello de Notingeham,
et Simone de Saintliz, et Willelmo de Albamarla, et Pagano
CHARTER OF STEPHEN 4^5
filio Johannis, et Hamone de Sancto Claro, et Ilberto de Laceio.
Apud Oxeneford. Anno ab incarnatione Domini M.C. XXXVI.,
set regni mei primo.
HI. CHARTER OF HENRY H.i
(circa 1 154.)
Henricus Dei gracia rex anglie, dux Normannie et Aquitanie,
et comes Andegavie, omnibus comitibus, baronibus et fidelibus
suis Francis et Anglicis, salutem. Sciatis me, ad honorem Dei
et sancte Ecclesie, et pro communi emendacione tocius regni
mei, concessisse et reddidisse et presenti carta mea confirmasse
Deo et sancte ecclesie et omnibus comitibus et baronibus et
omnibus hominibus meis omnes concessiones et donaciones et
libertates et liberas consuetudines, quas rex Henricus avus
meus eis dedit et concessit. Similiter eciam omnes malas
consuetudines, quas ipse delevit et remisit, ego remitto et deleri
concede pro me et heredibus meis. Quare volo et firmiter
precipio quod sancta ecclesia et omnes comites et barones et
omnes mei homines omnes illas consuetudines et donaciones et
libertates et liberas consuetudines habeant et teneant libere
et quiete, bene et in pace et integre, de me et heredibus
meis, sibi et heredibus suis, adeo libere et quiete et plenarie
in omnibus, sicut rex Henricus avus meus eis dedit et con-
cessit, et carta sua confirmavit. Teste Ricardo de Luci apud
Westmonasterium.
IV. THE SO-CALLED " UNKNOWN CHARTER OF
LIBERTIES " OF JOHN.2
(circa 1214-1215.)
I. Concedit Rex Johannes quod non capiet hominem absque
judicio, nee aliquid accipiet pro justitia, nee injustitiam faciet.
^ The text is taken from that given in Stattites of the Realm, I. 4, which is
founded on a copy of the original preserved in the British Museum (Cotton,
Claudius D. II., folio 107). Cf. Bemont, Charies, 12-14.
2 For a discussion of the nature, date, and historical context of this document see
supra, pp. 171 -5 and Index. The text is founded upon that published by Mr. J. H.
Round in the English Historical Review, VIII. 288, but effect has been given to
most of the emendations suggested by Mr. Hubert Hall and Mr. G. W. Pfothero.
Cf. ibid., IX. 117 and 326. The twelve clauses are here numbered for convenience
of reference, although no numbers appear in the MS.
486 APPENDIX
2. Et si contingat quod meus baro vel homo meus moriatur
pt haeres suus sit in aetate, terram suam debeo ei reddere per
rectum releveium absque magis capiendi.
3. Et si ita sit quod haeres sit infra aetatem, debeo inf^
militibus de legalioribus feodi terram bajulare in custodia, et
illi cum meo famulo debent mihi reddere exitus terrae sine
venditione nemorum et sine redemptione hominum et sine
destructione parci et vivarii; et tunc quando ille haeres erit
in aetate terram ei reddam quietam.
4. Si foemina sit haeres terrae, debeo eam maritare, con-
silio generis sui, ita non sit disparagiata. Et si una vice eam
dedero, amplius eam dare non possum, sed se maritabit ad
libitum suum, sed non inimicis meis.
5. Si contingat quod baro aut homo meus moriatur, concede
ut pecunia sua dividatur sicut ipse diviserit; et si prae-
occupatus fuerit aut armis aut infirmitate improvisa, uxor
ejus, aut liberi, aut parentes et amici propinquiores pro ejus
anima dividant.
6. Et uxor ejus non abibit de hospitio infra xl. dies et
donee dotem suam decenter habuerit, et maritagium habebit.
7. Adhuc hominibus meis concedo ne eant in exercitu extra
Angliam nisi in Normanniam et in Britanniam et hoc decenter;
quod si aliquis debet inde servitium decem militum, consilio
baronum meorum alleviabitur.
8. Et si scutagium evenerit in terra, una marca argenti
capietur de feodi militis; et si gravamen exercitus contigerit,
amplius caperetur consilio baronum regni.
9. Adhuc concedo ut omnes forestas quas pater meus et
frater meus et ego afforestaverimus, deafforesto.
10. Adhuc concedo ut milites qui in antiquis forestis meis
suum nemus habent, habeant nemus amodo ad herbergagia
sua et ad ardendum ; et habeant foresterium suum ; et ego
tantum modo unum qui servet pecudes meas.
11. Et si aliquis hominum meorum moriatur qui Judaeis
debeat, debitum non usurabit quamdiu haeres ejus sit infra
aetatem.
12. Et concedo ne homo perdat pro pecude vitam neque
membra.
THE ARTICLES OF THE BARONS 487
V. THE ARTICLES OF THE BARONS.i^
(1215.)
Ista sunt Capitula que Barones petunt et dominus Rex concedit.
1. Post decessum antecessorum heredes plene etatis habebunt
hereditatem suam per antiquum relevium exprimendum in carta.
2. Heredes qui infra etatem sunt et fuerint in custodia, cum
ad etatem pervenerint, habebunt hereditatem suam sine relevio
et fine.
3. Custos terre heredis capiet rationabiles exitus, consuetu-
dines, et servitia, sine destructione et vasto hominum et rerum
suarum, et si custos terre fecerit destructionem et vastum,
amittat custodiam ; et custos sustentabit domos, parcos, vivaria,
stagna, molendina et cetera ad terram illam pertinentia, de
exitibus terre ejusdem ; et ut heredes ita maritentur ne dispara-
gentur et per consilium propinquorum de consanguinitate sua.
4. Ne vidua det aHquid pro dote sua, vel maritagio, post deces-
sum mariti sui, sed maneat in domo sua per .xl. dies post mortem
ipsius, et infra terminum ilium assignetur ei dos ; et maritagium
statim habeat et hereditatem suam.
5. Rex vel ballivus non saisiet terram aliquam pro debito dum
catalla debitoris suflficiunt; nee plegii debitoris distringantur,
dum capitalis debitor suflRcit ad solutionem ; si vero capitalis
debitor defecerit in solutione, si plegii voluerint, habeant terras
debitoris, donee debitum illud persolvatur plene, nisi capitalis
debitor monstrare poterit se esse inde quietum erga plegios.
6. Rex non concedet alicui baroni quod capiat auxilium de
liberis hominibus suis, nisi ad corpus suum redimendum, et ad
faciendum primogenitum filium suum militem, et ad primo-
genitam filiam suam semel maritandam, et hoc faciet per
rationabile auxilium.
7. Ne aliquis majus servitium faciat de feodo militis quam
inde debetur.
8. Ut communia placita non sequantur curiam domini regis,
sed assignentur in aliquo certo loco ; et ut recognitiones capian-
tur in eisdem comitatibus, in hunc modum : ut rex mittat duos
justiciaros per .iiii*^'". vices in anno, qui cum .iiii^''. militibus
ejusdem comitatus electis per comitatum, capiant assisas de nova
^ The text is taken from that of the Statutes of the Realm, I. ^-Z, which is
founded on the original in the British Museum. See stipra, pp. 1 70- 1. Cf. Bemont,
Chartes, 15-23.
488 APPENDIX
dissaisina, morte antecessoris, et ultima presentatione, nee
aliquis ob hoc sit summonitus nisi juratores et due partes.
9. Ut liber homo amercietur pro parvo delicto secundum
modum delicti, et, pro magno delicto, secundum magnitudinem
delicti, salvo continemento suo ; villanus etiam eodem modo
amercietur, salvo waynagio suo; et mercator eodem modo,
salva marcandisa, per sacramentum proborum hominum de
visneto.
10. Ut clericus amercietur de laico feodo suo secundum
modunj aliorum predictorum, et non secundum beneficium
ecclesiasticum.
11. Ne aliqua villa amercietur pro pontibus faciendis ad
riparias, nisi ubi de jure antiquitus esse solebant.
12. Ut mensura vini, bladi, et latitudines pannorum et rerum
aliarum, emendetur; et ita de ponderibus.
13. Ut assise de nova dissaisina et de morte antecessoris
abbrevientur ; et similiter de aliis assisis.
14. Ut nullus vicecomes intromittat se de placitis ad coronam
pertinentibus sine coronatoribus ; et ut comitatus et hundred!
sint ad antiquas firmas absque nullo incremento, exceptis
dominicis maneriis regis.
15. Si aliquis tenens de rege moriatur, licebit vicecomiti vel
alii ballivo regis seisire et imbreviare catallum ipsius per visum
legalium hominum, ita tamen quod nichil inde amoveatur, donee
plenius sciatur si debeat aliquod liquidum debitum domino regi,
et tunc debitum regis persolvatur; residuum vero relinquatur
executoribus ad faciendum testamentum defuncti; et si nichil
regi debetur, omnia catalla cedant defuncto.
16. Si aliquis liber homo intestatus decesserit, bona sua per
manum proximorum parentum suorum et amicorum et per visum
ecclesie distribuantur.
17. Ne vidue distringantur ad se maritandum, dum voluerint
sine marito vivere, ita tamen quod securitatem facient quod non
maritabunt se sine assensu regis, si de rege teneant, vel domi-
norum suorum de quibus tenent.
18. Ne constabularius vel alius ballivus capiat blada vel alia
catalla, nisi statim denarios inde reddat, nisi respectum habere
possit de voluntate venditoris.
19. Ne constabularius possit distringere aliquem militem ad
dandum denarios pro custodia castri, si voluerit facere custodiam
illam in propria persona vel per alium probum hominem, si ipse
earn facere non possit per rationabilem causam; et si rex eum
THE ARTICLES OF THE BARONS 489
duxerit in exercitum, sit quietus de custodia secundum quanti-
tatem temporis.
20. Ne vicecomes, vel ballivus regis, vel aliquis alius, capiat
equos vel carettas alicujus liberi hominis pro cariagio faciendo,
nisi ex voluntate ipsius.
21. Ne rex vel ballivus suus capiat alienum boscum ad castra
vel ad alia agenda sua, nisi per voluntatem ipsius cujus boscus
ille fuerit.
22. Ne rex teneat terram eorum qui fuerint convicti de
felonia, nisi per unum annum et unum diem, sed tunc reddatur
domino feodi.
23. Ut omnes kidelli de cetero penitus deponantur de Tamisia
et Medeweye et per totam Angliam.
24. Ne breve quod vocatur " Precipe " de cetero fiat alicui de
aliquo tenemento unde liber homo amittat curiam suam.
25. Si quis fuerit disseisitus vel prolongatus per regem sine
juditio de terris, libertatibus, et jure suo, statim ei restituatur;
et si contentio super hoc orta fuerit, tunc inde disponatur per
juditium .xxv. baronum, et ut illi qui fuerint dissaisiti per
patrem vel fratrem regis, rectum habeant sine dilatione per
juditium parium suorum in curia regis ; et si rex debeat habere
terminum aliorum cruce signatorum, tunc archiepiscopus et
episcopi faciant inde juditium ad certum diem, appellatione
remota.
26. Ne aliquid detur pro brevi inquisitionis de vita vel
membris, sed libere concedatur sine pretio et non negetur.
2*^. Si aliquis tenet de rege per feodi firmam, per sokagium,
vel per burgagium, et de alio per servitium militis, dominus
rex non habebit custodiam militum de feodo alterius, occasione
burgagii vel sokagii, nee debet habere custodiam burgagii,
sokagii, vel feodi firme ; et quod liber homo non amittat militiam
suam occasione parvarum sergantisarum, sicuti de illis qui
tenent aliquod tenementum reddendo inde cuttellos vel sagittas
vel hujusmodi.
28. Ne aliquis ballivus possit ponere aliquem ad legem simplici
loquela sua sine testibus fidelibus.
29. Ne corpus liberi hominis capiatur, nee imprisonetur, nee
dissaisietur, nee utlagetur, nee exuletur, nee aliquo modo de-
struatur, nee rex eat vel mittat super eum vi, nisi per juditium
parium suorum vel per legem terre.
30. Ne jus vendatur vel differratur vel vetitum sit.
31. Quod mercatores habeant salvum ire et venire ad emen-
490
APPENDIX
dum vel vendendum, sine omnibus malis toltis, per antiquas
et rectas consuetudines.
32. Ne scutagium vel auxilium ponatur in regno, nisi per
commune consilium regni, nisi ad corpus regis redimendum,
et primogenitum filium suum militem faciendum, et filiam suam
primogenitam semel maritandam ; et ad hoc fiat rationabile
auxilium. Simili modo fiat de taillagiis et auxiliis de civitate
Londonie, et de aliis civitatibus que inde habent libertates, et ut
civitas Londonie plene habeat antiquas libertates et liberas con-
suetucjines suas, tam per aquas, quam per terras.
33. Ut liceat unicuique exire de regno et redire, salva fide
dominis regis, nisi tempore werre per aliquod breve tempus
propter communem utilitatem regni.
34. Si quis mutuo aliquid acceperit a Judeis plus vel minus,
et moriatur antequam debitum illud solvatur, debitum non
usurabit quamdiu heres fuerit infra etatem, de quocumque
teneat; et si debitum illud inciderit in manum regis, rex non
capiet nisi catallum quod continetur in carta.
35. Si quis moriatur et debitum debeat Judeis, uxor ejus
habeat dotem suam ; et si Hberi remanserint, provideantur eis
necessaria secundum tenementum ; et de residuo solvatur debitum
salvo servitio dominorum ; simili modo fiat de aliis debitis ; et ut
custos terre reddat heredi, cum ad plenam etatem pervenerit,
terram suam instauratam secundum quod rationabiliter poterit
sustinere de exitibus terre ejusdem de carucis et wainnagiis.
36. Si quis tenuerit de aliqua eskaeta, sicut de honore Walinge-
ford, Notingeham, Bononie, et Lankastrie, et de aliis eskaetis
que sunt in manu regis et sunt baronie, et obierit, heres ejus
non dabit aliud relevium, vel faciet regi aliud servitium quam
faceret baroni ; et ut rex eodem modo eam teneat quo baro
eam tenuit.
37. Ut fines qui facti sunt pro dotibus, maritagiis, heredi-
tatibus, et amerciamentis, injuste et contra legem terre, omnino
condonentur; vel fiat inde per juditium, .xxv. baronum, vel per
juditium majoris partis eorumdem, una cum archiepiscopo et
aliis quos secum vocare voluerit ita quod, si aliquis vel aliqui de
.XXV. fuerint in simili querela, amoveantur et alii loco illorum
per residuos de .xxv. substituantur.
38. Quod obsides et carte reddantur, quae liberate fuerunt
regi in securitatem.
39. Ut illi qui fuerint extra forestam non veniant coram
justiciariis de foresta per communes summonitiones, nisi sint in
THE ARTICLES OF THE BARONS 491
placito vel plegii fuerint; et ut prave consuetudines de forestis
€t de forestariis, et warenniis, et vicecomitibus, et rivariis,
emendentur per .xii. milites de quolibet comitatu, qui debent
eligi per probos homines ejusdem comitatus.
40. Ut rex amoveat penitus de balliva parentes et totam
sequelam Gerardi de Atyes, quod de cetero balliam non habeant,
scilicet Engelardum, Andream, Petrum, et Gyonem de Cancellis,
Gyonem de Cygony, Matheum de Martiny, et fratres ejus; et
Galfridum nepotem ejus et Philippum Mark.
41. Et ut rex amoveat alienigenas, milites, stipendiaries,
balistarios, et ruttarios, et servientes qui veniunt cum equis et
armis ad nocumentum regni.
42. Ut rex faciat justiciaries, constabularies, vicecomites, et
ballivos, de talibus qui sciant legem terre et eam bene velint
observare.
43. Ut barones qui fundaverunt abbatias, unde habent cartas
regum vel antiquam tenuram, habeant custodiam earum cum
vacaverint.
44. Si rex Walenses dissaisierit vel elongaverit de terris vel
libertatibus, vel de rebus aliis in Anglia vel in Wallia, eis statim
sine placito reddantur; et si fuerint dissaisiti vel elongati de
tenementis suis Anglie per patrem vel fratrem regis sine juditio
parium suorum, rex eis sine dilatione justiciam exhibebit, eo
modo quo exhibet Anglicis justiciam de tenementis suis Anglie
secundum legem Anglie, et de tenementis Wallie secundum
legem Wallie, et de tenementis Marchie secundum legem
Marchie ; idem facient Walenses regi et suis.
nisi aliter esse
debeat per cartas
quas rex habet per
* juditium archi-
episcopi et alio-
rum quos secum
vocare voluerit.
45. Ut rex reddat filium Lewelini et pre-
terea omnes obsides de Wallia, et cartas que
ei liberate fuerunt in securitatem pacis
46. Ut rex faciat regi Scottorum de obsi-
dibus reddendis, et de libertatibus suis, et
jure suo, secundum formam quam facit
baronibus Anglie .....
47. Et omnes foreste que sunt aforestate per regem tempore
suo deafforestentur, et ita fiat de ripariis que per ipsum regem
sunt in defense.
48. Omnes autem istas consuetudines et libertates quas rex
concessit regno tenendas quantum ad se pertinet erga sues,
omnes de regno tarn clerici quam laici observabunt quantum ad
se pertinet erga sues.
[Here, there occurs a blank space in the original.]
492
APPENDIX
49. Hec est forma securitatis ad observandum pacem et liber-
tates inter regem et regnum. Barones eligent .xxv. barones de
regno quos voluerint, qui debent pro totis viribus suis observare,
tenere et facere observari, pacem et libertates quas dominus rex
eis concessit et carta sua confirmavit ; ita videlicet quod si rex,
vel justiciarius, vel ballivi regis, vel aliquis de ministris suis, in
aliquo erga aliquem deliquerit, vel aliquem articulorum pacis aut
securitatis transgressus fuerit, et delictum ostensum fuerit .iiiio"".
baronibus de praedictis .xxv. baronibus, illi .iiii^''. barones acce-
dent ad dominum regem, vel ad justiciarium suum, si rex fuerit
extra regnum; proponentes ei excessum, petent ut excessum
ilium sine dilatione faciat emendari ; et si rex vel justiciarius
ejus illud non emendaverit, si rex fuerit extra regnum, infra
rationabile tempus determinandum in carta, predicti .iiii*^^. re-
ferent causam illam ad residuos de illis .xxv. baronibus, et illi
,xxv. cum communa totius terre distringent et gravabunt regem
modis omnibus quibus poterunt, scilicet per captionem castrorum,
terrarum, possessionum, et aliis modis quibus poterunt, donee
fuerit emendatum secundum arbitrium eorum, salva persona
domini regis et regine et liberorum suorum ; et cum fuerit emen-
datum, intendant domino regi sicut prius. Et quicumque
voluerit de terra jurabit se ad predicta exequenda pariturum man-
datis predictorum .xxv. baronum, etgravaturum regem pro posse
suo cum ipsis ; et rex pubblice et libere dabit licentiam jurandi
cuilibet qui jurare voluerit, et nuUi umquam jurare prohibebit.
Omnes autem illos de terra qui sponte sua et per se noluerint
jurare .xxv. baronibus de distringendo et gravando regem cum
eis, rex faciet jurare eosdem de mandato suo sicut predictum
est. Item si aliquis de predictis .xxv. baronibus decesserit, vel a
terra recesserit, vel aliquo modo alio impeditus fuerit quominus
ista predicta possint exequi, qui residui fuerint de .xxv. eligent
alium loco ipsius pro arbitrio suo, qui simili modo erit juratus
quo et ceteri. In omnibus autem que istis .xxv. baronibus com-
mittuntur exequenda,- si forte ipsi .xxv. presentes fuerint et inter
se super re aliqua discordaverint, vel aliqui ex eis vocati nolint
vel nequeant interesse, ratum habebitur et firmum quod major
pars ex eis providerit vel preceperit, ac si omnes .xxv. in hoc
consensissent ; et predicti .xxv. jurabunt quod omnia antedicta
fideliter observabunt et pro toto posse suo facient observari.
Preterea rex faciet eos securos per cartas archiepiscopi et epis-
coporum et magistri Pandulfi, quod nichil impetrabit a domino
papa per quod aliqua istarum conventionum revocetur vel
THE ARTICLES OF THE BARONS 493
minuatur, et, si aliquid tale impetraverit, reputetur irritum et
inane et numquam eo utatur.
VI. WRITS SUPPLEMENTARY OF JOHN'S GREAT
CHARTER.
(i) Writ to Stephen Harengod, dated 23rd June, 12 15, announc-
ing that terms had been arranged.'^
Rex Stephano Harengod etc., Sciatis quod firma pax facta
est per Dei gratiam inter nos et barones nostros die Veneris
proximo post festum Sancte Trinitatis apud Runemed. , prope
Stanes; ita quod eorum homagia eodem die ibidem cepimus.
Unde vobis mandamus firmiter precipientes quod sicut nos et
honorem nostrum diligitis et pacem regni nostri, ne ulterius
turbetur, quod nullum malum de cetero faciatis baronibus
nostris vel aliis, vel fieri permittatis, occasione discordie prius
orte inter nos et eos. Mandamus etiam vobis quod de finibus et
tenseriis nobis factis occasione illius discordie, si quid superest
reddendum, nichil capiatis. Et si quid post ilium diem
Veneris cepistis, illud statim reddatis. Et corpora prisonum
et obsidum captorum et detentorum occasione hujus guerre, vel
finium vel tenseriarum predictarum, sine dilatione deliberetis.
Hec omnia predicta, sicut corpus vestrum diligitis, faciatis.
Et in hujus etc., nobis mittimus. Teste meipso apud Runemed.,
xxiij. die Junii anno regni nostri xvij.
(2) Writ to Hugh de Bova, dated 2yd June, 12 15, ordering dis-
handment of mercenaries.^
Rex Hugoni de Bova, salutem. Mandamus vobis quod in
fide qua nobis tenemini non retineatis aliquem de militibus
vel servientibus qui fuerunt apud Dover., sed in patriam
suam in pace sine dilatione ire faciatis. Et in hujus, etc.
Teste meipso apud Runimed. xxiij. die Junii anno regni nostri
xvijmo.
*The text follows that of New Kyiiier^ I. 133, but has been collated with Rot.
Pat.^ I. 143 (17 John, m. 23) and two corrections made. This writ is referred to
stipra^ p. 41 n., where its date is discussed.
*See supra^ p. 42. The text is given in New Ryuner, I. 134, and in Rot. Pat.^
I. 144(17 John, m. 23).
494 APPENDIX
(3) Writs issued to the sheriffs of counties on igth June, 1215.1
Rex vicecomiti, forestariis, warennariis, custodibus ripari-
arum et omnibus baillivis suis in eodem comitatu, salutem.
Sciatis pacem firmam esse reformatam per Dei gratiam
inter nos et barones et liberos homines regni nostri, sicut
audire poteritis et videre per cartam nostram quam inde
fieri fecimus, quam etiam legi publice precepimus per totam
bailliam vestram et firmiter teneri ; volentes et districte pre-
cipientes quod tu vicecomes omnes de baillia tua secundum
formam carte predicte jurare facias xxv. baronibus de quibus
mentio fit in carta predicta, ad mandatum eorundem vel majoris
partis eorum, coram ipsis vel illis quos ad hoc atornaverint
per litteras suas patentes, et ad diem et locum quos ad hoc
faciendum prefixerint predicti barones vel atornati ab eis ad
hoc. Volumus etiam et precipimus quod xii milites de comitatu
tuo, qui eligentur de ipso comitatu in primo comitatu qui tenebi-
tur post susceptionem litterarum istarum in partibus tuis, jurent
de inquirendis pravis consuetudinibus tam de vicecomitibus quam
eorum ministris, forestis, forestariis, warennis et warennariis,
ripariis et earum custodibus, et eis delendis, sicut in ipsa
carta continetur. Vos igitur omnes sicut nos et honorem
nostrum diligitis, et pacem regni nostri, omnia in carta
contenta inviolabiliter observetis et ab omnibus observari
faciatis, ne pro defectu vestri, aut per excessum vestrum,
pacem regni nostri, quod Deus avertat, iterum turbari contin-
gat. Et tu, vicecomes, pacem nostram per totam bailliam
tuam clamari facias et firmiter teneri precipias. Et in hujus,
etc. vobis mittimus. Teste me ipso apud Runimede, xix. die
Junii, anno regni nostri xvij™°-
(4) Writs issued to the sheriffs of counties on 2'jth June, 12 15.2
Rex vicecomiti Warewic. et duodecim militibus electis in
eodem comitatu ad inquirendum et delendum pravas con-
suetudines de vicecomitibus et eorum ministris forestis et
forestariis warennis et warennariis ripariis et earum custodibus
salutem. Mandamus vobis quod statim et sine dilatione sais-
^See sup-a, p. 42. The text is given by New Rymer, I. 134, and in Rot.
Pat., I. 134 (17 John, m. 21). A French version appears in D'Achery, Spicilegium,
XII. 573, and in Bemont, Chartes, XXIV. n.
"See stcpra, pp. 42-3, and 440. The text is taken from Rot. Pat., I. 180
(17 John, m. 23 d.). It will be found also in New Rymer, I. 134, and in Stubbs'
Sel. Chart., 306-7.
SUPPLEMENTARY WRITS 495
iatis in manum nostram terras et tenementa et catalla omnium
illorum de comitatu Warewic. qui jurare contradixerint viginti
quinque baronibus secundum formam contentam in carta nostra
de libertatibus vel eis quos ad hoc atornaverint. Et si jurare
noluerint statim post quindecim dies completos preterquam
terre et tenementa et catalla eorum in manu nostra saisita
fuerint, omnia catalla sua vendi faciatis et denarios inde pre-
ceptos salvo custodiatis, deputandos subsidio terre sancte.
Terras autem et tenementa eorum in manu nostra teneatis,
quousque juraverint. Et hoc provisum est per judicium
domini Cantuar. archiepiscopi et baronum regni nostri. Et in
hujus etc. Teste meipso, apud Winton. xxvij die Junii anno
regni nostri xvij™°-
Idem mandatum est omnibus vicecomitibus Anglic.
(5) Conventio facta inter Re gem Anglic et bar ones ejusdcm
regni. ^
Hec est conventio facta inter dominum Johannem regem
Anglie, ex una parte, et Robertum filium Walteri, marescallum
exercitus Dei et sancte ecclesie in Anglia. et Ricardum comitem
de Clare, Gaufridum comitem Essex, et Glouc, Rogerum
Bigot comitem Northfolc. et Suthfolc, Saherum comitem
Wint., Robertum comitem Oxon., Henricum comitem Here-
ford., et barones subscriptos, scilicet Willielmum Mariscallum
juniorcm, Eustachium de Vescy, Willielmum de Mobray, Johan-
nem filium Roberti, Rogerum de Monte Begonis, Willielmum
de Lanvalay, et alios comites et barones et liberos homines totius
regni, ex altera parte, videlicet quod ipsi comites et barones et
alii prescripti tenebunt civitatem London, de baillio domini regis,
salvis interim domino regi firmis redditibus et claris debitis suis,
usque ad assumptionem beate Marie anno regni ipsius regis
xvii^o. gt dominus Cant, tenebit similiter de baillio domini regis
turrim London, usque ad predictum terminum, salvis civitati
London, libertatibus suis et liberis consuetudinibus suis, et salvo
cuilibet jure suo in custodia turris London., et ita quod interim
non ponat dominus rex munitionem vel vires alias in civitate
predicta vel in turri London. Fiant etiam infra predictum ter-
minum sacramenta per totam Angliam viginti quinque baronibus
sicut continentur in carta de libertatibus et securitate regno con-
^See supra, pp. 43, and 477. The text is taken from New Rymer^ I. 133, on
the authority of Rot. Claus.y 17 John, m. 27 d. It is printed by Blackstone,
Great Charter, 25-6.
496 APPENDIX
cessis vel attornatis viginti quinque baronum sicut continentur
in literis de duodecim militibus eligendis ad delendum malas
consuetudines de forestis et aliis. Et preterea infra eundem
terminum omnia que comites et barones et alii liberi homines
petunt a domino rege que ipse dixerit esse reddenda vel que
per XXV barones aut per majorem partem eorum judicata
fuerint esse reddenda reddantur secundum formam predicte
carte. Et si hec facta fuerint vel per dominum regem non
steterit quo minus ista facta fuerint infra predictum terminum
tunc civitas et turris London, ad eundem terminum statim
reddantur domino regi salvis predicte civitati libertatibus suis
et liberis consuetudinibus suis sicut prescriptum est. Et si
hec facta non fuerint et per dominum regem steterit quod ista
non fiant infra predictum terminum barones tenebunt civitatem
predictam et dominus archiepiscopus turrim London, donee
predicta compleantur. Et interim omnes ex utraque parte
recuperabunt castra terras et villas quas habuerunt in initio
guerre orte inter dominum regem et barones.
(6) Protest by archbishops of Canterbury and Dublin, and other
prelates, that chapter 48 of the Great Charter <was to be
interpreted by both sides as limited.'^
Omnibus Christi fidelibus ad quos presentes littere per-
venerint, Stephanus, Dei gracia Cantuar. archiepiscopus, tocius
Anglie primas et sancte Komane ecclesie cardinalis et H. eadem
gracia, archiepiscopus Dublin., W. quoque London., P. Winton.,
J. Bathon et Glaston., H. Lincoln., W. Wygorn., et W. Coventr.,
ejusdem gracie dono episcopi, salutem in Domino. Cum
dominus Rex concesserit et per cartam suam confirmaverit, quod
omnes male consuetudines de forestis, et forestariis et eorum
ministris, statim inquirantur in quolibet comitatu, per duodecim
milites juratos de eodem comitatu ; qui debent eligi per
probos homines ejusdem comitatus; et infra xl. dies post
inquisitionem factam penitus, ita quod nunquam revocentur,
deleantur per eosdem ; dum tamen dominus Rex hoc prius
sciat; universitati vestre notum fieri volumus, quod articulus
iste ita intellectus fuit ex utraque parte, quum de eo tractabatur,
et expressus, quod omnes consuetudines ille remanere debent,
sine quibus foreste servari non possint : et hoc presentibus
litteris protestamur.
*See supra^ pp. 43, and 440. The protest is recorded in KoL Claus.y 17 John
m. 27 d., and is printed in New Rymer^ I. 134.
SUPPLEMENTARY WRITS 497
(7) Protest by the arclibishops and other prelates that the barons
repudiated their promise to ratify their oaths by .formal
charters A
Omnibus Christi fidelibus etc. Stephanus, Dei gracia Cantuar.
archiepiscopus, totius Anglie primas et sancte Romane ecclesie
cardinalis, Henricus Dublin, archiepiscopus, Willielmus London.,
Petrus Winton., Joscelinus Bathon. et Glaston., Hugo Lincoln.,
Walterus Wigorn., Willielmus Coventr., Ricardus Cicestr.,
episcopi et magister Pandulfus domini Pape subdiaconus et
familiaris, salutem. Noverit universitas vestra, quod quando
facta fuit pax inter dominum regem Johannem et barones Anglie,
de discordia inter eos orta, idem barones, nobis presentibus et
audientibus, promiserunt domino Regi, quod quamcumque securi-
tatem habere vellet ab eis de pace ilia observanda, ipsi ei
habere facerent, preter castella et obsides. Postea vero quando
dominus Rex petiit ab eis, ut talem cartam ei facerent : —
" Omnibus etc. Sciatis nos astrictos esse per sacramenta et
homagia domino nostro Johanni Regi Anglie, de fide ei
servanda de vita et membris et terreno honore suo, contra
omnes homines qui vivere possint et mori ; et ad jura
sua et heredum suorum, et ad regnum suum custodien-
dum et defendendum."
Ipsi id facere noluerunt. Et in hujus rei testimonium id ipsum
per hoc scriptum protestamur.
VIL THE GREAT CHARTER OF HENRY UL^
(third reissue, IITH FEBRUARY, I225.)
Henricus Dei gratia rex Anglie, dominus Hibernie, dux Nor-
mannie, Aquitanie, et comes Andegavie, archiepiscopis, epis-
^ See sitpra, p. 43. The protest is printed in AW. Faf., I. 144 (17 m. 21 d.), and
also in New Rymer^ I. 134.
- This is the definitive form of the Great Charter, as confirmed by Edward I. in
1297 and many times thereafter. See supra^ p. 154. The text is taken from
Stahites of the Realm, I. 22-25. Words not found in the Charter of 1215 are here
printed in itahcs. The footnotes (in preparing which frequent reference has been
made to Bemont's Chartes) give the principal variants occurring in the Charters of
121 5, 1 216, and 1217, as compared with that of 1225. The numbers commonly
used (and here adopted) for the chapters of the issue of 1225 do not agree with
those used for similar chapters of the issue of 1217. The numbers in brackets are
those of corresponding chapters of 1215.
2 I
498 APPENDIX
copis, abbatibus, prioribus, comitibus, baronibus,! vicecomltibus^
prepositis, minlstris et omnibus ballivis et fidelibus suis presen-
tern cartam inspecturis, salutem. Sciatis 2 quod nos, intuitu
Dei et pro salute anime nostre et animarum antecessorum et
successorum nostrorum, ad exaltationem sancte ecclesie et
emendationem regni nostri, spontanea et bona voluntate nostra,
dedimus et concessimus archiepis copis, episcopis, abbatibus,
prioribus, comitibus, baronibus et omiiibus de regno nostro has
libertates subscriptas tenendas in regno nostro Anglie in per-
petuum.
I (i). In primis concessimus ^ Deo et hac present! carta nostra
confirmavimus ^ pro nobis et heredibus nostris in perpetuum quod
anglicana ecclesia libera sit, et habeat omnia ^ jura sua Integra
"^Justiciariis^forestaHis follow baronibus in the issues of 12 1 5 and 1216, both of
which om\t pHoribus.
2 The sentence following " sciatis'''' differs in each preamble. For that of 1215,
see supra^ pp. 185-6.
The Charter of 12 16 reads : Sciatis nos, intuitu Dei et pro salute anime nostre
et omnium antecessorum et successorum nostrorum, ad honorem Dei et exalta-
tionem sancte ecclesie et emendationem regni nostri, per consilium venerabilium
patrum nostrorum domini Gualonis titulo sancti Martini presbiteri cardinalis,
apostolici sedis legati, Petri Wintoniensis, Reineri de Sancto Asapho, Jocelini
Batthoniensis et Glastoniensis, Simonis Exoniensis, Ricardi Cicestriensis, Willelmi
Coventriensis, Benedicti Roffensis, Henrici Landavensis, Menevensis, Bangorensis
et Sylvestri Wygorniensis episcoporum, et nobilium virorum Willelmi Mariscalli,
comitis Penbrocie, Ranulfi comitis Cestrie, Willelmi de Ferrariis comitis Derebie,
Willelmi comitis Albemarle, Huberti de Burgo justiciarii nostri, Savarici de
Maloleone, Willelmi Brigwerre patris, Willelmi Brigwerre filii, Roberti de
Crutenay, Falkesii de Breaute, Reginaldi de Vautort, Walteri de Lascy, Hugonis
de Mortuomari, Johannis de Monemute, Walteri de Bellocampo, Walteri de
Clifford, Roberti de Mortuomari, Willelmi de Cantilupo, Mathei filii Hereberti,
Johannis Mariscalli, Alani Bassett, Philippi de Albiniaco, Johannis Extranei et
aliorum fidelium nostrorum : (i) Imprimis concessisse Deo et hac presenti carta
confirmasse ....
The Charter of 1217 reads: Sciatis quod, intuitu Dei et pro salute anime
nostre et animarum antecessorum et successorum nostrorum, ad exaltationem
sancte ecclesie et emendationem regni nostri, concessimus et hac presenti carta
confirmavimus pro nobis et heredibus nostris in perpetuum, de consilio venerabilis
patris nostri domini Gualonis titulo Sancti Martini presbiteri cardinalis et apostolice
sedis legati, domini Walteri Eboracensis archiepiscopi, Willelmi Londoniensis
episcopi et aliorum episcoporum Anglie, et Willelmi Mariscalli comitis Pembrocie,
rectoris nostri et regni nostri, et aliorum fidelium, comitum et baronum nostrorum
Anglie, has libertates subscriptas tenendas in regno nostro Anglie in perpetuum.
' Concessisse^ in 1215 and 1216. ^ Cojifir?nasse, in 12 15 and 1 216.
^ Omnia t omitted also in 121 5 and 1217.
THE GREAT CHARTER OF HENRY. HI. 499
et libertates suas illesas.^ Concessimus etiam omnibus liberis
hominibus regni nostri pro nobis et heredibus nostris' in per-
petuum omnes libertates subscriptas, habendas et tenendas eis et
heredibus suis de nobis et heredibus nostris in perpetuum.^
2 (2). Si quis comitum vel baronum nostrorum sive aliorum
tenencium de nobis in capite per servicium militare mortuus
fuerit, et, cum decesserit, heres ejus ^ plene etatis fuerit et
relevium debeat, habeat hereditatem suam per antiquum
relevium, scilicet heres vel heredes comitis de baronia comitis
Integra per centum libras, heres vel heredes baronis de baronia
Integra per centum libras, ^ heres vel heredes militis de feodo
militis integro per centum solidos ad plus ; et qui minus debuerit
minus det secundum antiquam consuetudinem feodorum.
3 (3). Si autem heres alicujus talium fuerit infra etatem,^
dominus ejus non habeat custodiam ejus nee terre sue antequam
homagiuni ejus ceperit; et, postquam talis heres fuerit in cus-
todia, cum ad etatem pervenerit, scilicet viginti et unius anni,
habeat hereditatem suam sine relevio et sine fine, ita tamen quod,
si ipse, dum infra etatem fuerit, fiat miles, nichilominus terra
remaneat in custodia dominorum suorum usque ad terminum
predictum.
4 (4). Gustos terre hujusmodi heredis qui infra etatem fuerit
non capiat de terra heredis nisi rationabiles exitus et rationabiles
consuetudines et rationabilia servicia, et hoc sine destructione et
vasto hominum vel rerum ; et si nos commiserimus custodiam
alicujus talis terre vicecomiti vel alicui alii qui de exitibus terre
illius nobis debeat respondere, et ille destructionem de custodia
fecerit vel vastum, nos ab illo capiemus emendam, et terra com-
mittetur^ duobus legalibus et discretis hominibus de feodo illo
qui de exitibus nobis respondeant vel ei cui eos assignaverimus ;
et si dederimus vel vendiderimus alicui custodiam alicujus talis
terre, et ille destructionem inde fecerit vel vastum, amittat ipsam
^For the important clause occurring in I2l$hetweem7/esas and Concessimus,
see sttpruy p. 190.
"^ In perpetuum, omitted also in 1216 and 121 7. ^ Suits, in 121 5.
*The Inspeximus of 1297 reads *^ marcas'" in place of "libras." See supra,
p. 201. Cf. Bracton, II. c. 36, and other authorities cited by Bemont, Chartes,
47 n.
^ Et connects etatem and fuerit in custodia in 12 15, the intervening words being
omitted.
^ Committatur, in 1215, 12 16 and 121 7.
500 APPENDIX
custodiam et tradatur duobus legalibus et discretis hominibus de
feodo illo qui similiter nobis respondeant, sicut predictum est.
5 (5). Custos autem, quamdiu custodiam terre habuerit, sus-
tentet domos, parcos, vivaria, stagna, molendina et cetera ad
terram illam pertinencia de exitibus terre ejusdem, et reddat
heredi, cum ad plenam etatem pervenerit, terram suam totam
instauratam de carucis^ et omnibus aliis rebus, ad minus secun-
dum, quod illatn recepit. Hec omnia observentur de custodiis
archiepiscopatuum, episcopatuum, abbatiarum, prioratuum,
ecclesiarum et dignitatum vacancium que ad nos pertinent,
excepto quod hujusmodi custodie vendi non debent.
6 (6). Heredes maritentur absque disparagatione.2
7 (7). Vidua post mortem mariti sui statim et sine difficultate
aliqua habeat maritagium suum et hereditatem suam, nee aliquid
det pro dote sua vel pro maritagio suo vel pro hereditate sua,
quam hereditatum maritus suus et ipsa tenuerunt^ die obitus
ipsius mariti, et maneat in capitali mesagio mariti sui ^ per quad-
raginta dies post obitum ipsius mariti sui, infra quos assignetur
ei dos sua, nisi prius ei fuerit assignata, vel nisi domus ilia sit
castrum; et si de castro recesserit, statim provideatur ei domus
competens in qua possit honeste morari, quousque dos sua ei
assignetur secundum quod predictum est,^ et habeat rationabile
estoverium suum interim de communi. Assignetur autem ei pro
dote sua tercia pars tocius terre mariti sui que sua fuit in vita
sua, nisi de minori dotata fuerit ad hostium ecclesie.
(8). Nulla vidua distringatur ad se maritandam,^ dum vivere
voluerit sine marito, ita tamen quod securitatem faciet quod se
non maritabit sine assensu nostro, si de nobis tenuerit, vel sine
assensu domini sui,^ si de alio tenuerit.
8 (9). Nos vero vel ballivi^ nostri non seisiemus terram ali-
quam nee redditum pro debito aliquo quamdiu catalla debitoris
presencia sufficiant ^ ad debitum reddendum et ipse debitor para-
^ Different ending in 121 5 ; see supi'a, p. 210.
^ An additional clause occurs in 1215 ; see supra, p. 212.
^ Tenuerint, in 1 2 16 and 1 21 7.
^ In domo mariti sui, in 1215 and 12 16; in capitali viesicagio mariti sui, in
1217.
^The words et habeat to ad hostium ecclesie are omitted also in 12 16.
^Maritandum, in 1216 ; a contraction occurs in 1217, which may stand for
either termination.
"^ De quo tenuerit, in 1 2 15. ^ Nee nos ncc ballivi, in 1215.
^ Sufficitmt, in 12 1 5, 1216 and 1217.
THE GREAT CHARTER OF HENRY HI. 501
tus sit inde satisfacere ; nee plegii ipsius debitoris distringantur
quamdiu ipse capitalis debitor sufficiat 1 ad solutionem debiti ; et,
si capitalis debitor defecerit in solutione debiti, non habens unde
reddat aut reddere nolit cum possit, plegii respondeant pro 2
debito; et, si voluerint, habeant terras et redditus debitoris
quousque ^ sit eis satisf actum de debito quod ante pro eo solve-
runt,^ nisi capitalis debitor monstraverit se inde esse quietum
versus eosdem plegios.
95 (13). Civitas^ Londonie habeat omnes antiquas libertates
et liberas consuetudines suas."^ Preterea volumus et concedi-
mus quod omnes alie civitates, et burgi, et ville, et barones de
quinque portubus, et omnes portus, habeant omnes libertates
et liberas consuetudines suas.
iqS (16). Nullus distringatur ad faciendum majus servicium
de feodo militis nee de alio libero tenemento quam inde debetur.
11 (17). Communia placita non sequantur curiam nostram, set
teneantur in aliquo loco certo.
12 (18). Recognitiones de nova disseisina et de morte ante-
cessoris ^ non capiantur nisi in suis comitatibus, et hoe modo :
nos, vel si extra regnum fuerimus, capitalis justieiarius noster,
mittemus^o justiciarios per unumquemque comitatum semel in
anno,^'^ qui cum militibus comitatuum capiant in comitatibus
assisas predictas. Et ea que in Ulo adventu suo in comitatu per
justiciarios predictos ad dictas assisas capiendas missos terminari
non possunt, per eosdem. terminentur alibi in itinere suo; et ea
que per eosdem propter difjicultatem aliquorum articulorum
term,inari non possunt, referantur ad justiciarios nostros de
banco, et ibi terminentur.
13. Assise de ultima presentatione semper capiantur coram
justiciariis nostris de banco et ibi terminentur.
14 12 (20). Liber homo non amercietur pro parvo delicto nisi
secundum modum ipsius delicti, et pro magno delicto, secundum
'^Stifficit, in 1215. ^De^ in 1215. ^ Donee, in 1215.
^'Solverint, in 1215, 1216 and 1217.
5 Three additional chapters (10, 11 and 12) occur in 12 15.
^ Et civitasy in 121 5. ' Tarn per terras., quam per aquas, in 1215.
'^Two additional chapters (14 and 15) occur in 1215.
^ Et de ultima presentacione, added in 12 1 5. ^^ Duos justiciarios, in 1215.
^^ Fer quatuor vices in anno, in 1215, which concludes somewhat differently, see
supra, p. 296. The charter of 1216 is practically the same, here, as that of 12x5 ;
while that of 1225 reproduces that of 12 17.
*2 An additional chapter (19) occurs in 121 5.
502 APPENDIX
magnltudinem delicti, salvo contenemento suo; et mercator
eodem modo salva mercandisa sua; et villanus alterius quam
noster^ eodem modo amercietur salvo wainagio suo, si inci-
dent 2 in misericordiam nostram : et nulla predictarum miseri-
cordiarum ponatur nisi per sacramentum ^ proborum et legalium
hominum de visneto.
(21). Comites et barones non amercientur nisi per pares suos,
et non nisi secundum modum delicti.
(22) A Nulla ecclesiastica persona amercietur secundum quanti-
tatem beneficii sui ecclesiastici, set secundum laicum tenementum
suum, et secundum quantitatem delicti.
15 (23). Nee villa, nee homo, distringatur facere pontes ad
riparias nisi qui ex antiquo et de jure facere debet. ^
16.6 Nulla riparia decetero defendatur, nisi ille que fuerunt
in defense tempore regis Henrici avi nostri, per eadem loca et
eosdem terminos sicut esse consueverunt tempore suo.
17 (24). Nullus vicecomes, constabularius, coronatores vel alii
ballivi nostri teneant placita corone nostre.
18^ (26). Si aliquis tenens de nobis laicum feodum moriatur,
et vicecomes vel ballivus noster ostendat litteras nostras patentes
de summonitione nostra de debito quod defunctus nobis debuit,
liceat vicecomiti vel ballivo nostro attachiare et inbreviare catalla
defuncti inventa in laico feodo ad valenciam illius debiti per
visum legalium hominum, ita tamen quod nichil inde amoveatur
donee persolvatur nobis debitum quod clarum fuerit, et residuum
relinquatur executoribus ad faciendum testamentum defuncti ; et
si nichil nobis debeatur ab ipso, omnia catalla cedant defuncto,
salvis uxori ipsius et pueris suis ^ rationabilibus partibus suis.
^Alterius quavi noster^ omitted also in 1216 ; first inserted in 1217.
^ Inciderint^ in 1215. ^ Sacramenta^ in 1217.
*This reads in 1215 : Nullus dericus amercietur de laico tenemento suo, nisi
secundum modum aliorum predictorum, et non secundum beneficii sui ecclesi-
astici.
In 1216 : Nullus dericus amercietur^ nisi secttndum fo7'mam predictorum, et
non secundum quantitatem beneficii sui ecclesiastici.
The Charter of 1217 is here identical with that of 1225, except that j-^a? takes the
place of set.
^Debent, in 1 2 15.
* Compare with last clause of c. 47 of 1215, which was omitted in 1216. The
Charter of 121 7 here resembles that of 1225.
'An additional chapter (25) occurs in 1215 ; see supra^ p. 317, which should be
compared with c. 35 of 1225.
•Omitted in 12 16.
THE GREAT CHARTER OF HENRY HI. 503
19 1 (28). Nullus constabularius vel ejus ballivus 2 capiat blada
vel alia catalla alicujus qui non sit de villa ubi castrum situm
est, nisi statim inde reddat denarios aut respectum inde habere
possit de voluntate venditoris ; si autem de villa ipsa fuerit, infra
quadraginta dies precium reddat.^
20 (29). Nullus constabularius distringat aliquem militem ad
dandum denarios pro custodia castri, si ipse earn facere voluerit*
in propria persona sua, vel per alium probum hominem, si ipse
earn facere non possit propter rationabilem causam, et, si nos
duxerimus eum^ vel miserimus in exercitum, erit quietus de
custodia secundum quantitatem temporis quo per nos fuerit in
€xercitu de feodo pro quo fecit servicium in exercitu.^
21 (30). Nullus vicecomes, vel ballivus noster, vel alius ^ capiat
equos vel carettas alicujus ^ pro cariagio faciendo, nisi ^ reddat
liberationem antiquitus statutam, scilicet pro caretta ad duos
equos decern denarios per diem, et pro caretta ad tres equos
quatuordecim denarios per diemA^ Nulla caretta dominica ali-
cujus ecclesiastice persone vel militis vel alicujus domine capia-
tur per ballivos predictos.
(31). Nee nos nee ballivi nostri nee alii^^ capiemus alienum
boscum ad castra vel alia agenda nostra, nisi per voluntatem
illius 12 cujus boscus ille fuerit.
22 (32). Nos non tenebimus terras eorurnl i^ qui convicti
fuerint de felonia, nisi per unum annum et unum diem ; et tunc
reddantur terre dominis feodorum.
23 (33)- Omnes kidelli decetero deponantur penitus per Tami-
siam et Medeweiam i^ et per totam Angliam, nisi per costeram
maris.
^ An additional chapter (27) occurs in 1215 ; see su/>ra, p. 326.
^ Alius ballivus noster^ in 12 15, which omits ejus.
3 In 1216 the last phrase reads : si atitem de villa fuerit ^teneatur injra tres
septimanas precium reddere.
*^ Si facere voluerit ctistodiam illam^ in 121 5.
^ Eum comes after viiseHmus in 121 5.
•'The words in italics here are omitted also in 1216.
' Aliquis alius^ in 1215. ^Alicujus liberi kominis, in 1215.
^ Nisi de voluntate ipsitis liberi hominis^ in 1 21 5.
^^This chapter is not in 12 16, but first occurs in 1217.
^'^ Nee aliiy omitted also in 12 16. '^'^Ipsius^ in place oi illius, in 1215.
'^^ Illorum, in place oi eorurn, in 1 21 5. ^^De Tamisia, et de Medewaye, in 1215.
504 APPENDIX
24 (34). Breve quod vocatur Precipe decetero non fiat alicui
de aliquo 1 tenemento, unde liber homo perdat 2 curiam suam.
25 (35). Una mensura vini sit per totum regnum nostrum, et
una mensura cervisie, et una mensura bladi, scilicet quarterium
London., et una latitudo pannorum tinctorum et russettorum
et haubergettorum, scilicet due ulne infra listas ; de ponderibus
vero 3 sit ut de mensuris.
26 (36). Nichil detur^ de cetero pro brevi inquisitionis ab eo
qui inquisitionem petit ^ de vita vel membris, set gratis conce-
datur et non negetur.
27 ^37)' Si aliquis teneat de nobis per feodifirmam vel socca-
gium, vel per burgagium, et de alio terram teneat per servicium
militare, nos non habebimus custodiam heredis nee terre sue que
est de feodo alterius, occasione illius feodifirme, vel soccagii, vel
burgagii, nee habebimus custodiam illius feodifirme vel soccagii
vel burgagii, nisi ipsa feodifirma debeat servicium militare. Nos
non habebimus custodiam heredis nee ^ terre alicujus quam tenet
de alio per servicium militare, occasione alicujus parve serjan-
terie quam tenet de nobis per servicium reddendi nobis cultellos,
vel sagittas, vel hujusmodi.
28 (38). Nullus ballivus ponat decetero aliquem ad legem
tnanifestam vel ad juramentum simplici loquela sua, sine testibus
fidelibus ad hoc inductis.
29 (39). Nullus liber homo decetero capiatur vel inprisonetur
aut disseisiatur de aliquo libero tenemento suo vel lihertatibus
vel liberis consuetudinibus suis^'^ aut utlagetur, aut exuletur aut
aliquo alio modo destruatur, nee supier eum ibimus, nee super
eum mittemus, nisi per legale judicium parium suorum, vel per
legem terre.
(40). NuUi vendemus, nulli negabimus aut differemus rectum
vel^ justiciam.
30 (41). Omnes mercatores, nisi publice antea prohibiti fue-
rint, habeant salvum et securum exire de Anglia, et venire in
Angliam, et morari, et ire per Angliam tam per terram quam
per aquam ^ ad emendum vel ^^ vendendum sine omnibus toltis
'^Libera follows aliquo^ in 1297.
^ Amittere possit, in 1215; omittere possit^ in 1216. ^ Auteni, in 1^15.
* Vel capiatur^ after detur, in 1215.
^The words in italics are omitted also in 12 16.
^ Et for nec^ in 1216 ; vel in 1217.
'The words in italics are omitted also in 1216. ^ Aut^ in 1215.
^ Aquas, in 1216. ^^ Et^ in 121 5 and 1217.
THE GREAT CHARTER OF HENRY HI. 505
malis 1 per antiquas et rectas consuetudines, preterquam in
tempore gwerre, et si sint de terra contra nos gwerrina; et si
tales inveniantur in terra nostra in principio gwerre, attachien-
tur sine dampno corporum vel rerum, donee sciatur a nobis vel
a capitali justiciario nostro quomodo mercatores terre nostre
tractentur, qui tunc invenientur in terra contra nos gwerrina ;
et, si nostri salvi sint ibi, alii salvi sint in terra nostra.
31 2 (43). Si quis tenuerit de aliqua escaeta, sicut de honore
Wallingefordie, Bolonie, Notingeham, Lancastrie, vel de aliis^
que sunt in manu nostra, et sint baronie, et obierit, heres ejus
non det aliud relevium nee fiat^ nobis aliud servicium quam
faceret baroni, si ipsa 5 esset in manu baronis ; et nos eodem
modo earn tenebimus quo baro eam tenuit; nee nos, occasione
talis baronie vel escaete, habebimus aliquam escaetam vel cus'
todiam aliquorum hominum nostrorum, nisi alibi tenuerit de
nobis in capite ille qui tenuit baroniam vel escaetam.^
32.'^ Nullus liber homo decetero det amplius alicui vel vendat
de terra sua quam ut de residuo terre sue possit sufficienter
fieri domino feodi servicium ei debitum quod pertinet ad feodum
illud.
33 s (46). Omnes patroni abbatiarum qui habent cartas regum
Anglie de advocatione, vel antiquam tenuram vel possessionem,
habeant earum custodiam cum vacaverint, sicut habere debent,
et sicut supra declaratum est.
34 9 (54). Nullus capiatur vel imprisonetur propter appellum
femine de morte alterius quam viri sui.
1 Ma/is toltis^ in 1 2 1 5 .
2 An additional chapter (42) occurs in 1215.
^Eskaetis, after aliisy in 12 15.
^ Faciat, ioxjiat, in 121 5.
^ Si baronia ilia esset ^ in 1215 ; si terra ilia esset ^ in 1216 ; si ilia esset, in 12 17.
^The words in italics are omitted also in 12 16.
'Two additional chapters (44 and 45) occur in 1215.
8 The charter of 121 5 reads: onuies barones qtii fundaverunt abbatias, unde
habent cartas regwii Anglie, vel antiquam tenuram, habeant eartim custodiam
cum vacaverint, sicut habere debent.
^ Seven additional chapters (47 to 53) occur in 1215. The charter of 1216 omits
CO. 32 and 33 of 1225, but has instead of them three chapters of which c. 36 is the
same as c. 44 of 121 5, except that stmt appears for sint', c. 37 is the same as c. 46
of 121 5, as quoted in last note, except that the words et sictit stipi-a declaratum est
are added at the end ; c. 38 is of the same tenor as c. 47 of 121 5, except that John's
tempore nostro is changed to meet the altered circumstances.
5o6 APPENDIX
35.^ Nullus comitatus decetero teneatur, nisi de mense in
mensem; et, ubi major terminus esse solehat, major sit. Nee
aliquis vicecomes vel ballivus faciat turnum suum per hundredum
nisi bis in anno et non nisi in loco debito et consueto, videlicet
semel post Pascha et iterum post festum sancti Michaelis. Et
visus de franco plegio tunc fiat ad ilium terminum sancti
Michaelis sine occasione, ita scilicet quod quilibet habeat liber-
tates suas quas habuit et habere consuevit tempore regis Henrici
avi nostri, vel quas postea perquisivit. Fiat autem visus de
franco plegio sic, videlicet quod pax nostra teneatur, et quod
tethinga integra sit sicut esse consuevit, et quod vicecomes non
querat occasiones, et quod contentus sit eo quod vicecomes
habere consuevit de visu suo faciendo tempore regis Henrici
avi nostri.
36. Non liceat alicui decetero dare terram suam alicui domui
religiose, ita quod eam resumat tenendam de eadem domo, nee
liceat alicui domui religiose terram alicujus sic accipere quod
tradat illam ei^ a quo ipsam recepit ^ tenendam. Si quis autem
de cetero terram suam alicui domui religiose sic dederit, et super
hoc convincatur, donum suum penitus cassetur, et terra ilia
domino suo illius feodi incurratur.
37. Scutagium decetero capiatur sicut capi solehat^ tempore
regis Henrici avi nostri. Et salve ^ sint archiepiscopis, epis-
copis, abbatibus, prioribus, templariis, ho spit alar iis , comitibus,
baronibus et omnibus aliis tarn ecclesiasticis quam secularibus
personis Ubertates et libere consuetudines quas prius habuerunt.
{60). Omnes autem istas consuetudines predictas et libertates
quas concessimus in regno nostro tenendas quantum ad nos
pertinet erga nostros, omnes de regno nostro tam clerici quam
laid observent quantum ad se pertinet erga suos.^ Pro hac
^The charter of 1216 (like that of 1215) omits cc. 35, 36, and 37 of 1225 (all of
which occur in 1217). C. 35 of 1225, however, should be compared with c. 25 of
1215, and c. 37 of 1225 with c. 12 of 1215.
The charter of 12 15 has five additional chapters here (55 to 59), of which c. 56
appears in 1216 as c. 40, with the alteration made necessary by the new reign.
"^ Eavi illi (in place oi illam et), in 1217.
^ Receperit (in place oi recepit)^ in 121 7.
* Consuevit (in place oi solebat), in 12 17.
*The last sentence, from Et salve onwards, is omitted in 1217.
« After erga stios, each charter proceeds differently. The charter of 1215 has
three additional chapters (61, 62, and 63, g.v.\
The charter of 12 16 has one additional chapter : Quia vero quedam capitula
in priori {sic) carta continebantur que gravia et dubitabilia videbantur, scilicet de
THE GREAT CHARTER OF HENRY HI. 507
autem concessione et donatione libertatum istaruni et aliarum
libertatum contentarum in carta nostra de libertatihus foreste,
archiepiscopi, episcopi, abhates, priores, comites, 'barones,
milites, libere tenentes, et omnes de regno nostro dederunt nobis
quintam decimam partem omnium mobilium suorum. Con-
cessimus etiam eisdem pro nobis et heredibus nostris quod nee
nos nee heredes nostri aliquid perquiremus per quod libertates
in hac carta contente infringantur vel infirmentur ; et, si de
aliquo aliquid contra hoc perquisitum fuerit, nichil valeat et
pro nullo habeatur.
Hiis testibus domino Stephano Cantuariensi archiepiscopo,
Eustachio Lundoniensi, Jocelino Bathoniensi, Petro Winto-
niensi, Hugoni Lincolniensi, Ricardo Sarrisberiensi, Benedicto
Roffensi, Willelmo Wigorniensi, Johanne Eliensi, Hugone
Herefordiensi, Radulpho Cicestriensi, Willelmo Exoniensi epis-
copis, abbate sancti Albani, abbate sancti Edmundi, abbate de
Bello, abbate sancti Augustini Cantuariensis, abbate de Eves-
hamia, abbate de W estmonasterio , abbate de Burgo sancti Petri,
abbate Radingensi, abbate Abbendoniensi, abbate de Maume-
buria, abbate de Winchecomba, abbate de Hida, abbate de
Certeseia, abbate de Sireburnia, abbate de Cerne, abbate de
scutagiis et auxiliis assidendis, de debitis Judeorum et aliorum et de libertate
exeundi de regno nostro, vel redeundi in regnum, et de forestis et forestariis,
warrenis et warrenariis, et de consuetudinis comitatum, et de ripariis et earum cus-
todibus, placuit supradictis prelatis et magnatibus ea esse in respectu quousque
plenius consilium habuerimus ; et tunc faciemus plenissime tarn de hiis quam de
aliis que occurrerint emendenda, que ad communem omnium utilitatem pertinuerint
et pacem et statum nostrum et regni nostri. Quia vero sigillum nondum habuimus,
presentem cartam sigillis venerabilis patris nostri domini Gualonis tituli sancti
Martini presbiteri cardinalis, apostolice sedis legati, et Willelmi Mariscalli comitis
Penbrocie, rectoris nostri et regni nostri, fecimus sigillari, Testibus omnibus pre-
nominatis et aliis multis, Datum per manus predictorum domini legati et Willelmi
Mariscalli, comitis Penbrocie apud Bristollum duodecimo die novembris anno
regni nostri primo.
The charter of 12 17 has two additional chapters (46 and 47) :
C. 46 : Salvis archiepiscopis, episcopis, abbatibus, prioribus, templariis, hospi-
talariis, comitibus, baronibus, et omnibus aliis, tam ecclesiasticis personis, quam
secularibus, libertatihus et liberis consuetudinibus quas prius habuerunt — (which
may be compared with the last clause of c. 37 of 1225, supra).
C. 47 : Statuimus etiam, de communi consilio tocius regni nostri, quod omnia
castra adulterina, videlicet ea que a principio guerre mote inter dominum
Johannem patrem nostrum et barones suos Anglie constructa fuerint vel reedi-
ficata, statim deruantur. Quia vero nondum habuimus sigillum, banc sigillis
domini legati predicti et comitis Willelmi Mariscalli rectoris et regni nostri
fecimus sigillari.
5o8 APPENDIX
Abbotebiria, abbate de Middletonia, abbate de Seleby, abbate
de Wyteby, abbate de Cirencestria, Huberto de Burgo justiciario,
Ranulfo comite Cestrie et Lincolnie, Willelmo comite Sarrisberie,
Willelmo comite Warennie, Gilberto de Clara comite Gloucestrie
et Hertfordie, Willelmo de Ferrariis comite Derbeie, Willelmo
de Mandevilla comite Essexie, Hugone Le Bigod comite Nor-
folcie, Willelmo comite Aubemarle, Hunfrido comite Herefordie,
Johanne constabulario Cestrie, Roberto de Ros, Roberto filio
Walteri, Roberto de Veteri ponte, Willielmo Brigwerre, Ricardo
de Munfichet, Petro filio Herberti, Matheo fUio Herberti,
Willielmo de Albiniaco, Roberto Gresley, Reginaldo de Brahus,
Johanne de Munemutha, Johanne filio Alani, Hugone de Mortuo-
mari, Waltero de Bellocampo, Willielmo de sancto Johanne,
Petro de Malalacu, Briano de Insula, Thoma de Muletonia,
Ricardo de Argentein., Gaufrido de Nevilla, Willielmo Mauduit,
Johanne de Baalun.
Datum apud Westmonasterium undecimo die februarii anno
regni nostri nono.
VIII. CARTA DE FORESTA.i
(6 NOVEMBER, 12 I7.)
Henricus Dei gratia rex Anglie, dominus Hibernie, dux
Normannie, Aquitanie et comes Andegavie, archiepiscopis, epis-
copis, abbatibus, prioribus, comitibus, baronibus, justiciariis,
forestariis, vicecomitibus, prepositis, ministris, et omnibus balli-
vis et fidelibus suis, salutem. Sciatis quod, intuitu Dei et pro
salute anime nostre et animarum antecessorum et successorum
nostrorum, ad exaltacionem Sancte Ecclesie et emendacionem
regni nostri, concessimus et hac presenti carta confirmavimus pro
nobis et heredibus nostris in perpetuum, de consilio venerabilis
patris nostri domini Gualonis tituli sancti Martini presbiteri
cardinalis et apostolice sedis legati, domini Walteri Eboracensis
archiepiscopi, Willelmi Londoniensis episcopi, et aliorum epis-
coporum Anglie, et Willelmi Marescalli comitis Penbrocie,
rectoris nostri et regni nostri, et aliorum fidelium comitum et
baronum nostrorum Anglie, has libertates subscriptas tenendas
in regno nostro Anglie, in perpetuum :
I. In primis omnes foreste quas Henricus rex avus noster
^See supra, pp. 146-7. The text is taken from the Statutes of the Realm,
I. 20-21. Bemont, Chartes, 64flf., gives in footnotes the variants in the reissue
of 1225.
CARTA DE FORESTA 509
afforestavit videaiitur per bonos et legales homines ; et, si boscum
aliquem alium quam suum dominicum afforestaverit ad damp-
num illius cujus boscus fuerit, deafforestentur. Et si boscum
suum proprium afforestaverit, remaneat foresta, salva communa
de herbagio et aliis in eadem foresta, illis qui eam prius habere
consueverunt.
2. Homines qui manent extra forestam non veniant decetero
coram justiciariis nostris de foresta per communes summoni-
ciones, nisi sint in placito, vel plegii alicujus vel aliquorum qui
attachiati sunt propter forestam.
3. Omnes autem bosci qui fuerunt afforestati per regem
Ricardum avunculum nostrum, vel per regem Johannem patrem
nostrum usque ad primam coronacionem nostram, statim de-
afforestentur, nisi fuerit dominicus boscus noster.
4. Archiepiscopi, episcopi, abbates, priores, comites et barones
et milites et Hbere tenentes, qui boscos suos habent in forestis,
habeant boscos suos sicut eos habuerunt tempore prime corona-
cionis predicti regis Henrici avi nostri, ita quod quieti sint in
perpetuum de omnibus purpresturis, vastis et assartis factis in
illis boscis, post illud tempus usque ad principium secundi anni
coronacionis nostre. Et qui de cetero vastum, purpresturam, vel
assartum sine licencia nostra in illis fecerint, de vastis et assartis
respondeant.
5. Reguardores nostri eant per forestas ad faciendum reguar-
dum sicut fieri consuevit tempore prime coronacionis predicti
regis Henrici avi nostri, et non aliter.
6. Inquisicio, vel visus de expeditacione canum existencium in
foresta, decetero fiat quando debet fieri reguardum, scilicet de
tercio anno in tercium annum ; et tunc fiat per visum et testi-
monium legalium hominum et non aliter. Et ille, cujus canis
inventus fuerit tunc non expeditatus, det pro misericordia tres
solidos; et de cetero nullus bos capiatur pro expeditacione.
Talis autem sit expeditacio per assisam communiter quod tres
ortilli abscidantur sine pelota de pede anteriori ; nee expedi-
tentur canes de cetero, nisi in locis ubi consueverunt expeditari
tempore prime coronacionis regis Henrici avi nostri.
7. Nullus forestarius vel bedellus decetero faciat scotale, vel
colligat garbas, vel avenam, vel bladum aliud, vel agnos, vel
porcellos, nee aliquam collectam f aciant ; et per visum et sacra-
mentum duodecim reguardorum quando facient reguardum, tot
forestarii ponantur ad forestas custodiendas, qudt ad illas cus-
todiendas rationabiliter viderint sufiicere.
510 APPENDIX
8. Nullum suanlmotum de cetero teneatur in regno nostro
nisi ter in anno; videlicet in principio quindecim dierum ante
festum Sancti Michaelis, quando agistatores conveniunt ad agis-
tandum dominicos boscos nostros ; et circa festum Sancti Martini
quando agistatores nostri debent recipere pannagium nostrum ;
et ad ista duo suanimota conveniant forestarii, viridarii, et agis-
tatores, et nullus alius per districtionem ; et tercium suanimotum
teneatur in inicio quindecim dierum ante festum Sancti Johannis
Baptiste, pro feonacione bestiarum nostrarum ; et ad istud
suanimotum tenendum convenient forestarii et viridarii et nulli
alii per districtionem. Et preterea singulis quadraginta diebus
per totum annum conveniant viridarii et forestarii ad videndum
attachiamenta de foresta, tarn de viridi, quam de venacione, per
presentacionem ipsorum forestariorum, et coram ipsis attachiatis.
Predicta autem suanimota non teneantur nisi in comitatibus in
quibus teneri consueverunt.
g. Unusquisque liber homo agistet boscum suum in foresta
pro voluntate sua et habeat pannagium suum. Concedimus
eciam quod unusquisque liber homo possit ducere porcos suos
per dominicum boscum nostrum, libere et sine inpedimento,
ad agistandum eos in boscis suis propriis, vel alibi ubi voluerit.
Et si porci alicujus liberi hominis una nocte pernoctaverint in
foresta nostra, non inde occasionetur ita quod aliquid de suo
perdat.
10. Nullus de cetero amittat vitam vel menbra pro vena-
cione nostra; set, si aliquis captus fuerit et convictus de
capcione venacionis, graviter redimatur, si habeat unde redimi
possit; et si non habeat unde redimi possit, jaceat in prisona
nostra per unum annum et unum diem ; et, si post unum
annum et unum diem pleglos invenire possit, exeat a prisona;
sin autem, abjuret regnum Anglie.
11. Quicunque archiepiscopus, episcopus, comes vel baro
transient per forestam nostram, liceat ei capere unam vel duas
bestias per visum forestarii, si presens fuerit ; sin autem, faciat
cornari, ne videatur furtive hoc facere.
12. Unusquisque liber homo decetero sine occasione faciat in
bosco suo, vel in terra sua quam habeat in foresta, molendinum,
vivarium, stagnum, marleram, fossatum, vel terram arabilem
extra cooperatum in terra arabili, ita quod non sit ad nocumen-
tum alicujus vicini.
13. Unusquisque liber homo habeat in boscis suis aereas,
ancipitrum et spervariorum et falconum, aquilarum, et de hey-
CARTA DE FORESTA 511
rinis et habeat similiter mel quod inventum fuerit in boscis
suis.
14. Nullus forestarius de cetero, qui non sit forestarius de
feudo reddens nobis firmam pro balliva sua, capiat chiminagium
aliquod in balliva sua ; forestarius autem de feudo firmam nobis
reddens pro balliva sua capiat chiminagium, videlicet pro careta
per dimidium annum duos denarios, et pro equo qui portat
sumagium per dimidium annum unum obolum, et per alium
dimidium annum obolum, et non nisi de illis qui de extra
ballivam suam, tanquam mercatores, veniunt per licenciam suam
in ballivam suam ad buscam, meremium, corticem vel carbonem
emendum, et alias ducendum ad vendendum ubi voluerint : et
de nulla alia careta vel sumagio aliquod chimunagium capiatur :
et non capiatur chiminagium nisi in locis illis ubi antiquitus
capi solebat et debuit. Illi autem qui portant super dorsum
suum buscam, corticem, vel carbonem, ad vendendum, quamvis
inde vivant, nullum de cetero dent chiminagium. De boscis
autem aliorum nullum detur chiminagium foristariis nostris,
preterquam de dominicis bocis nostris.
15. Omnes utlagati pro foresta tantum a tempore regis
Henrici avi nostri usque ad primam coronacionem nostram,
veniant ad pacem nostram sine inpedimento, et salvos plegios
Inveniant quod de cetero non forisfaciant nobis de foresta
nostra.
16. Nullus castellanus vel alius teneat placita de foresta
sive viridi sive de venacione, sed quilibet forestarius de feudo
attachiet placita de foresta tam de viridi quam de venacione,
et ea presentet viridariis provinciarum et cum irrotulata fuerint
et sub sigillis viridariorum inclusa, presententur capitali fores-
tario cum in partes illas venerit ad tenendum placita foreste, et
coram eo terminentur.
17. Has autem libertates de forestis concessimus omnibus,
salvis archiepiscopis, episcopis, abbatibus, prioribus, comitibus,
baronibus, militibus et aliis tam personis ecclesiasticis quam
secularibus, Templariis et Hospitalariis, libertatibus et liberis
consuetudinibus in forestis et extra, in warennis et aliis, quas
prius habuerunt. Omnes autem istas consuetudines predictas
et libertates, quas concessimus in regno nostro tenendas quan-
tum ad nos pertinet erga nostros, omnes de regno nostro tam
clerici quam laici observent quantum ad se pertinet erga suos.
Quia vero sigillum nondum habuimus, presentem cartam sigillis
venerabllis patris nostri domini Gualonis tituli Sancti Martini
512 APPENDIX
presbiteri cardinalis, apostolice sedis legati, et Willelmi Mares-
calli comitis Penbrok, rectoris nostri et regni nostri, fecimus
sigillari. Testibus prenominatis et aliis multis. Datum per
manus predictorum domini legati et Willelmi Marescalli apud
Sanctum Paulum London., sexto die Novembris, anno regni
nostri secundo.
SELECT BIBLIOGRAPHY AND LIST OF
AUTHORITIES REFERRED TO.
I. COMMENTARIES AND OTHER WORKS ON MAGNA CARTA
(CHRONOLOGICALLY ARRANGED).
The Mirror of Justices, edited by Whittaker, W. J. (Selden Society) ;
1895.
Coke, Sir Edward, Second Institute, 1641 ; 17th edition, 1817.
Cooke, Edward, Magna Charta made in the ninth year of King Henry III.
and confirmed by King Edward I. in the twentieth year of his reign ;
1684.
Blackstone, Sir William, The Great Charter and Charter of the Forest, to
which is prefixed the History of the Charters ; 1759.
Barrington, Daines, Observations upon the Statutes from Magna Charta to
21 James I. ; 1766.
Sullivan, F. S., An Historical Treatise on the Feudal Law, with a Com-
mentary on Magna Charta ; 1772.
Thomson, Richard, An Historical Essay on the Magna Charta of King John ;
1829.
Lau, Thaddaeus, Die Entstehungsgeschichte der Magna Charta ; 1856.
Bemont, Charles, Chartes des Libertes Anglaises ; 1892.
Hantos, Elem6r, The Magna Carta of the English and of the Hungarian
Constitution ; 1904.
II. CHRONICLES AND ANNALS.
Annals of Dunstable, edited by H. R. Luard (Rolls Series) ; 1866.
Annals of Waverley, edited by H. R. Luard (Rolls Series) ; 1865.
Benedict Abbot, Gesta Regis Henrici Secundi, edited by William Stubbs
(Rolls Series) ; 1867.
Histoire de Guillaume le Marichal, edited by Paul Meyer ; 1891.
Histoire des dues de Normandie et des rois d'Angleterre, edited by F. Michel ;
1840.
Jocelyn of Brakelond, Chronica de rebus gestis Samsonis Abbatis Monasterii
Sancti Edmundi, edited by J. G. Rokewode (Camden Society) ; 1840.
Matthew Paris, Chronica Majora, edited by H. R. Luard (Rolls Series) ;
1872.
Memorials of St. Dunstan, edited by WilHam Stubbs (Rolls Series) ; 1874.
Ralph of Coggeshall, Chronicon Anglicanum, edited by Joseph Stevenson
(Rolls Series) ; 1875.
2 K
514 BIBLIOGRAPHY
Roger of Hoveden, Chronica, edited by William Stubbs (Rolls Series) ;
1868-1871.
Roger of Wendover, Chronica sive Flores Historiarum, edited by H. O.
Coxe (Eng. Hist. Society) ; 1841.
Walter of Coventry, Memoriale, edited by William Stubbs (Rolls Series) ;
1872.
Walter of Hemingburgh, Chronicon de Gestis Regum Angliae, edited by
H. C. Hamilton (Eng. Hist. Society) ; 1848-9.
William of Malmesbury, Gesta Regum Anglorum, edited by William Stubbs
(Rolls Series) ; 1887-9.
III. COLLECTIONS OF STATUTES, CHARTERS, AND
TREATIES.
Statutes of the Realm (Record Commission) ; 1810-28.
Statutes at Large.
Die Gesetze der Angelsachsen, edited by F. Liebermann ; 1 898-1912.
Acts of the Parliament of Scotland from 1124 to 1707, edited by Thomas
Thomson and Cosmo Innes ; 1814-75.
Rotuli Litter arum Clausarum in Turri Londinensi Asservata, edited by
Thomas Duffus Hardy (Record Commission) ; 1833.
Rotuli Litter arum Patentum in Turri Londinensi Asservata, edited by T. D.
Hardy (Record Commission) ; 1835.
Rotuli Chartarum in Turri Londinensi Asservata, edited by T. D. Hardy
(Record Commission) ; 1837.
Rotuli de Oblatis et Finibus, edited by T. D. Hardy (Record Commission) ;
1835-6.
Rotuli Parliamentorum ; 1832.
Rotuli Hundredorum (Record Commission) ; 1812-18.
Testa de Neville sive Liber Feodorum (Record Commission) ; 1807.
The Red Book of the Exchequer, edited by Hubert Hall (Rolls Series) ; 1896.
Munimenta Gildhallae Londoniensis : Liber Albus, Liber Custumarum et
Liber Horn, edited by H. T. Riley (Rolls Series) ; 1859-62.
Rymer, Thomas, Foedera, Conventiones, Litterae, et cujuscunque generis
acta publica ; 4th edition (Record Commission) ; 1816-69 (referred to
throughout as " New Rymer "),
Ancient Charters, Royal and Private, edited by J. H. Round (Pipe Roll
Society, vol. 10) ; 1888.
D'Achery, J. L., Vetorum Scriptorum Spicilegium ; 1655-77.
Hemingi Chartularum Ecclesiae Wigornensis, edited by Thomas Hearne ;
1723-
Potthast, A., Regesta Pontificum Romanorum ; 1874-5.
Teulet, Alexandre, Layettes du Trisor des Charles ; 1863-1902.
Stubbs, William, Select Charters and other Illustrations of English Con-
stitutional History ; 7th edition, 1890.
Prothero, G. W., Select Statutes and other Constitutional Documents illus-
trative of the reigns of Elizabeth and James I. ; 1894.
Gardiner, S. R., The Constitutional Documents of the Puritan Revolution ;
1889.
Birch, W. de G., Historical Charters and Constitutional Documents of the
City of London ; 1887.
BIBLIOGRAPHY 515
IV. COLLECTIONS OF PLEAS, TRIALS, AND OTHER RECORD
EVIDENCE.
Placitorum Abbreviatio, Richard I. to Edward II. (Record Commission) ;
1811.
Bigelow, M. M., Placita Anglo-Normannica ; 1879.
Bracton's Note Book : a Collection of Cases, edited by F. W. Maitland ;
1887.
Howell, T. B and T. J., Complete Collection of State Trials ; 1809-28
(referred to as " State Trials ").
Pleas of the Crown for the Comity of Gloucester, edited by F. W. Maitland ;
1884.
Select Pleas of the Crown, edited by F. W. Maitland (Selden Society) ;
1888.
Select Pleas in Manorial and other Seignorial Courts, edited by F. W. Mait-
land (Selden Society) ; 1889.
Select Pleas of the Forest, edited by G. J. Turner (Selden Society) ; 1901.
Select Pleas, Starrs, and other Records from the Rolls of the Exchequer of the
Jews, edited by J. M. Rigg (Selden Society) ; 1902.
Year Books of the Reign of Edward I., edited by A. J. Horwood and L. O.
Pike (Rolls Series) ; 1863-1901.
Year Books of Edward II., 1307- 1309, edited by F. W. Maitland (Selden
Society) ; 1903,
Great Roll of the Pipe for the Twelfth Year of Henry II. (Pipe Roll Society,
vol. 9) ; 1888.
Madox, Thomas, History and Antiquities of the Exchequer of the Kings of
England ; 2nd edition, 1769 (referred to throughout as " Madox ").
Madox, Thomas, Firma Burgi ', 1726.
Madox, Thomas, Baronia Anglica ; 1741.
V. LEGAL TREATISES— MEDIEVAL.
Glanvill, Ranulf, Tractatus de Legibus et Consuetudinibus Regni Angliae.
Richard, son of Nigel, De necessariis Observantibus Scaccarii Dialogus (com-
monly called Dialogus de Scaccario), edited by A. Hughes, C. G. Crump,
and C. Johnson ; 1902.
Bracton, Henry de, De legibus et consuetudinibus Angliae, edited by Sir
Travers Twiss (Rolls Series) ; 1878-83.
Fleta, Commentarius Juris Anglicani ; edition of 1647.
Littleton, Thomas, Treatise of Tenures \ edition of 1841.
VI. LEGAL TREATISES— MODERN.
Anson, Sir W. R., The Law and Custom of the Constitution ; 2nd edition,
1892.
Blackstone, Sir William, Commentaries on the Laws of England ; edition
of 1826.
Coke, Sir Edward, Institutes of the Laws of England ; 17th edition, 1817.
(The First Institute is generally referred to as " Coke on Littleton.")
Encyclopaedia of the Laws of England, edited by A. W. Renton ; 1897-8^
Hale, Sir Matthew, Historia Placitorum Coronae ; 1 736.
Jenks, Edward, Modern Land Law ; 1899.
2K2
5i6 BIBLIOGRAPHY
Manwood, John, A Treatise and Discourse of the Laws of the Forest ; 1598.
Stephen, H. J., Commentaries on the Laws of England ; 13th edition, 1899.
Thayer, J. B., A Preliminary Treatise on Evidence at the Common Law ;
VII. LEGAL AND CONSTITUTIONAL HISTORIES.
Bigelow, M. M., History of Procedure in England ; 1880.
Brunner, Heinrich, Die Entstehung der Schwurgerichte ; 1871.
Creasy, Edward, Progress of the English Constitution ; 1874.
Gneist, Rudolf, The History of the English Constitution, translated by
P. A. Ash worth ; edition of 1891.
Gneist, Rudolf, The English Parliament in its Transformations through a
Thousand Years, translated by A. H. Keane ; 1887.
Holdsworth, W. S., A History of English Law, vol. i ; 1903.
Medley, D. ]., A Student's Manual of English Constitutional History ; 2nd
edition, 1898.
Moore, S. A. and H. S., The History and Law of Fisheries ; 1903.
Pollock, Sir F,, and Maitland, F. W., The History of English Law before
the time of Edward I. ; ist edition, 1895 (referred to throughout as
" Pollock and Maitland ").
Pike, L. O., A Constitutional History of the House of Lords, from original
sources ; 1894.
Reeves, John, History of English Law ; 3rd edition, 1783-4.
Stephen, Sir J. F., A History of the Criminal Law in England ; 1893.
Stubbs, William, The Constitutional History of England in its Origin and
Development : (a) vol. i, 6th edition, 1897 ; (&) vol. 2, 4th edition,
1894 ; (c) vol. 3, 5th edition, 1896.
Taswell-Langmead, T. P., English Constitutional History from the Teutonic
Conquest to the Present Time ; 5th edition, 1896.
Taylor, Hannis, The Origin and Growth of the English Constitution ; 1898.
VIII. GENERAL HISTORIES.
Brady, Robert, Complete History of England ; 1685.
Care, Henry, English Liberties in the Freeborn Subjects' Inheritance ; 17 19.
Green, J. R., A Short History of the English People ; edition of 1875.
Henry, Robert, History of Great Britain ; 6th edition, 1806.
Lingard, John^ A History of England to 1688 ; 1819-30.
Mackintosh, James, History of England; edition of 1853.
Smith, Gold win, The United Kingdom : a Political History ; 1899.
Tyrrell, James, History of England, 1697-1704.
IX. HISTORIES OF SPECIAL PERIODS.
Bateson, Mary, Mediaeval England (Story of the Nations Series) ; 1903.
Freeman, E. A., The Norman Conquest of England ; 1870-9.
Freeman, E. A., The Reign of William Rufus ; 1882.
Gardiner, S. R., History of England from the Accession of James I. to the
Outbreak of the Civil War ; 1883-4.
Hallam, Henry, View of the State of Europe during the Middle Ages ; 7th
edition, 1837.
BIBLIOGRAPHY 517
Kemble, J. M., Saxons in England ; 1849.
Norgate, Kate, England under Angevin Kings ; 1887.
Norgate, Kate, John Lackland ; 1902.
Norgate, Kate, The Minority of Henry III. ; 191 2.
Pearson, Charles, A History of England during the Early and Middle Ages ;
1867.
Petit-Dutaillis, Charles, ^tude sur la vie et la regne de Louis VIII. ; 1894.
Powicke, F. M., The Loss of Normandy ', 191 3.
Prothero, G. W., The Life of Simon de Montfort, Earl of Leicester ; 1877.
Ramsay, Sir J. H., The Foundations of England ; 1898.
Ramsay, Sir J. H., The Angevin Empire ; 1903.
X. MISCELLANEOUS.
Adams, G. B., The Origin of the English Constitution ; 191 2.
Brady, Robert, A Full and Clear Answer ; 1683.
Boutmy, fimile, Etudes de Droit Constitutionnel ; 1885.
Burke, Edmund, Works ; edition of 1837,
Chadwick, H. Munro, Studies on Anglo-Saxon Institutions ; 1905.
Dowell, Stephen, History of Taxation and Taxes in England ; 1884.
Gross, Charles, Preface to Select Cases from the Coroners' Rolls (Selden
Society) ; 1896.
Hall, Hubert, History of the Customs Revenue in England ; 1885.
Harcourt, L. W, V,, His Grace the Steward and Trial of Peers ; 1907.
Hearnshaw, F. J, C, Leet Jurisdiction in England ; 1908.
Lapsley, G. T., The County Palatine of Durham ; 1900.
Luard, H. R., Preface to vol. 2 of Matthew Paris, Chronica Majora (Rolls
Series) ; 1872.
Luchaire, Achille, Communes Frangaises ; 1890.
Luffman, John, Charters of London ; 1793.
Neilson, George, Trial by Combat ; 1890.
Noorthouck, John, A New History of London ; 1773.
Macy, J., The English Constitution ; a Commentary on its nature and growth ',
1897.
Maitland, F. W., Township and Borough ; 1898.
Maitland, F. W., in Social England, edited by Henry Duff Trail, vol. i ;
ist edition, 1893.
Maitland, F. W., Preface to Select Pleas of the Crown (Selden Society) ;
1888.
Maitland, F. W., Preface to Select Pleas in Manorial and other Seignorial
Courts (Selden Society) ; 1889.
Maitland, F. W., Preface to The Mirror of Justices (Selden Society) ; 1895.
Maitland, F. W., Collected Papers, 3 vols. ; 191 1.
Maitland, F. W., Equity ; 1909.
Orpen, G. H., Ireland under the Normans ; 1911.
Petit-Dutaillis, Charles, Studies and Notes Supplementary to Stubbs' Con-
stitutional History, translated by W. E. Rhodes ; 1908.
Pollock, Sir F., Essays in Jurisprudence and Ethics ; 1894.
Poole, R. L., The Exchequer in the Twelfth Century ; 1912.
5i8 BIBLIOGRAPHY
Rigg, J. M., Preface to Select Pleas, Starrs, and other Records from the Rolls
of the Exchequer of the Jews (Selden Society) ; 1902.
Rossler, Oskar, Kaiserin Mathilde und das Zeitalter der Anarchie in Eng-
land ; 1897.
Round, J. H., editorial notes to Ancient Charters, Royal and Private (Pipe
Roll Society, vol. 10) ; 1888.
Round, J. H., Geoffrey de Mandeville : a Study of the Anarchy ; 1892.
Round, J. H., Feudal England : Historical Studies of the Eleventh and
Twelfth Centuries ; 1895.
Round, J. H., The Commune of London and other Studies ; 1899.
Round, J. H., Peerage and Pedigree ; 19 10.
Round, J. H., The Kings' Serjeants and Officers of State ; 191 1.
Seebohm, Frederic, The English Village Community : an Essay on Economic
History ; 1883.
Stubbs, William, Preface to Walter of Coventry, Memoriale (Rolls Series) ;
1972.
Turner, G. J., Preface to Select Pleas of the Forest (Selden Society) ; 1901.
VinogradofE, Paul, Villainage in England : Essays in English Mediaeval
History ; 1892.
Vinogradoff, Paul, The Growth of the Manor; 1905.
Vinogradoff, Paul, English Society in the Eleventh Century ; 1908.
XI. REPORTS, BIBLIOGRAPHIES, AND DICTIONARIES.
Reports from the Lords' Committee appointed to search the Journals of the
House, Rolls of Parliament, and other Records for all matters touching
the Dignity of a Peer ; ist Report, 1820.
Reports from the Select Committee appointed to inquire into the state of the
Public Records of the Kingdom (Record Commission) ; 1900.
Report on Manuscripts in Various Collections (Historical Manuscripts Com-
mission) ; 1 90 1.
Gross, Charles, The Sources and Literature of English History ; 1900.
Watt, Robert, Bibliotheca Britannica ; 1824.
Lowndes, W. T., The Bibliographer's Manual of English Literature ;
1857-64.
Dictionary of National Biography, edited by Leslie Stephen and Sidney
Lee ; 1885-1900.
INDEX OF STATUTES.
20 Henry III.
PAGE
c. 6, - 63, 213.
c. 7, - - 64 n.
c. 8, - - 64 n.
c. 9, - - 432 n.
c. 11, 210 n., 423 n.
52 Henry III.
c. 15, - - 223 n.
c. 16, 64 n., 211,
275 n.
c. 29, - - 355 n.
3 Edward I.
c. 1, - - 311 n.
c. 6, - - 293.
c. 7. - - 332 n.
c. 10, - - 316.
c. 11. - - 365.
c. 12, - 341, 375 n.
c. 21, - - 209.
c. 32, 330. 332 n..
335 n.
c. 36, - - 66.
c. 48, - - 211.
4 Edward I.
(Rageman), - 281 n.
6 Edward I.
c. 4, - 55, 219 n.
c. 5, - - 209.
c. 9, 364 n., 365 n.
7 Edward I.
- 150 n.
13 Edward I.
c. 5, - - 276 n.
c. 12, - - 365 n.
c. 13, - - 309 n.
c. 29, 361 n., 365 n.
c. 30, - 277 n., 283.
c. 39, - - 281.
c. 47, 210 n., 304 n.
18 Edward I.
- 150 n.
27 Edward I.
c. 3, 281.
28 Edward I.
c. 4, - - 268 n.
c. 5, - - 267 n.
c. 7, - - 315 n.
c. 12, - - 223.
c. 14, - - 319.
Stat. 3, c. 13, 311 n.
Statutes of uncertain date :
Statute of
Jewry, - 229, 231.
Praerogativa Regis, 339 n.
Consuetudines et Assisae de
foresta,
- 430.
1 Edward III. stat. 2, c. 1, 438 n.
c. 13,4130.
c. 17, 3090.
2 Edward III. c. 2, - - 281 n.
c. 9, 400 n., 406 n.
c. 11, - - 263 n.
4 Edward III. c. 15, - - 320.
5 Edward III. c. 9, - - 380 n.
9 Edward III. c. 1, - - 407 n.
14 Edward III. stat. 1, c. 9, 320.
c. 21, 400 n.
stat. 2, - 238.
stat. 4, c. 4, 212.
c. 5, 212.
25 Edward III. stat. 3, c. 4, 344.
stat. 4, c. 7, 407 n.
stat. 5, c. 4, 380-1 n.
c. 11, 66.
27 Edward III. stat. 2, c. 8, 378 n.
28 Edward III. c. 13, - - 378 n.
37 Edward III. c. 9, - - 364 n.
c. 18, - - 381 n.
38 Edward III. c. 3, - - 381 n.
42 Edward III. c. 1, - 159, 364 n.
c. 3, - - 381 n.
2 Richard II. stat. 1, c. 1, 407 n.
6 Richard II. c. 5, • - 284 n.
c. 6, - - 453 n.
11 Richard II. c. 7, - - 407 n.
c. 11, - - 284 n.
17 Richard II. c. 6, - - 381 n.
c. 9, - * 345 n.
I Henry IV. c. 14, - - 366 n.
4 Henry IV. c. 5, - - 320 n.
c. 11, - - 345 n.
5 Henry IV. c. 10, - - 315 n.
I Edward IV. c. 2, - - 310 n.
12 Edward IV. c. 7, - - 345 n.
I Richard III. c. 8, - 356-7 n.
3 Henry VII. c. 1 (s. 11), - 366 n.
32 Henry VIII. c. 46, - - 211.
I Edward VI. c. 4, - - 413.
3 Charles I. c. i, 333, 381 n., 394.
16 Charles I. c. 16, 430 n., 438.
12 Charles II. c. 24, 55, 211, 331 n.
520
INDEX OF STATUTES
13 Charles II. c. 8, 331 n., 335 n.
12 George III. c. 20, - - 342 n.
54 George III. c. 145, - 342 n.
57 George III. c. 61, - - 43011.
59 George III. c. 46, - 362 n., 367.
7 and 8 George IV., c. 28, - 342 n.
3 and 4 William IV. c. 27, - 272 n.
c. 106 (s. 10), 342 n.
5 and 6 William IV. c. 76 (s. 14),
407 n.
14 and 15 Victoria, c. 42, - 431 n
15 and 16 Victoria, c. 76, - 371 n.
33 and 34 Victoria, c. 14, - 411.
c. 23, - 342 n.
50 and 51 Victoria, c. .35, - 306 n.
INDEX.
Abbeys (rights of founders of),
433-5. 451-
Advowson, 276, 434. (See also
Presentment.)
Aesop, 4.
Agistors, 419.
Aids (feudal), 59, 65-7, 232-4, 248,
256-60, 261.
Aids (general), 234-9,
Alexander II. (King of Scots), 213,
459, 461-3-
Aliens. (See Foreigners.)
Allegiance, 409, 410-11.
Amercements, 71, 118, 284-99,
453-6-
Anglo-Saxon period, 4-5, 6, 8.
Anselm, 17.
Anson, Sir William R., 107 n.,
249 n., 253 n.
Appeal (to a higher court), 8, 12.
Appeal (or accusation), 89, 135-6,
316, 360. 365-7, 451-3.
Arbitrary imprisonment. (See Im-
prisonment.)
Archbishops. (See Dunstan, Lan-
franc, Anselm, Becket, Hubert
Walter, Stephen Langton.)
Aristotle, 5.
Arthur (John's nephew), 188, 207,
442, 461 n.
Arthur (King), 24.
Articles of the Barons ; impressed
with John's seal, 37-8.
relation to schedule of 27 April,
1215, 37.
relation to " unknown charter,"
175-
relation to Magna Carta, 37, 39,
41, 42 n., 196, 198, 236, 248,
282-3, 308, 382, 449, 451, 454,
469, 477-
leave some points undecided,
38-9.
selfish in motive, 119.
omits provisions of Charter, 129-
130, 191, 248, 451.
history of MS. of, 170-1.
as affecting the Church, 191.
on relief, 198.
on wardship, 213.
on rights of London, 234, 236.
on tallage, 236.
on petty assizes, 282-3.
on coroners, 308.
on judgment of peers, 377, 382,
449. 458-
on estates unjustly disseised, 383,
449.
on unjust fines, 454-5.
on rights of Welshmen, 458.
on rights of King of Scots, 462 n.
on " the form of security," 468 n.
text of, 487-493.
Articuli super Cartas, 223, 319.
Ashford v. Thornton, 366-7.
Assize (various meanings of), 272 n.
Assize of Arms, 12.
Assize of Clarendon, 250 n., 340.
on ordeal, 60, 373.
on Crown pleas, 88, 339.
on jury of accusation, 135.
on chattels of felons, 339, 340.
article 4 of, 373.
article 5 of, 339.
Assize of Measures, 316 n., 356-8.
Assize of Northampton, 271 n.,
278 n.
on Crown pleas, 88.
on jury of accusation, 135, 339.
on wardship, 206.
on mort d' ancestor, 275.
on powers of castellans, 315.
article 4 of, 206, 275, 278 n.
article 12 of, 315.
Assize of Wine, 356-8.
Assize of Woodstock, 286 n., 415-6,
418, 426, 427 n,, 428.
Assize (Grand), 90, 137, 272-6, 353.
Assizes (Petty), 90-2, 137-8, 269-83,
347. (See also Justice of
Assize.)
Attainder, 60, 342.
522
INDEX
Bailiffs. 304, 316-7, 319, 369-75-
definition of, 316-7.
duties of, 369-75.
abuses by, 329, 369, 431.
type of men to be appointed, 431.
Baronia, meanings of, 200, 251 n.
Barons (legal position of), 85, 130-1,
196, 200, 251.
amercement of, 295-8.
Barons (minor), 251-3.
Barons (under John and his son),
their motives, 49, 51, 116-7, 119.
their grievances, 48-92, 389.
their policy, 51, 67-8, 88, 120, 389.
their factions, 189.
Barons' War (against Henry III.),
142-3, 405.
Barony, 200. (See also Honour.)
Barrington (Daines), 179, 208 n.,
217.
Bateson, Miss Mary, 56, 241 n.,
360 n.
Battle. (See Trial by combat.)
Bee, Abbey of, 17.
Becket (Thomas a), 13, 192-3, 250.
Bemont, Charles, 177, 180.
Bench. (See King's Bench and
Common Pleas.)
Benefit of clergy. (See Clergy.)
Bigelow, M. M., 268 n., 349 n.,
370 n., 377 n.. 379 n.
Bishops.
as holders of baronies, 17, 62, 68,
202, 211.
election of, 17-19, 141, 194, 213-4.
wardship over their lands, 62,
97-8, 103, 192-3, 211-2, 434.
liable for scutage, 70.
not for relief, 202-3, 211-2,
Blackstone, Sir William.
on the sensation caused by Char-
ter of Henry I., 28 n.
on the barons' diffidatio, 34.
compares John's Charter with
Articuli, 39.
on forest clauses in John's char-
ter, 44.
on antecedents of Charter, 47.
on tenure, 53.
on feudal incidents, 59 n., 64.
on nature of Magna Carta, iii,
113-
on loss of John's seal, 144 n.
on different versions of Charter,
155 n., 168, 171 n., 210 n., 278,
386 n.
his book, 176, 179.
on estovers, 219 n.
on justices of assize, 271 n., 278.
on c. 33, 344 n.
on c. 39, 376 n., 386 n.
on feudal appeal, 360 n.
on forest courts, 430.
on executors of charter, 469 n.
Blench tenure (in Scotland), 56 n.
Bloodfeud, 284-5.
Boroughs, 234-48.
monopoly of, 357.
privileges of, 240-1, 357, 404-7.
(See also under Firma.)
Bot, 285.
Boutmy, fimile, 106.
Bouvine (battle of), 30.
Bracton, 208, 214, 219 n., 230, 293,
297» 329, 340. 360 n., 364 n.,
413 n., 452 n.
Breaute, Falkes de, 446.
Bridges, obligation to repair, 299-
304-
Brewer, William, 153.
Burgage tenure, 57, 62, 367-9.
Burgess, 115, 213-4. (See also
Merchant.)
Burgh, Hubert de.
at Runnymede, 36. I
becomes Justiciar, 42.
Canonical election, 18, 25, 32. (See
also Bishops.)
Canute (King), 6, 14.
Capitulary of Kiersey, 59.
Carta de foresta. (See Forest.)
Cartae (of 1166), 12.
Castellans. (See Constables.)
Castle-guard (tenure of), 333-4.
Castles (private), 218, 442.
Castles (royal), 226-7, 333-4-
Central government, problem of,
13-16.
Champions (in duellum), 451.
King's, 56.
Chancery, 11, 89, 342.
Court of, 264-5, 410.
Charter, John's Great.
date when sealed, 37-41.
historical antecedents of, 3-5, 48-
92.
its prototypes, 93-104.
historical sequel to, 139-64.
form of, 104-9.
contents of, 109-20.
characteristics of, 109-20.
how far a baronial manifests,
1 1 3-6.
its value, 120-9.
its defects, 129-31.
its traditional interpretations,
132-3-
ho.w^ far declaratory, 11 1-2,
310-27 345-<^.
INDEX
523
Charter, John's Gredit.—conid.
relations to Henry II. 's reforms.
1 1 1-2, 270-8.
how far reactionary, 112-3. '
its practical nature, 120, 131 n.,
376.
attitude towards Church and
clergy, 116, 190-4, 408-11, 434.
towards boroughs and traders,
42, 117, 234-9, 240-8, 398-407-
towards tenants of mesne lords,
1 16-7, 255-60, 41 1-4, 463-5.
towards villeins, 113, 118-9.
towards lower classes generally,
113-4, 118.
its sanction, 115-6, 129-31, 465-
70.
variations from articles of Barons,
39, 116-7, 130, 191, 234, 236,
248, 282-3, 299. 308, 451, 458-9.
exaggerated estimates of, 120-3.
four copies of, 41.
manuscripts of, 165-75.
restored by Prince Louis, 146.
(See also Table of Contents.)
Charter, the so-called " unknown
charter of John." (See Un-
known.)
Charter of Henry I. ; granting of,
97-
exhibited by Langton, 28, 32.
Blackstone on, 28 n., 48.
a model for Magna Carta, 32,
95-
its general tenor, 97.
its chief provisions, 97-101.
on rights of church, 203.
on marriage, 98.
on wardships, 98, 203, 206.
on reliefs, 197, 203.
on amercements, 286.
on forests, 424.
on rights of sub -tenants, 464.
text of, 481.
Charter of Henry II., loi.
text of, 485.
Charter, Henry III.'s first reissue
(1216), 139-44-
its additions, 204, 218, 223 n.,
434-
its omissions, 141-4, 228, 231, 233,
236, 255, 258.
its respiting clause, 143.
on homage, 204.
on wardship, 204, 212, 434.
on Church and clergy, 212, 434.
on marriage, 213.
on widows' quarantine, 218.
on debtors, 223 n.
on Jews, 228, 231.
on aids and scutages, 233, 236,
255, 258.
on merchants, 404.
Charter, Henry III.'s second reissue,
(1217), 145-52.
its omissions, 130, 151, 326.
its additions, 147-51, 216, 219,
331, 440, 465.
reactionary clauses of, 148-9.
on widows' rights, 216, 219.
on scutages, 148-9, 233.
on petty assizes, 277-8.
on villeins, 119, 292, 386.
on clergy, 299.
on Commune Concilium, 130.
on purveyance, 331, 335.
on forests, 302.
on amercements, 299.
Charter, Henry III.'s third reissue
(1225), 152-7.
the final form of Magna Carta,
155-
price paid for, 154-5.
text of, 497-508.
Charter of Stephen, 101-3, 188.
text of, 483-5.
Chase, distinguished from " forest,"
422.
Chatham, Lord, 113.
Chattels, forfeit of felon's, 339.
in intestacy, 326-9.
Church, English.
growth of conception of National
Church, 1 91 -2.
gains power under Stephen, 18.
Henry II. 's struggle with, 13.
relations to State, 16, 17, 194, 324.
relations to Rome, 16, 17, 194,
408-9.
freedom of, 18, 32, 97, 102, 116,
141, 191-4.
national character of, 192.
question of investitures, 17.
influence upon the Great Charter,
191, 326-9, 408.
relations to Crown, 96, 194, 408.
relations to John, 32, 116.
John's charter to, 32.
relations to barons, 50.
its control of wills and property
of intestates, 326-9.
canonical election. (See Bishops.)
(See also Clergy.)
Cigogne, Engelard de, 417 n., 444-7.
Clarendon. (See Constitutions of
and Assize of.)
Clergy, amercement of, 298-9.
their right to go to Rome, 407-11.
" benefit of clergy," 18, 214 n.
Cnut. (See Canute.)
524
INDEX
Coinage, private, 8.
Coke (Chief Justice).
on tenures, 54, 57 n.
on primer seisin, 65 n.
on Magna Carta, 113, 133, 178,
181, 190, 464.
on c. 39, 381, 385, 386.
on c. 40, 398.
on c. 54, 453.
on villeins, 118, 287 n., 386.
on Knights' fees, 201.
on waste, 209 n., 338.
on disparagement, 214.
on rights of widows, 217, 219.
on justices of assize, 280 n.
on attainder, 342.
his unhistorical method, 178, 385.
Combat. (See Trial by combat.)
Comitatus. (See Shire, Shire Court,
and also Earldom.)
Commissions of Justices, 280-2.
Comniittee of Executors. (See Exe-
cutors.)
Common law, growth of, 12.
courts of, 263-9.
Common pleas, 90, 261-9.
Court of, Ti, 90, 265-7.
Commune, 116, 243-5, 471.
Commune Concilium, 231-40, 248-
55.
functions of, 130-1, 163, 254-5.
composition of, 250-5.
Commune of London. {See London.)
Compurgation, 85, 370.
Confirmatio Cartarum, 69, 238.
Constables, different uses of word,
313-5-
their functions and powers, 314,
329, 333-
and purveyance, 329-31.
and castle-guard, 333.
Constitutions of Clarendon, 18-19,
202 n., 211, 298-9, 408.
Contenement, 284.
meaning of, 293-4.
Copyhold, 53.
Coronation oath, 95-6.
Coroner, 16, 304, 315.
his functions, 315-6.
Council of St. Albans (1213), 28, 432.
Council of Oxford (1213), 30.
Councils, Lateran. (See Lateran.)
County. (See County Court, also
Firma Comitatus.)
County Court, 77, 81, 282-3.
to appoint knights to reform
abuses, 42, 439-40.
to appoint knights for assizes,
277-8.
Courts, three rival systems of, 77-92.
Courts, Christian, 323-4.
Courts, local. (See County and
Hundred.)
Courts, royal. (See Curia Regis,
Chancery, Common pleas. Ex-
chequer, King's Bench.)
Criminal law. (See Pleas of the
Crown.)
Crusaders, 449-50.
effects of vow, 33, 449.
Curia Regis.
appeals to, 8.
altered by Henry I., 9.
under Henry II., 11, 50.
powers of, 188-9, 262, 264, 388.
relation to Witenagemot, 188,
250.
Danish influence, 6.
Darrein presentment. (See Pre-
sentment.)
David I. (King of Scots), 460.
Debtors, 223-5, 321-3.
and Court of Exchequer, 268.
Diffidatio.
nature and effects of, 34.
by barons on 5th May, 12 15, 34.
Disparagement, 212-3, 461.
Disseisin, 119, 274-5,' 383, 386, 450.
Disseisin (Assize of Novel), 269,
274-5. 277.
Domesday Book, 10.
Dominus, meaning of, 187-9.
Dower, 215-7, 230-1, 454.
Duellum. (See Trial by combat.)
Dunstan, 14, 95.
Dymoke (Family of), 55.
Earldormen, 8, 14.
Earls, amercement of, 130, 295-8.
definition of, 201.
relief due from, 196-202.
members of Commune Concilium,
248-51.
relation to county, 270 n.
Earls Palatine, 7, 15, 64 n.
Edgar (King), 14.
Edward Confessor, 6, 7, 14.
his laws confirmed, 27, 32, 11 1-2.
Edward I.
his reforms, 159-64.
his attitude to the Charter, 161 -4.
his Parliaments, 162-3.
his Inspeximus of the Charter,
168-9, 198 n., 299 n.
quarrels with Constable and
Marshal, 69, 261.
relations with London, 405.
his influence on the Constitution,
162-4, 475-
INDEX
525
Edward II., 406, 430, 437.
Edward III., 238, 303, 332, 406,
409, 438.
Election, meaning of word, 278.
in relation to kingship, 118-9.
of jurors, 277-8.
of knights to reform abuses, 42,
438-40.
of knights to hold assizes, 278.
England, growth of a united, 4-6.
English Church. (See Church.)
English law. (See Law.)
English Monarchy. (See Mon-
archy.)
Escheat, 59-61, 336-9, 41 1-3.
Estovers, definition of, 218-9.
widows' rights of, 218.
of firewood, 426.
Exchequer, under Henry I., 9-10.
Court of, 263-5, 267-9.
Executors of the Charter, 465-77.
references to, 40, 42, 44.
their excesses, 44.
their local agents, 440.
their powers, 449, 455, 465-77.
Eyres. (See Justices of Eyre.)
Falconry. (See Foivling.)
Fee-farm, 55, 57, 368-9.
Felony, 336-43-
Feudal aids. (See Aids.)
Feudal grievances, 58-92, 121.
Feudal incidents, 59-65. (See also
Reliefs, Escheats, Wardships,
Marriages, Primer Seisins, and
Fines for Alienations.)
Feudal jurisdictions. (See under
Jurisdictions and under Courts.)
Feudal obligations, 52-76, 116-7.
(See also Services, Incidents,
and Aids.)
Feudal services, 12, 52-3, 67-9, 199,
260-1.
Feudal tenures. (See Tenures.)
Feudahsm in England, 7-9, 52-3, 54,
57-
Fictions (legal), 269 n.
Fines, 454-6.
differ from amercements, 292-3.
Fines for alienations, 65.
Firma burgi. 241-2, 320, 445.
Firma comitates, 317-21.
Fishing, 303-4, 343-6.
Fitz-Aylwin, Mayor of London, 34.
Fitz-Peter, Geoffrey, 29.
Fitz-Walter, Robert.
escapes from John, 25.
reinstated, 26.
commands barons' army, 34.
outlawed, 384.
one of the executors of Charter,
469.
Flambard, Ralph.
his feudal innovations, 9, 58, 203,
206.
Foreign merchants.
their position in London, 247.
favoured by Magna Carta, 400-7.
Foreign service, 69, 260-1.
Forest Charters, issued in 121 7, 146.
its repudiation feared, 155-6.
provisions of, 425, 426, 428, 430,
436.
text of, 508-12.
Forest Courts, 420-2, 427.
Foresters, 417-20, 428-30.
Forests, 414-31, 435-40.
protests by prelates regarding, 43.
text of protests, 496-7.
boundaries of, 156, 437.
Fowling, King's rights of, 299-304,
435-8.
Frankalmoin, 55, 298-9.
Free socage. (See Socage.)
Freehold, 53-8.
Freeman, definition of, 114-5, 195.
287 n., 299 n., 386.
rights of, 284, 287-9, 334-5, 346-7.
375 ff-
obligations of, 300.
Fyrd, 12, 300.
Godwin, 14.
Government, ideal form of, 5.
" mixed," 5.
(See also Local goi^ernnient.)
Grand Assize. (See Assize.)
Grand Jury. (See Jury.)
Grand Serjeanty. (See Serjeanty.)
Great Charter. (See Charter.)
Green, John Richard, on Richard
I.'s reign, 21.
Grey, John de, 23.
Gualo (papal legate), 46, 47, 139,
144. 145-
Guardian, 219. (See also Ward-
ship.)
Guilds, 400, 404, 407.
Habeas corpus. (See Writ of.)
Harden (Wilham), Mayor of
London, 245 n., 469.
Harold, King, 7.
Hawking. (See Fowling.)
Henry I.
his character, 9.
his title, 96, 196.
reforms, 9-10, 15.
relations with church, 17, 192,
203.
526
INDEX
Henry I.
and canonical election, i8.
his attitude towards the forests,
424.
scutage under, 70.
charter to Jews, 227.
charter to London, 241-2.
organizes Exchequer, 9.
his Charter of Liberties. (See
Charter.)
Henry II.
his character, 11.
and Becket, 19, 250.
policy in regard to church, 18-19,
211.
military reforms, 12.
legal reforms, 11-12, 88-92, 273,
306, 347.
administrative reforms, 11 -12,
19-20.
his new legal procedure, 11-12,
88-92.
opposes hereditary principle,
15-
discourages trial by combat, 273,
360-1.
undermines private jurisdictions,
12, 273, 347.
takes aids, 66.
levies scutages, 69-70.
founder of trial by jury, 134-8,
273-
charter to London, 242-3.
opens royal courts to freeholders,
12, 77, 80-1, 87, 395.
Henry III.
his coronation, 139.
his minority, 143, 266.
his advisers accept the Charter,
47. 139.
takes aids, 66.
levies scutage, 143, 258.
takes fees for writs, 259, 397.
his attitude to Magna Carta,
153-9, 160.
his attitude to the Jews, 228.
protects foreign merchants, 405.
his alliance with Rome, 328 n.
his rights of fowling, 303.
his attitude towards the forests,
437-8.
declared of age, 153.
(See also under Charter.)
Honorius III., 153.
Hostages, 25, 441-4, 458, 459.
Hubert de Burgh. (See Burgh.)
Hubert, Walter (Archbishop).
his reforms, 21, 307.
his testament, 324 n.
his death, 23.
Hundred, 317-8.
Court of, 78-9.
Imprisonment (arbitrary), 133,
401-2.
Incidents. (See Feudal incidents.)
Indictment, 340-1.
Innocent III. nominates Stephen
Langton, 23-4.
releases John's subjects from
allegiance, 24, 25.
excommunicates John, 26.
his reconciliation with John, 26.
appealed to by malcontents, 33.
supports John, 33, 46, 462, 477.
suggested as umpire by John,
34-
annuls the Great Charter, 45.
excommunicates barons, 46.
interferes with Scots King, 462.
Inquest of Service (12 12), 24, 75-6.
Inquest of Sheriffs. (See Sheriff.)
Inquisitio. (See Recognitio.)
Interest, 223-5. (See also Jews and
Usury.)
Interregnum (theory of), 96.
Investitures (question of), 17.
Isabel, Princess. (See Scotland.)
Ivo, Bishop of Chartres, 17.
James I., 58 n., 336.
Jarls (Danish), 15.
Jews, 223-31, 378.
Joan (John's natural daughter), 25.
John, his character, 48.
his bad faith, 42-3.
his power, 24.
his policy, 21, 48, 50.
his incapacity, 22.
his title to the Crown, 186-9.
his exactions, 22, 31, 49.
his straits for money, 32.
his levies of scutage, 31, 69-76,
charters to English Church, 32,
33, 194, 211-2.
charters to London, 34, 244-5,
345-
relations with his barons, 24, 29,
31. 34. 49.
refuses their demands, 33.
meets barons at Runnymede,
36-41.
takes Crusader's vow, 33, 46.
relations with English Church, 23,
24, 50, 212.
relations with Rome, 23.
the quarrel, 24.
the reconciliation, 26, 50.
his surrender to Innocent, 26, 29,
50.
INDEX
527
John asks Innocent to annul char-
ter, 44.
hangs Welsh hostages, 25, 442-3.
loses Normandy, 22.
confiscates church property, 24.
restores it, 28.
issues writs for a Council in Nov.
1213, 29.
and county representation, 29.
his death, 47.
Judgment (in medieval law), 84-6.
Judgment of peers. (See Peers.)
Jurisdiction, royal, 80-1. (See also
Courts.)
Jurisdictions, private, 78-80. (See
also Courts.)
Jury, trial by.
relation to recognitio, 87, 135.
relation to writ de odio, 89, 136,
361-2.
relation to Magna Carta, 134-8.
Jury (of accusation or presenting
jury), 135-7.
Jury (Civil), 137-8.
Jury (Grand), 136.
Jury (Petty), 136-8, 360-1.
Justice, three systems of, 77-81.
(See also Courts and Juris-
dictions.)
Justices of the Peace, 16, 281.
Justices, itinerant (or of Eyre),
270-1.
Justices of Assize, 270-83.
commissions of, 280-1.
Kiersey, capitulary of, 59.
King's Bench, 11, 263-7.
Knight's fee, 199-200, 260.
Knight's service, 54, 199.
Ky dells- (See Weirs.)
Lanfranc, 17, 96.
Langton. (See Stephen.)
Lateran Council (1099), 17.
Lateran Council (1215), 46, 341.
Law (lex), technical meaning of, 84,
369-70, 379-81. (See also
Trial.)
Leasehold, 53-6.
Legal fictions. (See Fictions.)
Legal procedure. (See Procedure.)
Leofric, 14.
Letters testimonial, 41.
text of, 478 n.
Llywelyn, 25, 458.
I-ocal government, problem of,
13-16.
London.
receives rebel army, 35.
accepts Louis as King, 47.
liable to tallage, 117, 234-9.
granted by John as security, 42,
477-
John's charters to, ^^4, 244-7, 345-
taxation of, 231-2, 234-9.
privileges of, 240-8, 345.
Longchamp, 51.
Lords (House of), 392.
Louis, son of French king, 46, 140,
145-
confirms Magna Carta, 47.
Magna Carta. (See Charter.)
Manorial courts, 77-80, 87. (See
also Courts and Jurisdictions.)
Maritagium, 216-8, 454.
Marriage, feudal incident of, 62 3,
212-4.
Marshal (William, Earl Marshal),
32, 33. 35-
mediator between King and
barons, 33, 35, 36, 476.
acts as Regent, 139-41, 145, 255.
Matilda (daughter of Henry I.).
supplanted by Stephen, lo-ii.
her prodigal grants, 137.
Merchants. (See Foreign merchants,
London, and Boroughs.)
Ministerial responsibility, doctrine
of, 129, 161-3, 433.
Minorities, rights of, 254-5.
Monarchy, two types of, 4-5.
Monarchy (English),
growth of, 5-13.
strong under Normans, 5.
and Angevins, 13.
weak under Anglo-Saxon kings,
4-5. 13-
Monopolies.
attitude of Magna Carta towards,
384.
claimed by native merchants,
117. 398-407-
Montfort, Simon de, 161 -2.
M or td' ancestor, 269-70, 275-6, 283 n.
Municipalities. (See Boroughs.)
National Church. (See Church.)
National unity in England, 5-6.
Nicholas (Papal legate), 311.
Norgate, Miss Kate.
on writs of November 121 3, 29.
on date of Magna Carta, 41 n.
on John's misconduct, 48 n.
on John's levies of scutage, 74 n.,
7.5 n.
on privileges of London, 245.
on the twenty-five executors, 476.
on the price paid for the charter
of 1225, 154 n.
528
INDEX
Norman Conquest, effects of, 5.
Novel disseisin. (See Disseisin.)
Oligarchic elements in England, 4.
Ordeal, 84-5, 136, 369-75.
Original writs. (See Writs.)
Outlaws, 384-5.
Pandulf (Papal legate), 26, 45.
Papal Bull, annulling Charter, 45-6.
Papal legates. (See Gualo, Nicholas,
Pandulf.)
Parks, distinguished from forests,
422.
Parliament.
its development, 162.
functions of, 162, 253-4.
composition of, 162.
Parliamentary taxation. (See Tax-
ation.)
Parliamentary representation. (See
Representation.)
Peasantry. (See Villeins.)
Peers, amercement by, 130, 295-8,
as a separate class, 186 n.
Peers, judgment of, 375, 377-9, 386-
95.
confused with jury trial, 134 n.
Peter des Roches (Bishop of Win-
chester), 30, 41, 45, 48. 255,
389, 431-
Peter of Wakefield, 26.
Peter's pence, 17.
Petition of Barons (1258), 198 n.,
214, 283 n., 414, 435.
Petty Assizes. (See Assize.)
Petty Jury. (See Jury.)
Petty Serjeanty. (See Serjeanty.)
Philip (King of France), 26, 27, 30,
46.
Pipe Rolls, of 1 1 30, 9.
revived by Henry II., 11.
Pleas. (See Common pleas.)
Pleas of the Crown, 88-92, 266-7,
304-21. -
in Scots law, 306.
Pleas of the forest, 313-4.
Popes. (See Honorius and Inno-
cent.) •
Presentment (darrein), 193, 269,
276-7, 233-
nature of, explained, 276.
Primer seisin, 63-5.
Procedure (legal), 81-6.
Henry II. 's reforms, 86-92.
Ptoof (in medieval law), 84-6. (See
also Trial.)
Protests by the prelates in John's
favour, 43, 476.
text of, 496-7.
Provisions of Oxford (1258), 160,
433-
Purprestures, 425.
Purveyance, 329-32, 334-6.
Quarantine, widow's, 218,
Queen's gold, 198.
Recognitio (principle of), 28, 86,
135-8.
Regarders, 417-22.
Relief (feudal incident of), 59-61,
203-5, 211-2.
not due from bishop-elect, 97-8,
202-3.
not due after wardship, 203.
definition of sums due, 196-
203.
Representation, parliamentary, 248-
55-
Responsibility. (See Ministerial.)
Richard I., 51, 66, 73, 197, 243, 460.
lessons of his reign, 20-1.
Richard II., 407.
Roches (Peter des). (See Peter.)
Roger, bishop of Salisbury. 9.
Rolls. (See Pipe Rolls.)
Royal justice. (See Courts and
Jurisdiction.)
Royal writs. (See Writs.)
Runnymede, 36-8, 44, 52, 479.
Saladin tithe, 20.
Salisbury, Roger, bishop of. (See
Roger.)
Schedule of grievances (27 April,
1215), 33, 37, 38, 173.
Scotland, relations to England,
459-63.
kings of. (See Alexander, David,
and William.)
Princesses Margaret and Isabel,
459.
Scutage, 69-76, 231-4.
grand sergeanties not liable for,
55-6.
in charter of 1215, 231-4, 248.
in charter of 1217, 148-9.
by sub-tenants, 256.
Scutum. (See Knight's fee.)
Seal (king's great), 143-4. 153-
Seisin. (See Primer Seisin and
Disseisin.)
Serjeanty (grand), 55-7, 64 n,, 199,
261.
Serjeanty (petty), 56-7, 64 n., 199,
261, 367-8.
Service. (See under Feudal service,
Foreign service, and Knight's
service.)
INDEX
529
Sheriff, 31 1-3, 431.
history of the office of, 31 1-3.
chief magistrate of county, 311.
his powers, 16, 304-10.
curtailed by the charter, 304, 317,
334.
his " tourn," 320-1.
his tyrannies, 15-16, 311, 320, 334.
his qualifications, 431-3.
his gradual decline, 16, 312-3.
forbidden to hold pleas of the
Crown, 304.
in America, 313.
in Scotland, 313.
accountable to Exchequer, 9.
under William I., 15.
under Henry II., 15-16.
Inquest of, 16.
Shire. (See County.)
Socage, 54-5, 57, 66, 199, 261, 367-9.
Statutes. (See separate Index of
Statutes.)
Stephen (King).
supplants Matilda, 10, loi.
his title, 102, 188.
relations with Church, 18, 102.
(See also Charters of.)
Stephen Langton.
nominated by Innocent, 23.
received by John, 27.
his alliance with the barons, 28.
exhibits charter of Henry I., 28,
32.
consolidates the opposition, 51.
mediates between King and
barons, 29, 32, 33, 36, 38, 45.
at Runnymede, 36, 38.
trusted by both sides, 38-9, 43, 51,
462 n.
influence on Great Charter, 39,
191, 212.
Tower of London placed in his
custody, 43.
suspended from office, 45.
goes to Rome (1215), 45.
asks Henr^' III. to confirm the
Charters, 153.
cares for interests of English
Church, 191, 212.
Stubbs (Bishop William).
on Henry I.'s compromise with
Anselm, 18.
on the coalition of John's oppon-
ents, 44.
on John's submission to Innocent,
51-
on scutage, 69-71.
on charter of Henry I., 97, 464.
on " national '* character of Char-
ter, 113, ii7n., 125.
on the barons' unselfishness,
117 n., 464.
on villeins under the Charter,
119 n.
on reissue of 12 16, 140 n., 143 n.
on treaty of Lambeth, 145.
on witnesses to the Charter, 189.
on maritagium, 217 n.
on estovers, 219.
on constitution of London, 241.
on the Common Council, 252, 254.
on c. 24, 310.
on c. 45, 431.
on meaning of " lex," 370 n.
on judgment of peers, 390 n.
on meaning of " warren," 423.
on the twenty-five executors, 476.
Succession, intestate, 141-2, 326-9.
testate, 321-6.
Suit and service, 67-9.
Suit (or forewitnesses), 84, 369-75.
Swanimotes, 421.
Synod of Whitby, 6.
Tallage, 235-9.
Taxation.
common Council's rights over,
231-40.
parliamentary, 238-9.
Tenserie, definition of, 40 n.
Tenure by barony, 54.
Tenure by castle- ward, 57 n.
Tenure in chivalry. (See Knight's
service.)
Tenures, 53-9. (See also Burg-
age, Fee-farm, Frank almoin.
Knight's service, Serjeanty,
Socage, Villeinage, also under
Freehold.)
Tourn. (See Sheriff's tourn.)
Treason, 60, 337-9.
Trial (in medieval procedure), 83-6,
136, 359, 369.
Trial by combat, 89, 136-7, 272,
360-7, 451-3-
Trial by jury. (See Jury.)
" Unknown charter of Liberties "
(so-called), referred to, 29.
theories as to its nature and
origin, 171 -5.
text of, 485-6.
Usury, 223-5. (See a^lso Jews.)
Verderers, 314, 418.
Vesci, Richard de, 25, 26, 33, 384,
443, 469.
Vicecomes. (See Sheriff.)
Villeinage, 54, 290, 386.
530
INDEX
Villeins, 289-92.
not freemen, 11 8-9.
methods of " wasting," 208 n.
amercement, of, 289-92.
attitude of charter to, 119, 335,
386, 386 n.
Wales, 25, 456-9.
Wapentake, 78, 135, 317.
Warden, 417. (See also Constable
and Guardian.)
Wardship, 61-2, 63.
tenures subject to, 55, 57, 367-8.
conditions of, 204.
attitude of charter to, 205-12,
367-9, 450.
over vacant sees, 97-8, 21 1-2.
over abbeys, 433-5.
prerogative wardship, 368-9,
450-1.
Warrens, meaning of, 210 n., 423.
distinguished from forest, 423.
charter on, 438-440.
Waste, of ward's lands, 207-9.
of felon's lands, 336-9.
of forests, 426-7.
Wathng Street, 6.
Waynage, 291-2.
Weirs, 343-6.
Welshmen. (See Wales.)
Wergeld, 285-6.
Whitby, Synod of, 6.
215-21, 230-1, 325.
13-14.
Widows' rights,
W^ilham I., 6.
outline of reign, 6-9.
his policy of balancing, 7-8.
relations to feudalism, 7.
to Domesday Book, 10.
local government under,
15-
relations with Rome, 17.
his innovations, 86.
title to English Crown, 7, 95-6.
relations with Witenagemot, 7-8,
14.
William II., relations with the
Church, 17, 203.
features of reign, 8-9.
William the Lion, King of Scots,
460-3.
Wills, 321-6.
Witenagemot, 14.
attitude of William I. to, 7-8.
powers of, 188-9.
Wites, 285-6.
Witnesses, 84.
Writ de odio et atia, 89,
. 361-5-
Writ ne exeat regno, 409-10.
Writ of habeas corpus, 362-3.
Writ of entry, 354-5.
Writ of right 347-9, 353-4-
Writ praecipe, 346-55.
Writs of November 12 13, 29-30.
136,
GLASGOW : PRINTED AT THE UNIVERSITY PRESS BV ROBERT MACLEHOSE AND CO. LTD.
PRESS NOTICES OF FIRST EDITION
MAGNA CARTA : A Commentary on the Great Charter
of King John, with an Historical Introduction. By William Sharp
McKechnie, M.A., LL.B., D.Phil., Lecturer on Constitutional
Law and History in the University of Glasgow.
" A book remarkable alike for its solid learning, its fertility in suggestion,
and its characteristic note of moderation and sweet reasonableness. . . .
No one will close the book without feeling that his knowledge of Magna
Carta has been greatly amplified and enriched. . . . One of the most
important contributions to medieval history in recent years." — The Times.
" Of the great merit and value of the present essay there can, we think,
be little doubt. The historical and juridical methods adopted by the
author are sound and intelhgent, and naturally in striking contrast with
those of earlier commentators. His style is attractive and sufficiently
concise, whilst his equipment for such a formidable task is very consider-
able. As the result of these elaborate studies, we have an entirely new
view of the Great Charter in its legal and constitutional aspects and political
environment. ' ' — A thenaeum.
" Mr. McKechnie's work is of such great value that no historian can
afford to neglect its clear and reasoned statements on all the questions
which centre round the scene at Runnymede, and on the economic facts
which led up to, and away from, that momentous conference." — Spectator.
" Bears with it the collective wisdom of all the commentaries, ancient
and modern, dispassionately scrutinised and sifted with close and varied
learning on medieval institutions. ... It is a great theme, and to say
that it is at last adequately treated, from the constitutional side, is well
merited praise." — Juridical Review.
" Dr. McKechnie has placed students of constitutional law and history
under deep obligations. . . . There can be no doubt that his work comes
to fill a long-felt want in legal and historical Uterature."— Om^/ooA (New
York).
" Few more readable books — except to those who can read nothing
but fiction — have ever been written." — Liberty Review.
" Students of the historical development of the law of the British Con-
stitution owe a welcome to the appearance of this learned treatise, a work
which brings the interpretation of a prime document of political liberty
abreast of recent advances in knowledge which, in legal history no less
than in other departments of inquiry, have brought in new views of old
facts. ' ' — Scotsman.
" Dr. McKechnie has given very satisfactory explanations of most of
the knotty points of the Charter, and we may specially commend the
sanity of his exposition of the constitutional clauses. Altogether his
book is a useful and solid help in the study of the numerous difficulties
involved in the text of the Great Charter." — Manchester Guardian.
PRESS NOTICES OF FIRST EDITION
•' One of the most important contributions to English diplomatics that
has appeared within the last five years." — Westminster Review.
" It has already won its place as the standard treatise on the Great
Charter." — Scottish Law Review. ^,
*' Dr. McKechnie has performed a real and needed service to historical
and legal scholarship. The manner in which he has performed his task
is as admirable as the matter. He writes in a clear and direct style, well
adapted to his subject. The book is printed in attractive form." — Law
Notes (New York).
" It may be said at once that our author has succeeded in giving us a
very sound and useful commentary instead of the antiquated ones of
Blackstone and of Thomson. . . . His book is sure to be used and to
last." — Professor Vinogradoff in the Law Quarterly Review.
" The labor of anyone who comments in the future on the clauses of
the Great Charter has been greatly lightened by the careful and learned
work of Dr. W. S. McKechnie." — Professor G. B. Adams of Yale, in The
Origin of the English Constitution.
" Of permanent value to students of EngUsh constitutional history." —
Annual Register.
" An admirable commentary on Magna Carta." — H. A. L. Fisher,
Fellow of New College, Oxford, in an editorial note to Maitland's Con-
stitutional History of England.
" A work on Magna Carta of unparalleled excellence in both its historical
and expository matter." — Hon. Fred. A. Baker in The Gateway (Michigan).
" We question whether another commentary will venture to attempt
to displace this one until some generations of historical students have
been at work on new material. . . . Anyone who turns to these pages
for help in particular difficulties will find enough to persuade him that
he had better read every section." — The late Miss Mary Bateson, Fellow
of Newnham College, Cambridge, in Scot. Hist. Rev. iii. p. 229.
" Mr. McKechnie may justly claim to have provided us with a most
adequate commentary on Magna Carta." — H. W. C. Davis, Fellow and
Tutor of Balliol College, Oxford, in English Historical Review, xxi. p. 150.
" Im ganzen eine Arbeit, die jeder Erforscher der mittelalterlichen
Rechts- und Verfassungsgeschichte Englands dankbar brauchen wird." —
Professor FeUx Liebermann, of Berhn, in Historische Vierteljahrschrift.
" C'est un livre qu'il faudra consulter pour I'histoire des institutions
politiques et admin istratives de I'Angleterre au xii® et au xiii® sidcle ;
il a sa place marquee a cote de Stubbs dans la biblioth^que des erudits." —
Professor Charles B6mont, of the Sorbonne, in La Revue Historique,
v,l
PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET
UNIVERSITY OF TORONTO LIBRARY
JN
U7
M2
1914
cop. 2
McKechnie, William Sharp
Magna carta