Skip to main content

Full text of "Swift The Man, His Works, And The Age Volume One Mr Swift And His Contemporaries"

See other formats


Tite JVHaxi.., his "Works., arxdL tHe; 




Mr Swift and his Contemporaries 
by Irvin Ehrenpreis 



F*rinteel in Greezt JBritctin 

will <ai>;p!&ur 4ett: t;lbiL<s rntci of 


PREFACE page ix 




3. DUBLIN 21 

4. INFANCY 27 



7. STUDIES 57 


Part II 

1. TEMPLE 91 

2. MOOR PARK 102 

4. EARLY POEMS (ll) 132 


6. KILROOT 150 


Part III 

1. A TALE OF A TUB (l) 185 

2. A TALE OF A TUB (ll) 204 

3. A TALE OF A TUB (ill) 2l6 




















Compared with the ideal of a definitive biography, the pre- 
sent volume has more faults than its author has blushes ; 
compared with available biographies, it aims at a new 
standard of thoroughness and accuracy. The two remaining 
volumes are already complete in a preliminary draft, and will 
appear in reasonably quick succession to the first. 

Biographers find Swift's character so fascinating that often 
they treat him in comparative isolation, telling the single story of 
his inner development and employing other people, as well as 
public events, only as these bear unavoidably upon the man or his 
works. Presented with so stark an image, the reader must come 
with unusual resources if he hopes to judge both the degree to 
which Swift was representative of his generation and the degree 
to which he was either independent or eccentric. Since I think 
this judgment is important, I have drawn many parallels be- 
tween Swift and his contemporaries. I have tried, by revealing 
unexpected connections and relationships, to suggest the narrow, 
close-knit nature of the social fabric to which he belonged. I have 
further tried to indicate how far intellectual traditions and pub- 
lic events could, as it were, endow Swift with principles which 
might seem arbitrary to us. 

I have been less concerned to add than to eliminate fables ; and 
those readers who look for my views on a long train of legendary 
Swiftiana will search in vain. Here, neither Swift nor Stella is 
made a bastard ; Swift does not say, c My uncle gave me the educa- 
tion of a dog' ; Dryden does not say, 'Cousin Swift, you will never 
be a poet* ; and Temple does not seat Swift and Stella at the ser- 
vants' table. But I have looked minutely into Swift's intentions 
and principles. Since his early works contain bold expressions 
of his ideals and intricate examples of his satirical methods, 
I have given them a detailed examination. Because A Tale of a 
Tub, his hardest and most brilliant work, has been misunder- 



stood by many critics, I have gone over it with unusual care. 

For early encouragement, steady guidance, and innumerable 
kindnesses, I am indebted to my teachers, George Sherburn, 
Louis Landa, and Sir Harold Williams. 

For generous support of the work, I am indebted to Indiana 
University, the U.S. Educational ('Fulbright') Commission, the 
John Simon Guggenheim Memorial Foundation, the American 
Council of Learned Societies, and the American Philosophical 

For their incredible patience and hospitality, I am indebted to 
Dr William O'Sullivan, Dr Richard Hayes, Mr L. W. Hanson, 
and the other librarians of those sanctuaries where my work was 
mainly done: Indiana University, the British Museum, Trinity 
College (Dublin), the National Library of Ireland, and condi- 
tion to which all others naturally aspire the Bodleian Library, 

For generously permitting me to use manuscript materials, I 
thank the Board of Trinity College, Dublin, the Trustees of the 
British Museum, the officers of the Leicester City Museum, Bod- 
ley's Librarian, and Mr James Osborn. I thank the Syndics of 
the Cambridge University Press for allowing me to quote from 
R. L. Colic, Light and Enlightenment > and A. R. Hall, Ballistics in 
the Seventeenth Century >; the Delegates of the Clarendon Press for 
L. A. Landa, Swift and the Church of Ireland] and the officers of the 
University Press, Dublin, for C. E. Maxwell, A History of Trinity 
College^ Dublin, 

For many kinds of assistance over many years of work, 1 am 
indebted to Mr Giles Barber, Professor Frederick L. Itaaty, Mi- 
James T. Boulton, Mr John Russell Brown, Mr G. A, Chinnery, 
Professor James L. Clifford, Professor Rosalie L, Colic, Professor 
Ronald S, Crane, Professor Philip B. Daghlian, Professor Her- 
bert Davis, Professor Oliver W* Ferguson, Mr Alastair I). S. 
Fowler, the Rev, J* G. Frostick, Professor Rudolf B- Gottfried, 
Professor Donald J* Gray, Professor John C. Hodges, Professor A. 
Rupert Hall, Professor Colin J. Home, Mr Emrys Jones, Pro* 
fessor Alexander C* Judson, Mr Hugh F. Kearney, Professor 



George P. Mayhew, the late Professor William Thomas Morgan, 
Major P. D. Mundy, Mr James M. Osborn, Professor Gordon N. 
Ray, Professor Robert W. Rogers, Lord Rothschild, Professor 
Edward L. Ruhe, Miss Nicolete Shawyer, Mr John Gerald 
Simms, Professor Charles H. Taylor, Jr, the late Mr W. H. Welp- 
ly, Miss Kathleen Williams, the Rev. R. G. Williams, Mr David 
Woolley, Mr Jonathan Wordsworth. And finally I must acknow- 
ledge an extraordinary obligation to Mr M. R. Ridley, whose 
sagacity has saved the reader from being exposed to some scores 
of pedantries and who would have liked to preserve him from 

I. E. 



Chapter One 

For a man who claimed that his family were c of all mortals 
what I despise and hate 5 , 1 Swift surely had much to do with 
them. His early years were sheltered by an uncle's hospi- 
tality; the last years were eased by a cqusin's devotion. He paid a 
lifetime allowance to a needy sister; he- supported and regularly 
visited a widowed and distant mother. In his era of intimacy with 
peers and statesmen, he saw and gave help to humble relations. 
When, ageing and ill, he lived withdrawn from the world, he lent 
a fortune to a young cousin; another cousin's husband was for 
a while his curate. Probably, Swift's ironical, ostentatious con- 
tempt for c what the world calls natural affection' 2 betokens an 
instinct grown too powerful for him to handle directly. 

Contrary both to received opinion and to the hints dropped by 
Swift himself, his relatives influenced in fundamental ways his 
literary ambitions, his political sympathies, and his religious con- 
victions. He was born into a family allied with two of the great 
names of seventeenth-century literature, Dryden and Davenant: 
his cousinship with Dryden he repeatedly mentioned; Dave- 
nant's grandson played the part of a brother to Swift during his 
childhood and youth. The high church, anti-Whig political 
alignment of Swift as an adult follows the course of his father's 
generation and of the father's father. In his vocation of priest and 
dean Swift, with belligerent persistence, supported policies which 
not only tie him to one grandfather but oppose him to the other; 
for though both were parsons, one had been persecuted by Puri- 
tans and the other by Laudians. 
But Swift has not only misled us as to the effect of his forebears 

* Ball vi. 113. * Ibid., p. 126. 
B [3] 


upon him; he has even misled us as to plain facts of his ancestry; 
and several early biographers have deepened the darkness by as- 
sertions and conjectures which can now be dismissed. Swift's ex- 
act relationship to Dryden, the identity of his mother's father, the 
large number of clergymen in his background, the early division, 
in associations, between his sister and himself these are some of 
the points established by a survey of his descent. We shall dis- 
cover how, in a literal sense, 'natural* it was for Swift to grow up 
into a high church Anglican priest, bitterly opposed to noncon- 
formity, separated from his sister, and preoccupied with literary 

One ancestor he venerated above all, his parson grandfather 
Thomas Swift. Writing toward the end of his lifc% when he was 
almost seventy-two, Swift still flamed at the idea of that royalist 
vicar's sufferings from the Puritans 'persecuted and plundered 
two and fifty times by the barbarity of Cromwell's hellish crew 1 . 1 
Yet Swift knew little about his own extraction, and that little was 
often wrong. In a fragment of autobiography, he traced his pa- 
ternal line to a Yorkshire family which had really a negligible tie 
or none with the Reverend Thomas Swift (1595-1658), vicar of 
Goodrich and rector of Bridstow, Herefordshire. 

His own maternal grandparents Swift skips over in this ac- 
count, blandly tying his mother, Abigail Erick, to *the most an- 
tient family of the Bricks, who derive their lineage from Krick the 
Forester'- Genealogists believe it more likely that she belonged to 
a modest branch of the Krick or Herrick family of Leicestershire, 
'very private gentlemen* In her son's odd phrase- ; her father was 
no doubt the Reverend James Ericke (B.A., Cambridge, 1624), 
vicar of Thornton, Leicestershire, from 1627 to 1634^ (Although 
she is supposed to have been related to Dorothy Osbornc, the 
connection has not been traced*) 4 

Swift owed his Christian name, in the last instance, to neither 
the Bricks nor the Swifts, but to the family of his father's mother. 

* Ball vi 127, * Autob.t f, 6 V . * Sec Appendix B. 

* When Sir William Temple died, Swift'* mother was one of tho*e who received 
an allowance for mourning. 



From these more remote ancestors were descended cousins with 
whom he was to spend much time when he lived in England ; and 
from the same ancestors, as Swift readily pointed out, John Dry- 
den was also derived. Both writers Dry den in the male line. 
Swift in the female are traced to a Northamptonshire gentle- 
man, John Dryden of Canons Ashby ; for one of his sons was Dry- 
den's grandfather, and another, Nicholas, was Swift's great 
grandfather. This second cousinship once removed is what 
Swift termed a 'near relation 3 . 1 

The name 'Jonathan* appears only after Nicholas Dryden's 
marriage; and with it we meet other names belonging to the 
generation of Swift's father. Nicholas Dryden married a Mary 
Emyley, both whose grandfather, Thomas Godwin, and uncle, 
Francis Godwin, were bishops. Nicholas and Mary Dryden 
called their eldest son Jonathan and their eldest daughter Eli- 
zabeth. A son who died in infancy was called Godwin, after 
(one assumes) either or both of his episcopal forebears. 

Here begins the tale which Swift knew: for it was Jonathan 
Dryden's sister Elizabeth who married the Reverend Thomas 
Swift of Goodrich; and their fifth son, Swift's father, was named 
Jonathan, probably after his Dryden uncle. The eldest and by 
far the most important uncle of Swift himself was named God- 
win; another uncle, Dryden. Such names and such connections 
show that Swift's links with literature and the church go back in 
one direction as far as the female side of this favourite grand- 
father's family. 

On the other side, Thomas Swift came not of Northampton- 
shire gentry but of Kentish clerics* Both his father, William Swift 
(1566-1624), and his grandfather, Thomas Swyfte (1535-92), 
had been rectors of St Andrew's, Canterbury; and his great 
grandfather, William Swyfte, had also lived in Canterbury. 2 In 

x Ball v. x62, 452-3; P. D. Mundy, N. & Q., 4 Oct. 1924, pp. 243-4; 18 Oct. 
1924, pp. 279-80, 334; 30 Oct. 1948, pp. 470-4; J. M. Osbom,Johnryden: Some 
Biographical Facts and Problems (New York 1940), p. 237. Swift was not as has been 
said also related to Dryden through descent from Bishop Thomas Godwin : see 
Mundy, JV. fi? Q., i Sept. 1951, pp. 383-4, 

1 Mundy, JV. 6? Q. 9 i Sept. 1951, pp. 381-7. 



naming his many sons, however, Thomas Swift seems to have 
drawn more on his wife's family than his own. Godwin, the eldest 
son (1628-95), must go back ultimately to one of the bishops 
probably Francis (1562-1633, Bishop of Hereford), who had col- 
lated Thomas Swift (his cousin) to the Goodrich living and, being 
alive, might still do some duty. 1 Dry den, the second son (born 
1629), obviously perpetuates Elizabeth Swift's maiden surname. 
Thomas, the third (born 1633), continued one Swift tradition; 
and William, the fourth (1637-^. 1705), continual another. 
Jonathan, the fifth (baptized 24 May 1640,, died March or April 
1667)5 has already been linked to Mrs Swift's brother. 

Swift's fragment of autobiography reflects more concern with 
the parson Thomas than the parson's children ; for Uncle God- 
win is dismissed in five sentences, Thomas in three, and Drydcn, 
William, and Adam,- all together, in one. Of the three last, Swift 
remarks that 'none of them left male issue*, but that Jonathan, 
'besides a daughter left one son*. 3 This daughters name may yet 
again have significance: it was Jane, and it must belong to Abi- 
gail Swift's family; wo know that Mrs Swift had a niece, Jane, 
daughter of a brother, the Reverend Thomas Errick, whose wife 
was named Jane as well. On his father's side, Swift had many 
female cousins, and they had many female offspring ; but none of 
them seem, like his sister, to be called Jane. By every token, Swift 
counted himself as belonging to his paternal grandfather's fa- 
mily, not to his mother's. By sex and by name, by earlier birth, 
and as we shall find by constant association, Jane would 
belong to his mother's side* 

It is not certain that Swift's mother knew much of her father, 
since he may have died when she was a child ; it is far less certain 
that her son did* Yet Swift made many visits to Leicester; so he 
must have been familiar with the city's Puritan tradition and its 
anti-royalist role in the Civil War. Thomas Errick, Swift's 
mother's brother, only died in 1681, but there were other ways as 
well to pick up hints of James Ericke's sympathies; and no 
admirer of the Laudian tradition would have welcomed these 

* Mundy, JV. & Q. t i Sept. 1951, p. 384, * See Appendix A. * Autofr,, f. 6. 



hints. For after seven years as vicar of Thornton, James Ericke 
had confessed to holding an unlawful conventicle in his brother- 
in-law's house. He had been brought before the Court of High 
Commission; and though that court had first 'resolued to make 
tryall of him for a tyme to see how and in what manner he did de- 
meane and carry him selfe in the execution of his ministry'/ he 
was probably, in the end, deprived of his living. 2 If Swift was at all 
aware of the Puritan strain in his mother's background, his re- 
markably intense devotion to the royalist Anglican of Goodrich 
would seem to imply a corresponding repudiation of the vicar of 
Thornton's family. 3 I am confident that he was indeed aware of 
the strain and that while Swift's lifelong polemic against non- 
conformity could not be simply due to this element in his back- 
ground, it does reflect the degree to which he considered himself 
a 'Swift' rather than an 'Erick'. 

1 P.R.O. MS. SP i6/ 261, 8 May 1634. 

2 On the Induction Mandate for his successor, John Summerneld, the parish is 
stated to be vacant 'per cessionem derelictionem sive deprivationem Jacobi 
Ericke% which suggests but does not confirm deprivation (Leicester City Archives 
MS. 1041/28/442). 

3 See Appendix C. 


Chapter Two 

The meaning of Swift's work will escape anybody who for- 
gets that his English career, long and important as it was, 
only interrupted an Irish life. Although a "residence in 
Whitehaven during infancy was to remain a cherished fact for 
Swift (and though he may perhaps have briefly visited England 
during his adolescence) , he had no real experience of Leicester, 
London, or Surrey till he left the university of his native city as a 
young adult. The traditions of his Herefordshire and Kentish an- 
cestors had to receive a strange Irish setting before he was intro- 
duced to them; and around the immediate frame of a Protestant 
English family living in Ireland extended the larger frame of the 
so-called 'English interest' there. Let us understand the evolution 
of the unstable social order to which Swift belonged with its 
splits between the native Irish and their conquerors, noncon- 
formity and the Established Church, landlord arid priest, old 
settlers and new administrators and we shall sec many of his 
distinguishing features emerge not as eccentricities but as intel- 
ligible reflections of the backgrounds of his career: for example, 
his conservative morality, or his uneasiness about property and 
wealth, or his unwillingness to call himself a Tory; his attacks 
upon Presbyterianism, his sympathy with the sufferings of the 
native Irish, coupled with his contempt for Roman Catholicism; 
his aggressive identification of himself with England, matched 
by his violent criticism of English policies in Ireland; his love 
of the church and his loathing of bishops, 

In the development of that uneasy social order, however, cer- 
tain forces did touch the dependent and half-orphaned boy more 
than most of his contemporaries : one was the shaky condition 



both of property titles and of money; another, the mutual dis- 
trust of northern, Scottish Presbyterians and the Anglican popu- 
lation of Ireland. Under such influences, Swift struggled to ac- 
cumulate an estate for himself; he battled all his life for the 
strengthening of the Church of Ireland ; and in his greatest politi- 
cal essays. The Drapier's Letters, he evoked the principles of human 
freedom out of a controversy over the coinage of money. Yet it is 
in the more general pattern of Irish history that we can observe 
not only what brought the Swifts to Dublin originally but also 
how they became involved with two great families, the Temples 
and the Ormondes; for the Temples directly and the Ormondes 
by way of the institutions they controlled were to guide young 
Jonathan Swift through the first stages of his career. At the same 
time, moreover, these families embodied the polar traditions of 
the English in Ireland : the Ormondes or Butlers, anciently es- 
tablished there, accepting responsibility for the whole popula- 
tion's welfare ; the Temples, come over as administrators, regard- 
ing the kingdom only as a province that should be of some use to 
the rulers and their dependents, with no care for the condition of 
the vast, stubborn majority. We shall see that Swift, assigned by 
birth to the social philosophy of the Temples, educated himself 
to transcend it and to support that of the Ormondes. In remark- 
ing this movement, however, as in studying any aspect of Res- 
toration Ireland, we shall acknowledge that it derives in turn 
from the ultimate principle of the history of the kingdom, the 
difference in number and religion between the rulers and the 

Swift, then, grew up in a middle-class, Anglican community 
within a much greater population of rural Irish Roman Catho- 
i lics^The Ireland he knew was the result of several violent but in- 
^conclusive military and colonizing projects of the English! The 
.earliest effort, accompanying the invasion of Strongbow and the 
Normans in the twelfth century, planted among the Irish a num- 
ber of great ruling families whose interests merged only very 
gradually with those of the natives. Into the seventeenth century 
the descendants of these invaders continued to wield something 



of their traditional power. From one of them, the Butlers, came 
the first Duke of Ormonde, who dominated the institutions which 
shaped Swift's career; and the second Duke, his grandson, whom 
Swift well knew and too much admired. By the end of the thir- 
teenth century such families possessed (at least, nominally) most 
of arable Ireland. 

During the next three hundred years the English settler did 
not exterminate the Irish ;Qie did not admit the Irish to civil or 
political rights} and he did not multiply so as to outnumber the 
Irish(lnstead, he continued to live in the conquered land as over- 
lord, holding high offices under the crown, and despising the 
natives until their resentment boiled over into armed rebellion^ 
A ferocious suppression would regularly end each uprising, with 
confiscation of the land of the leaders. Undisputed English su- 
premacy, however, hovered around the dimensions of the Pale. 
This enclave, where English law was regularly enforced, had 
been established by Henry II; it centred on Dublin, but varied 
in extent, generally including most of what is now the county 
of Dublin and much of Louth, Meath, and Kildaro, Sir John 
Temple, writing about 1646, described it as *a large circuit of 
land possessed at the time of the first conquest of Ireland by the 
English, and ever since inhabited by them; it contains several 
counties, viz., the counties of Dxiblin, Meath, Lowth, Kildarc, 

&C.' 1 

The feckless but mercenary administration, by the king's 
officers, of the English colony itself, fed a spirit of separatism. 
During the fifteenth century this grew into a resentful hostility 
toward England and English interests ; what mainly irritated the 
settlers was the steady neglect of the authorities to assist in their 
defence. "By Elizabeth's time three of the most pervasive elements 
in Irish history were established : the political alienation of the 
natives, who remained by far in the majority; the distrust of the 
so-called Anglo-Irish and the natives for each other a hatred 
nourished by the long history of mutual terrorism; and the clash 
of interests between the descendants of the English born in Ire- 

1 The Irish Rebellion, 1679, p. 63. 



land, and those Englishmen either directing Irish affairs from 
England or recently arrived in Ireland on administrative or 
plundering missions. 

It is always to the interest of a mother country to prevent in- 
dividual colonists from winning entrenched power for their fami- 
lies through the cumulative effect of intermarriage, legacies, 
hereditary perquisites, and traditional apportionments of privi- 
lege. At least, it is to the interest of ambitious statesmen, who 
must control every source of preferment. For this end, and to keep 
the colonists' regime generally subservient, it is common for the 
home government to bring in, as administrators, new men who 
will not direct their superiors but obey them. Thus the Domviles, 
Boltons, Percivals, Doppings, and Temples, whom Swift was to 
consider proper (if imperfect) leaders of Irish society, would only 
have risen to high office in the latter half of the sixteenth cen- 
tury. 1 In Ireland, moreover, just as c old 3 English Catholic land- 
owners saw themselves ousted, under Elizabeth and James, by 
new Anglican families associated with the army or the planta- 
tions, so the heirs of those recently established families were to 
find themselves displaced, under the later Stuarts and George I, 
by fresh carpet-baggers sent over from England. 

A further complication was introduced with the plantation of 
Ulster, early in the seventeenth century. In that northern pro- 
vince a tremendous conspiracy had been smashed, and the usual 
confiscations had taken place. These in due course opened the 
way to a new settlement of loyal Britishers; and because careful 
limits were put on the type of participation allowed, the project 
did achieve permanent success. Of all the undertakers, the Scots 
were the most energetic. Since, in addition, the counties of An- 
trim and Down, bordering on the escheated lands, were already 
well colonized by Scots, the enduring character of the new settle- 
ment was Scottish, When the Presbyterian ministers, persecuted 
by James's episcopacy, came after their people, they established 
a religious bent which has remained a distinguishing mark of 
northern Ireland* 

* H. F. Kearney, Strqffbrdin Ireland (Manchester 1959), p. 18. 


The strongest block to Anglo-Irish amity may also be blamed 
on the Tudors. This was the failure of the Reformation in Ire- 
land. Although a parliament sat in Dublin 1536-7 for the pur- 
pose of establishing the new church, such legislation had negli- 
gible effect outside the Pale. The Roman Catholic Church had 
always commiserated the sufferings of the Irish, and there was in 
the country no sentiment favouring a break. The suppression of 
the regular clergy and the reversion of ecclesiastical property to 
the crown gave the whole movement a venal, lawless stamp 
which was attended by neither the piety nor oven the proselytiz- 
ing which might have relieved it. Recognizing their advantage, 
the Roman Catholic clergy laid down a policy of complete 
loyalty to Ireland and relentless opposition to English Protestant 
ascendancy. This strategy was so successful that, by Swift's time, 
to be Irish was to be Roman Catholic, although to be English 
was not always to be Anglican. 

The forces behind all these developments came into play dur- 
ing the vast rebellion which started in 1 641 in Ulster, spread over 
the whole of Ireland, and was not finally suppressed until 1652. 
The results of it gave Irish history its direction at least up to 
Swift's death. At the centre of the conspiracy was a group of 
prominent Irish Catholic clan leaders and landowners. Worried 
by the progress of the Puritans but encouraged by the vacillations 
of Charles I, they acted as though they thought the last oppor- 
tunity had come for them at one stroke to preserve their church 
from extinction, to frustrate a final seizure of their estates, and to 
rectify the abuses of centuries. As the Civil War went on in 
England, a split developed within both this group and the British 
power which opposed them, so that the Irish divided between 
those who remained loyal to the king himself and those who de- 
manded complete independence, while the British in Ireland 
divided between Parliamentarians (including, naturally, the 
Scottish Presbyterians) and loyalists (including the Anglicans). 
In August 1649 s** an< i a half months after the execution of the 
king the Irish Rebellion reached the point where Cromwell 
himself marched in* But though he spread a ferocious, if uniform, 



terror for nine months, active resistance endured into the spring 
of 1652. Then at last, under the Articles of Kilkenny, the utter 
conquest of a nation was done. 

By this date, unfortunately, the cost to Parliament of putting 
down the Rebellion had swollen so high that out of the twelve 
million acres of arable land in Ireland, five would be wanted to 
pay the accumulated debt. In the summer of 1652, therefore, an 
Act of Settlement was passed as arrogant and arbitrary as it 
was inevitable by which the whole territory of Ireland was 
treated as confiscated property. The basis of this expropriation 
(and later ones) was no longer racial but religious. Every Irish- 
man, whether English, Scots, or Gaelic, who could not demon- 
strate his innocence in the Rebellion and his constant good 
affection to the Commonwealth of England, was to suffer punish- 
ment by loss either of life or of property or of both, "wholly or 
partial according to the degree of their guilt'. 1 Such was the so- 
called 'Cromwellian Settlement'. But if the Rebellion was thus 
extinguished, its primitive causes remained untouched and con- 
tinued to flourish : land titles were still not secure; the English re- 
newed their mistreatment of the Irish and their neglect of 'Anglo- 
Irish' families ; and the Roman Catholic Church still underwent 
a harsh, though unsystematic, suppression. 

While the main victim of Cromwell's army was the native Irish 
Catholic population, the Parliamentary troops were not, of 
course, careful of Anglican property. When Swift went to school 
at Kilkenny, he was to see a monument to their energy in the ca- 
thedral of St Canice: 'They left it roofless,' writes the first Res- 
toration, bishop, 'took away five great and goodly bells, broke 
down all the windows and carried away the glass, also broke 
down the doors, the font, and many goodly marble monuments.' 2 
Cromwell's administration of Ireland, intended, in Macaulay's 
words, c to make Ireland thoroughly English, to make it another 
Yorkshire or Norfolk', went far toward its goal. There was a c con- 

x Dunlop, p. 1 1 6, 

* George Seaver, Tke Cathedral Church qfStCaniee, Kilkenny (Kilkenny 1953), p. 19, 
quoting Griffith Williams. 



stant and large emigration from England to Ireland', and c the 
native race was driven back before the advancing van of the 
Anglo-Saxon population.' 1 In general, hideous agonies were in- 
flicted upon the Irish Catholics ; for although some managed to 
find cither farms they could rent or else other employment, many 
were transported to the West Indies and many became simply 

At the Restoration, the trend naturally reversed itself, in a 
movement which did not cease until 1 69 1 . The Act of Settlement 
(1662) and the Bill for the Explanation of the Act of Settlement 
(1665) re-established many of the old proprietors. Numbers of 
them had indeed returned to their homes when Charles II be- 
came king, since they had no reason to think he would deal kindly 
with the Cromwellians. Very roughly, the outcome was that a 
third of the pasture and plough land went to native Roman 
Catholic landlords, a third to the older Protestant colonial fami- 
lies, and a third to the more recent 'adventurers' and soldiers. All 
serious attempts to modify this Act failed until the death of 

With the accession of James II, the native Irish, led by Richard 
Talbot, now Earl of Tyrconnel, found their opportunity. A 
younger brother of the Roman Catholic Archbishop of Dublin, 
and a crony of James II when Duke of York, Tyrconnel had long 
been the most influential spokesman for the Irish Catholics at 
the court of Charles II. Not until the Revolution, however, did 
he have his way. Then, during the War of Williamltcs and 
Jacobites in Ireland, an irregular Irish Parliament packed 
with Catholics repealed the 1662 Act of Settlement, Yet even 
this change had no real effect, since in October 1691 the war 

On this occasion the famous Treaty of Limerick was signed, 
among the civil provisions of which appeared the last grand at- 
tempt to regulate the tenure of land in Swift's lifetime. This was 
the agreement that all submissive Roman Catholics should *be 
secured in the free and undisputed possession of their estates as 

1 Macaulay, Constitutional Essays, World's Classics cd., p. 332. 


they possessed them under the Act of Settlement'. 1 However, the 
civil Treaty (unlike the military) was never ratified by the Irish 
i.e., 'Anglo-Irish' Parliament, which on the contrary rejected 
it at last in 1697 5 an< 3- so ft remained invalid. 

Under William and Mary, the confiscations, grants, and re- 
sumptions made by crown and parliament had effects too elabor- 
ate to be detailed ; but they can be roughly summarized. In 1 688, 
between a quarter and a fifth of the profitable land of Ireland had 
belonged to Roman Catholics, whether of Gaelic Irish or of e old 
English 9 extraction. By 1703, this fraction had declined to some- 
thing like a seventh. Yet the area forfeited over these fifteen years 
was much less than what such proprietors had lost through the 
combined effects of the Cromwellian and the Restoration settle- 
ments. 'After 1703 there were no more confiscations on the 
wholesale scale of the sixteenth and seventeenth centuries. Sta- 
bility had at last been reached/ 2 

Through the evolution of this land problem, the fundamental 
instability of Irish society during Swift's early years becomes un- 
pleasantly clear. The basic form of wealth was real estate. From 
the outbreak of the Rebellion, however, the titles to thousands of 
acres changed hands with crumbling rapidity. Not regard for 
justice, but political and ecclesiastical expediency, guided these 
shifts. And if even property in land seemed shaky, money was 
more so. From the year of the Restoration until well into the 
reign of William and Mary, Irish coinage passed through al- 
chemical transformations. Under Charles II a series of private 
persons obtained licences to supply the kingdom with copper and 
brass small change; but as each licencee failed to honour his 
pledge to redeem these pieces with gold and silver, the public 
always suffered from the consequent depreciation- James II, 
during his Irish campaigns, made a chaos of the coinage. He 
raised the price of gold and silver, struck brass sixpences, issued 
coins from two mints as fast as materials could be collected: 
church bells, cooking utensils, old cannon; he even recalled his 

1 Dunlop, p. taS. 

2 J G, Siimns, The Wittiamitc Confiscation, 1956, pp. 17* 160-3. 



own large half-crowns and restamped them as crowns. Then in 
1691 all these coins ceased to be current. But still under William 
III one patentee flooded the kingdom with halfpence until it be- 
came common a historian writes for creditors to compound 
for 'one fourth copper'. 1 And all the time, of course, by inexor- 
able, universal process, the value of gold and silver fell gradually 
in both England and Ireland; for Swift well knew that it took 
thirty pounds under Queen Anne to buy the equivalent of five 
pounds under Henry VI. 2 

To Swift's private reasons for worrying about his material for- 
tunes, such a history would have given a special sharpness. The 
English for centuries have found prestige in the ownership of 
land, and put a price upon estates beyond the economic value. In 
Swift's day it was a truism that 'power follows property* (i.e., 
ownership of land) ; and although the maxim was often realized 
in reverse through the alleged effect's giving rise to the sup- 
posed cause (power likes to be respectable) Swift's obsession 
with 'real' property, as superior both morally and substantially to 
moneyed wealth, would have had all the weight of tradition be- 
hind it. If to this common tradition and to Swift's early poverty 
and dependence we join the peculiar course of Irish history, we 
shall not feel puzzled by his fear of inflationary trends ('the per- 
petual decrease of the value of gold and silver 9 ) 3 and his conse- 
quent insistence that land is the only sound bottom of a man's 
prosperity. As he struggled to build 'some little occonomy of [his] 
own*, 4 Swift was to reflect that within his memory even the most 
stable form of capital had several times been shaken. 

Over this same period, of course, the 'settlement* acts and 
penal laws (against the Papists' acquiring large estates) operat- 
ed so severely that there were few Roman Catholic freeholdings 
of any size, and those were constantly dwindling. 5 Yet for ordin- 
ary natives, in, this period, emigration was hardly a practicable 

* Davis, Drapitr, p. 233. 

* WilUam Fieetwood, Ckronicon Prtcioswn, 1 707, p, 1 67. Swift owned and appar- 
ently used a copy of this book. 

* Davis DC. 48. * Sherburn m. 96. 

* For a detailed analysis of Protestant and Catholic landowners, tee Simms. 


possibility ; they generally lived so near starvation that a failure 
of crops meant immediate famine. With conditions of existence 
that could not be much depressed by war, and with normal ways 
of improvement closed to them, the Irish were indifferent to 
prospects and obstacles alike which would have moved a happier 

To Englishmen who were seeking financial opportunities, the 
case was alluring. *In those days/ a cousinly biographer of Swift 
says, with exquisite though unintentional irony, 'Ireland was 
very moderately supplied with lawyers.' 1 Particularly at the 
Restoration, men whom the Irish must have considered carpet- 
baggers found their chance to swarm into the gaps which a suc- 
cession of upheavals had opened in officialdom and the profes- 
sions. For Swift's father's family these attractions were probably 
improved both by the confidence they felt in their record of 
loyalty, and by some direct ties with a few high personages, es- 
pecially the great Duke of Ormonde and Sir John Temple. 

At the head of the Anglican, royalist interest, almost through- 
out the Rebellion, had stood James Butler, Marquess, and in 
1 66 1 Duke, of Ormonde (whose grandson was to be one of Jona- 
than Swift's heroes). The great duke's lineage, his vast fortune, 
handsome appearance, intelligence, charm, honesty, and energy 
made him the first man of his age in Ireland. For the opening five 
years of the Rebellion he led the campaign on behalf of the king, 
first against the Irish Catholic rebels, later against both them and 
the Parliamentarians. When in 1647 he had to choose between 
the two, he handed Dublin over to the Parliamentary troops, 
aligning himself with the England of Puritan republicans rather 
than the Ireland of the Popish separatists. Afterward, *in foreign 
fields he won renown', through military and diplomatic errands 
which took him to England, France, and Germany. Upon the 
Restoration, which he helped to plan, the king heaped him with 
honours and estates ; Dryden celebrated him as Barzillai 'Large 
was his wealth, but larger was his heart.* Nevertheless, his in- 
tegrity became a fault, and the devious politics of Charles and 

1 D. Swift, p. 15. 


James finally forced him out of public affairs. Ormonde was, we 
are told, connected with the Swifts through his wife (one of her 
relatives married Swift's eldest uncle, Godwin) ; and he gave 
essential aid in establishing the family in Ireland. 

The other great support for Godwin and his brothers was al- 
legedly Sir John Temple, Master of the Rolls in Ireland, whose 
son, Sir William, was to become Swift's patron. The older Temple, 
according to Swift, was c a great friend' to the whole Swift family. 1 
Sir William Temple, describing Swift, said that 'his whole 
family 5 had been 'long known to me'. 2 Even stronger expressions 
were used by Deane Swift; for he averred that Godwin (his own 
grandfather) had been on terms of the warmest friendship with 
Sir John, that the two men passed a great deal of time together, 
and that they worked in each other's interest. This account might 
seem more convincing if it were less minute. 3 

At the outbreak of the Rebellion, Sir John Temple happened 
to be in Ireland as a member of the Privy Council. His father, Sir 
Philip Sidney's friend, had brought the Temple family to this 
country upon his own appointment as provost of the young 
Trinity College, Dublin. Sir John, born in Ireland, was educated 
at Trinity College, but eventually made his home in London, and 
visited his native land only on business. 4 His notable achieve- 
ment in the 1641 crisis was to supply Dublin Castle with food. 
Since the merchants were deeply anxious for their property, he 
not only invited some of the foremost provisioners to use the 
castle as a warehouse, but also made himself responsible for the 
safety of their stores. As they accepted the offer and moved in 
large supplies of beef, herring, and wheat, Temple (then Master 
of the Rolls) was able to victual the Castle and feed the army. 5 

In spite of his ingenuity, Sir John took the most bigoted view of 
the uprisings and freely exaggerated both the atrocities commit- 
ted by the rebels and the stubbornness of their leaders. The langu- 
age of the preface to a reprint (1679) of his book will convey his 

1 Autob., f. 8. * Ball r. 2. * D. Swift, pp. 33-4. 

4 Woodbridge, pp, 3-4. 

* Sir John Tcmp^ The Irish Rebellion . . . JV<w Reprinted, 1679, P- 3** 


attitude; here spotless virtue shrieks with that indignation which 
wolves always mouth against sheep who bite: he places entire 
blame upon the natives for 

a rebellion so execrable in itself, so odious to God and the whole 
world, as no age> no kingdom, no people can parallel the horrid 
cruelties, the abominable murders, that have been without num- 
ber, as well as without mercy committed upon the British inhabi- 
tants throughout the land, of what sex or age, of what quality or 
condition soever they were. 1 

In an age when at the most a few hundred men directed all the 
affairs of Ireland, and the same figures appeared now as lords 
justices, now as bishops, and now as college provosts, nobody in 
Sir John's station could fail to have dealings with Ormonde. Even 
outside the continual shoulder-rubbing of a small viceregal 
society. Temple as a privy councillor would regularly meet the 
lord lieutenant, argue with him^ and help draft official messages 
from the council to the king's own deputy. 2 But while Ormonde 
felt sympathetic with the people among whom his ancestors had 
lived and died, Temple had two feet in England. Carte, the bio- 
grapher of Ormonde 3 indicates that the councillors did all they 
could to aggravate the differences between the Irish and the king. 
Evidently they expected that with the end of the war great terri- 
tories in Ireland would, as they wrote to the lord lieutenant, lie 
the more open to his majesty's free disposal, and to a general 
settlement of peace and religion by introducing of English'. 3 

By their own hopes of greater gains from forfeited estates by the 
spreading of the Rebellion, they did not care to have it crushed in 
the bud. 4 

The venal calculations were not wholly belied by the event. 
Only three years after the Cromv/ellian (1652) Settlement, Gen- 

1 J. Temple, Tlie Irish Rebellion, p. 31. 

2 Among the Carte MSS. in the Bodleian Library are many messages from the 
Irish Privy Councillors, including Sir John Temple, to Ormonde; there are also 
letters from Sir William Temple, much later, to Ormonde. 

* Carte n. 140. 4 Ibid., p. 6. 

o [19] 


eral Fleetwood was writing that the country was quiet : English 
people might come over and buy land confidently, for they would 
find Ireland little different from home ; considering what the de- 
vastation had been, the 'plenty' that had sprung up for the 
'Anglo-Irish' was 'wonderful*. 1 Among those who took advan- 
tage of this situation were, of course, the Swifts. 

1 Froude I. 1 38-9. 


Chapter Three 

Amt the immediate setting, material and human, of 
Swift's childhood, we do know a little. It was either at 
the Restoration or shortly before when the royalist 
reaction was obviously setting in and Ireland was calm that 
Swift's father and uncles crossed St George's Channel. Deane 
Swift, grandson of Godwin, says they left England shortly after 
their father's death in 1 658, x but allusions to them in Ireland seem 
to commence only about 1660. Godwin, the eldest (and executor 
of his father's will), was called to the English bar July 1660, and 
to the Irish bar not until May 1 663- 2 But William, the fourth son, 
was admitted a solicitor in Dublin in November 166 1 ; and Jona- 
than had begun to work at the King's Inns, Dublin (hall of the 
lawyers' corporation of Ireland), about 1 660. 3 Adam, the young- 
est, seems to have come over last; and Dryden, the second son, 
does not seem to have reached Ireland at all. Thomas stayed in 
England and followed his father's profession. Although Jona- 
than, like Dryden, died young, the remaining three immigrants 
managed to set themselves up in prosperous comfort. 

When Godwin and his brothers arrived in Ireland, they settled 
in Dublin, the seat of government and centre of wealth. Even 
Adam, -who eventually got himself an estate in the north (where 
he became a Member of Parliament), kept a place in the capital. 4 
To an unpleasant degree, Dublin was an island of Anglican civili- 

1 D. Swift, p. 15. a Ball i. 371-2, n.; Fletcher, Gray's Inn i. 4-31. 

9 D. Swift, p. 15; Ball i. 8-9, n. 4; JV. & Q., 15 Nov. 1947, p. 497; see Appendix 

4 JV". & Q. 9 15 Aug. 1857, pp. 124-5; 10 Jul. 1858, p. 24. In his will, 27 May 
1 704, Adam Swift describes himself as of Grcencastle, county Down, but he prob- 
ably lived much of the year in Dublin (Ball i. 24, n. 2). 


zation. Yet during the decade following the Restoration it held 
upwards of thirty or forty thousand inhabitants possibly as 
many as seventy thousand and was considered one of the first 
five or six cities of western Europe. 1 In the British Isles it was 
larger than any city outside the capital supposedly more popu- 
lous even than Bristol, then the second city of England. On the 
Continent only Paris, Rome, Amsterdam, and Venice seem to 
have been larger. 

The city had long since burst through its medieval walls but 
was not yet distinguished by the great neo-classical or Palladian 
edifices which in the eighteenth century came to dominate its 
architecture. The landmarks were still the two cathedrals, the 
castle, and the single-colleged university. Nevertheless, a Paris- 
ian touring Ireland about the time of Swift's birth called Dublin 
c one of the greatest and best-populated towns in Europe'. 2 

The river Liffey crossed the city from west to east, going on to 
empty into the Bay of Dublin. South of the river lay the oldest 
quarter, including what remained of the medieval walls, with 
their seventeen towers and gate-towers generally in poor repair. 
Within these walls and nearly at the centre of the ancient town 
stood Christ Church Cathedral. South-east of it, forming a corner 
of the fortifications, Dublin Castle rose, 'with the miserable 
trickle of the Poddle river serving as a ditch round two of its 
sides'. 3 St Patrick's Cathedral (not yet steepled) was outside this 
district, about half a mile south of Christ Church; and due east of 
the castle, on rising ground near the widening Liffey and in 
Swift's youth on the border of the expanding city, lay the build- 
ings of Trinity College. 

Close to the west edge of the castle, and containing its parish 
church, was the parish of St Werburgh's, one of the oldest sec- 
tions of Dublin. Here, adjoining St Werburgh's Street, and just 
under the old south-east wall (then still standing), was the group 

1 Jouvin, p. 414; Petty n. 498, 538-9; Maxwell, p. 102, n. Petty, <?. 1686, esti- 
mated the Dublin population at 69,000 ; Conncll, p. 25 and passim, shows that 
Petty tended grossly to underestimate population figures, Edmund Lloyd, in his 
Description of Dublin, 1732, reckoned the city's population to be 150,000 (p. 4). 

8 Jouvin, p. 414. * Maurice James Craig, Dublin 1660-1860, 1952, p. 5. 



of buildings called Hoey's Court, where Swift's mother was living 
when he was born. Describing the location, Sir Henry Craik re- 
marks that it 

was then one of the best [neighbourhoods] in Dublin, and the 
houses, though conveniently close to the principal street, were ap- 
proached only by sedan chairs, and thus relieved from the noise of 
the thoroughfare. The street close by (St. Werburg Street), was the 
busy street of Dublin. 1 

During Swift's early years Hoey's Court (erected in the seven- 
teenth century) and St Werburgh's Street were where prominent 
lawyers lived. At the fashionable parish church the services were 
thronged, and there was a state seat for the lord lieutenant. In 
Fishamble Street, the northerly continuation of St Werburgh's, 
lived some of the Grattans, whose illustrious men were to be 
cherished and admired by Swift. 2 In the southerly continuation, 
Bride Street, lived William Swift, close by his brothers Godwin 
and Adam in Bull Alley (which belonged to the same street) . 
Their parish church was St Bride's, where Godwin and Adam 
were vestrymen. Richard Steele's father, an attorney, established 
his family in this parish; and Steele was baptized in St Bride's. 3 

When Abigail Erick came to Dublin, and how she met her hus- 
band, we are not sure. Since her parents apparently emigrated to 
Ireland in 1634, however, she was probably born there. 4 In her 
marriage licence she was described as *of the city of Dublin 
spinster*. 5 With a persecuted Nonconformist minister for a father, 
she must have had a sober, devout upbringing. Of her character 
and opinions when she was betrothed to Swift's father, nothing 
more definite can be said. 

But several hints are preserved concerning the affiliations of 
the Swifts as her son was to see them. Those of his uncles who 
moved to Dublin carried with them an admiration for their 
father who had just died. For this, the main reasons were his own 

1 Craik I. 1 1. 2 Gilbert i. 6, 13, 14, 30, 36, 44, 57. 

3 G, A. Aitkcn, The Life of Richard Steele 9 1889, i. 13, 14, n. i ; also C. Winton in 
3EGP., Lvm (1959), 364-9. 

* See Appendix G. 8 Ball rv. 475. 



devotion to the cause of Charles I, and the drastic punishments 
which that loyalty had brought down upon him. He was certainly 
ejected from his two livings in 1646, and imprisoned for a time in 
Raglan Castle ; and some of his belongings were certainly seized ; 
he was accused of being 'active*, e incendious*, and in arms against 
Parliament. 1 The Cromwellians are less reliably reported to have 
looted his home several times, attacked his large family, and 
ruined his property. The harassment may not have been quite so 
devilish as portrayed in the accounts of royalist sympathizers. 
That his wife and children did undergo painful hardships, how- 
ever, is clear. Rooted in the family and household where Swift 
was to grow up, there lived a deep pride in this record of truth to 
the crown. The Goodrich vicar's son, the elder Jonathan Swift, 
showed not only a respect for the legend but an appreciation of 
its political utility, in a petition for the stewardship of the King's 
Inns : * Your petitioner, his father and theire whole family have 
been, always very loyall and faithfull to his sacred majestic and his 
royall father and have beene very greate sufferers upon that ac- 
count.' 2 Swift himself, writing his fragment of an autobiography, 
went into much more detail about this ancestor and his trials 
than about anyone else except Swift. In 1726 he presented to the 
Goodrich church a chalice which his grandfather had used, with 
a Latin inscription running in part, c notus in historiis ob ea quae 
fecit et passus est pro Carolo primo'. 3 Thirteen years later, writ- 
ing to Pope, he recalled the 'poor old gentleman's* sufferings.* 
Even Deane Swift, a great-grandson, kept up the proud tone, and 
called his forebear 'one of the greatest examples of patience, con- 
stancy, and spirit' to have appeared in England. 6 

In addition to these social and political traditions, the family 
had two literary ties which Swift would not consider negligible. 
The obvious one was the Dryden connection. But besides that, 
Sir William Davenant was allied to them through his daughter 
Mary, who married Swift's non-migrating uncle Thomas, the 

1 A, G. Matthews, Walker Revised (Oxford 1948), p. 196, where there are further 

* King's Inns, Dublin, MS. The Black JBooJe, f. 203. 

* D. Swift, Appendix, p. 22, * Sherburn rv. 174, * D. Swift, p. 14. 



clergyman. This uncle took his B.A. degree at Balliol, establish- 
ing a pattern which his son was eventually to imitate for the M, A. 
degree and which through him in turn influenced Jonathan Swift 
to follow a similar course at Hart Hall, Oxford. Uncle Thomas, 
after some service as a country parson, managed to obtain a 
church in London; but like Swift's father he died both unex- 
pectedly and without accumulating much wealth. At the time he 
was only thirty-six, 1 and left an infant son, also named Thomas, 
who had probably been born in the Davenant family's Oxford 
home. Since this boy, a year or two older than Swift, was to share 
most of his early experiences, the Davenant connection could 
not have seemed remote. In a dry comment upon the aftermath 
of the death, Swift suggests that he appreciated the degree of 
analogy between the bereaved family and his own : c His widow 
lived long, was extremely poor, and in part supported by the 
famous Dr South, who had been her husband's intimate friend.' 3 
The child of such a marriage would have seemed to live in a 
situation close enough to Swift's to allow each boy easily to 
identify himself with the other. 

For a happier story. Swift could have looked to the career of his 
uncle Godwin (1628-95). Yet Godwin's hopeful beginnings re- 
veal just the possibilities which failed to excite his nephew. This 
uncle had received a legal education, entering Gray's Inn as a 
student in 1 6so. 3 His success as a lawyer was enhanced by his mar- 
riage to Elizabeth Wheeler, first of his four wives; for she had a 
distant connection with the Duchess of Ormonde. 4 There is no 
doubt that the great Duke 'who well the noblest objects knew 
to choose' used his influence on. Godwin's behalf. In Ireland, 
Godwin flourished. His application for the office of filacer in the 
Court of Common Pleas was granted in i66o. 6 This post, origin- 
ally charged with the filing of writs, the issuing of processes, etc,, 

1 See Foster; Henry Isham Longden, Northamptonshire and Rutland Clergy (North- 
ampton 1938-52), xnr, 131; George Hennessy, Novttm Repertorium Ecclesiasticwn 
Parochial e Londinense > 1898, p. 143; Arthur H. Nethercot, Sir William Davenant 
(Chicago 1938), pp. 345, 406; Ball vi. 215. 

* Autob., f, 5 V . * Ball i, 371-2, n. 3; vx. 212. 

4 Mason, p. 227; Ball i. n, n. i. * SPL> 1660-2, p. 83; ztfS^-j, p. 469. 


had become a sinecure ; and it was later held by Godwin's young- 
er brother, William. 1 Furthermore, among the estates which Or- 
monde regained was the county palatine of Tipperary, to which 
an attorney-generalship belonged. This office too he bestowed 
upon Godwin Swift, though only until i668. 2 Neither appoint- 
ment should imply that Godwin actually appeared much in 
courts of law, whether in his public duties or his private career: 
Swift says he was 'an ill pleader, but perhaps a little too dextrous 
in the subtil parts of the law'. 3 His business was that of attorney or 
solicitor, and people often employed him as Deane Swift, his 
grandson, writes to draw up 'great settlements'. 4 When in 1664 
he was slandered as having cheated when drafting a will, he pe- 
titioned Ormonde for freedom to take action against his accusers. 5 

1 Ball i. a-9, n. 4. 

3 Ibid. vi. 2 1 a. A letter of 1 3 June 1668 from Ormonde to Godwin Swift, removing 
him, is among the unpublished Ormonde MSS. in the National Library, Dublin. 
3 Autob., f 9-9 v . 4 D. Swift, Appendix, p. 32. 
6 HMC., rx, pt. n, pp, i27a, 

Chapter Four 

When her son was born, 30 November 1667, Abigail 
>wift was a widow. Her husband had died in March or 
pril; but, as Swift is reported to have said, the new 
child 'came time enough to save his mother's credit 3 . 1 The elder 
Jonathan Swift had settled in Ireland about 1660 and perhaps 
through his brother Godwin's influence obtained a post at the 
King's Inns. This was the legal society of Ireland, corresponding 
to such English bodies as Lincoln's or Gray's Inns ; Sir John 
Temple was on the governing board. Here the young newcomer 
began as assistant to Thomas Wale, the steward of the society. In 
the summer of 1664 he married Abigail Erick: both bride and 
groom were twenty-four years old. When, more than a year later, 
Thomas Wale died, Jonathan Swift petitioned for his place. By 
an order dated 25 January 1665/6 and signed, among others, by 
Sir John Temple, he was Admitted steward of this house' ; and 
the following day he was made a member of the society. A few 
months later the first child of Jonathan and Abigail Swift was 
born, a daughter named Jane, baptized i May i666. 2 

Swift wrote of his father as a person only that 'he had some em- 
ployments, and agencyes; his death was much lamented on ac- 
count of his reputation for integrity with a tolerable good under- 
standing 9 . 3 On the improvidence of the marriage, however, he 
was almost eloquent; and the personal note in his comment 
should temper any surprise one might be inclined to feel over 
Swift's later attitude toward wedlock : 

1 Pilkington, p* 57. 

* For the facts in this paragraph, see JV. & Q. t 15 Nov. 1947, pp. 496-8. 

* Autob,, f. 6. 


This marriage was on both sides very indiscreet, for his wife brought 
her husband little or no fortune, and his death happening so 
suddenly before he could make a sufficient establishment for his 
family: and his son (not then born) hath often been heard to say 
that he felt the consequences of that marriage not onely through 
the whole course of his education, but during the greatest part of 
his life, 1 

The death of her husband left Mrs Swift dependent upon the 
generosity of her brothers-in-law. Yet even with the assistance of 
William Swift and the encouragement of the officers of the so- 
ciety, she recovered c but a very inconsiderable sum' 3 out of the 
fees totalling something like a hundred pounds due her late 
spouse. If (as her husband's great-nephew unreliably reports) the 
income she was left with came to only twenty pounds a year, it 
could not have supported her family of three. 3 

The relations between Swift and his mother make one of the 
puzzles in his life. His words about her are all in praise, and by 
every account she was an unusual woman. However, only one 
direct indication remains of his attitude towards her, the private 
comment which he wrote in his memoranda when he was noti- 
fied of her death: 

On Wednesday, between seven and eight, in the evening. May 10, 
1710, 1 received a letter in my chamber at Laracor, (Mr. Percivall 
and [Jo] Beaumont being by) from Mrs. Fenton [i.e., his sister 
Jane], dated May gth, with one enclosed, sent from Mrs. Worrall 
at Leicester to Mrs, Fenton, giving an account, that my dear 
mother Mrs. Abigail Swift died that morning, Monday, April 24, 
1710, about ten o'clock, after a long sickness, being ill all winter, 
and lame, and extremely ill a month or six weeks before her death. 
I have now lost my barrier between me and death; God grant I 
may live to be as well prepared for it, as I confidently believe her to 
have been! If the way to Heaven be through piety, truth, justice, 
and charity, she is there.* 

Oixe does not doubt the sincerity of this tribute. There is no 

* Auto&. 9 f. 6. a Black Book, f. 25O V . s D. Swift, p. 23. 
4 Nichols x. 104-5, from a page now lacking in Swift's account book for 



equivocation, and Swift expected no one besides himself to read 
the words. His affection and respect for Abigail Swift are deliver- 
ed with intense feeling. To the 'piety, truth, justice, and charity* 
which he lists as characteristic of her, he added, in a letter almost 
two decades later, the trait of prudence. 1 People who knew the 
family have told some anecdotes which give her also a liveliness 
unmentioned in her son's allusions. The most entertaining is John 
Lyon's story of her visit to Dublin, after Swift received a benefice 
not far from the capital. She stayed with Mrs Brent, a printer's 
wife, who was later to be Swift's housekeeper. Upon her arrival, 
Mrs Swift disclosed to her landlady that she had an. admirer in 
Dublin who had been corresponding with her, and that he would 
soon come to pay his addresses. Of course, the gentleman who 
presently appeared was Swift. His mother spoke with him alone 
for a while, then called Mrs Brent in. The son was presented as the 
lover, supposedly the only one, of Mrs Swift. 'The doctor smiled 
at his mother's humour', writes Lyon, and he 'afterwards paid his 
duty to her every day as expected by Mrs. Brent'. 2 

It is easy to accept the portrait of Mrs Swift as a good Christian, 
honest and prudent, but cheerful and willing to make up an in- 
nocent practical joke. Swift's personality reflects such influences. 
The difficulty is in her treatment of him during his childhood. Or 
perhaps it is in her lack of treatment. Among good English fami- 
lies it was customary for infants to be boarded out with wet nurses 
who were usually country women. How common the practice 
was, may be judged from Jeremy Taylor's protest against the 
physical and moral corruptions to which the vulgarity, ignor- 
ance, and carelessness of most nurses exposed their victims. 3 Al- 
though his attack has antecedents going back to Plutarch and be- 
yond, the custom persisted long after the reign of Queen Anne. 
The Spectator's complaint is well known. 4 Swift's nurse was not, 
however, the monster whom Taylor and Steele feared. His mo- 
ther, moreover, was evidently wise enough to have him breast- 
fed rather than submitted to the usual alternative, rearing by 

1 Ball rv. 55. * Lyon, f. 9 of preliminaries. 

9 Works* 3rd cd., 1839, ir, 38. 4 is Dec. 1711, by Steele. 



hand on bread-and-water pap, or a flour-and-sugar posset, or 
syrup of violets. 1 

In Swift's case the difficulty was too much, not too little, atten- 
tion from the foster-parent. 'Irish nurses', an English visitor wrote 
in 1 68 1, *are very tender and good to the children of others of 
higher degree, and most commonly their love is more to them 
than to their own.' 2 Although Swift said his nurse was English by 
birth, her attitude fits the native pattern. In fact, she fell in love 
with him. To quote the victim's own words, 

his nurse who was a woman of Whitehaven, being under an abso- 
lute necessity of seeing one of her relations, who was then extremely 
sick, and from whom she expected a legacy; and being at the same 
time extremely fond of the infant, she stole him on ship board un- 
known to his mother and uncles, and carried him with her to White- 
haven, where he continued for almost three years. For, when the 
matter was discovered, his mother sent orders by all means not to 
hazard a second voyage, till he could be better able to bear it. 3 

He adds that the nurse taught him to spell and that when he was 
three years old, he could read e any chapter in the Bible' (meaning, 
perhaps, the New Testament only). While the authority for the 
kidnapping is unimpeachable, the comment of Emile Pons, that 
it sounds like 'un conte defee$\ seems pertinent. 4 Swift wrote out 
the received version when he was over seventy and notoriously 
forgetful ; his fragment of an autobiography has as many errors as 
facts. 5 The details must be handled with caution. In 1734 he 
wrote, *I happened indeed by a perfect accident, to be born here 
[i.e., in Ireland] , my mother being left here from returning to her 

1 Rosamond Bayne-Powell, The English Child in the Eighteenth Century, x 939, p. 163. 

2 Thomas Dineley, from extracts in Journal of the Kilkenny . . . Archaeological 
Society, new ser., i (1856-7), 185. 

3 Autob., ff. 7-7 v . The V in 'uncles* is not certain, * Pons, p. 117. 

5 D. Swift, c. 1 754, says the fragment was composed 'about six or eight and 
twenty years ago* ; but in the fragment itself Swift says his grandfather's house, 
dated 1636, 'is above a hundred years old* (f. a v ); and in a letter of April 1739 he 
echoes the language of the fragment in a way that suggests he was writing both 
near the same time (Sherburn xv. 174). Monck Mason also mentions an extract 
from the Mercurius Rusticus account of the vicar of Goodrich, endorsed by Swift, 
'Memoirs of my grand-father, Thomas Swift, by Mr. Lyon; April 1738* (p, 228, 



house at Leicester, and I was a year old before I was sent to 
England. 91 

That Swift accepted the kidnapping as truth cannot be denied. 
He tells it in a straightforward passage of a factual account. An 
acquaintance close to him in his last years says that Swift loved 
Whitehaven as though he had been born there. Even toward the 
end of his life, in 1 740, he was happy to hear that a merchant from 
the town was in Dublin with his son and daughters. Swift 'in- 
vited them to dine with him and paid them many civilities while 
they stayed 5 . 2 Deane Swift, who speaks here with authority, says, 

His return to Ireland about three years after [the kidnapping], 
gave occasion to many ludicrous whims and extravagancies in the 
gaiety of his conversation. Sometimes he would declare, that he 
was not born in Ireland at all ; and seem to lament his condition, 
that he should be looked upon as a native of that country; and 
would insist, that he was stolen from England when a child, and 
brought over to Ireland in a band-box. 3 

Whenever the child was returned to his mother, he did not re- 
main long with her. At the age of six around 1673* he was 
sent to the excellent grammar school in Kilkenny (a little over 

1 Ball v. 64. a Lyon, p. 10. 

3 P. 26. Letitxa Pilkington gives what reads like a garbled recollection of Swift's 
stories about his infancy: 'The account I have frequently heard the Dean give of 
himself was that he was born in Hoy's Alley, in Warburgh's parish, Dublin; his 
father was a lawyer, and, returning from the circuit, he unfortunately brought 
home the itch with him, which he had got by lying in some foul bed on the road* 
Somebody advised him to use mercury to cure it, which prescription cost him his 
life in a very few days after his return. The Dean was a posthumous son to this 
gentleman, but, as he said, came time enough to save his mother's credit. He was 
given to an Irish woman to nurse, whose husband being in England and writing to 
her to come to him; as she could not bear the thoughts of parting with the child, 
she very fairly took him with her, unknown to his mother or any of his relations, 
who could learn no tidings cither of him or her for three years; at the end of which 
time she returned to Ireland, and restored the child to his mother, from whom she 
easily obtained a pardon, both on account of the joy she conceived at seeing her 
only son again, when she had in a manner lost all hope of it, as also that it was plain 
the nurse had no other motive for stealing him but pure affection, which the women 
of Ireland generally have in as eminent degree for the children they nurse as for 
their own offspring* (pp. 57-6) . 

4 So Swift says. He may be wrong, since boys admitted to Kilkenny College 
1684-6 ranged in age from nine to fifteen; see Hodges, p. 14, n. 


seventy miles south-west of Dublin), which had been established 
by the Ormonde family. While her son was at school, Mrs Swift 
did not remain in Ireland but moved with his sister to Leicester. 1 
She may have visited Dublin and seen him there from time to 
time; however, there is no proof that they met again until he left 
the university. It is certain that if Swift went to England during 
the interval, he did so only once. 2 

One is left with the impression that Mrs Swift had access to her 
sou for the first year of his life, for a year or two before he entered 
school, and for short periods during infrequent visits afterwards. 
The behaviour did not appear odd, as no early biographer of 
Swift picked it out for special comment. He himself implies that 
his mother's conduct seemed normal to him. 3 During the late 
seventeenth century, of course, the rate of infant mortality, the 
teachings of religion, and the demands of a harsh environment 
were still working to make parents less reverent of children than 
they have since become. Yet, even granting such mitigations of 
Mrs Swift's apparent coolness, nobody could argue that Swift's 
early years were suffused with a warm maternal glow. 

If we assume a childhood to be natural when it is passed with 
a father and mother, in a secure family and a settled residence, 
Swift had to deal with shattering conditions. His father was gone 
before the son even appeared. Such a loss would give fatherhood 
unique meaning : missing it so deeply, Swift would expect much 
from those whom he set in his own father's place; and he would 
therefore feel repeatedly disappointed by these older men. On 
those whom he loved, he would bestow, as his best gift, a father- 

1 Forster, p. 40; although this fact is always assumed, it has not been proved; one 
might have expected the mother to wait until the son entered the university. Since 
Jane Swift was married in Dublin 1699 (though she settled at last in England), her 
youthful connection with Ireland could not have ended in the 1670*8. Lyon says 
that she was at her uncle's (presumably William) in Bride Street, Dublin, when she 
wrote her letter of a6 May 1699 (Lyon, f, 7 of preliminaries; Ball i. 30 and n. a). 

* See MLN., LXV (Apr. 1950), 256-7. 

3 None of his letters to her is extant; but the heading of one which he never 
completed is preserved on a leaf used for notes: 

Moor Park August the 5th 1698 
Dear Mother. 
(Bulletin ofihejohn Rytands Library, xxxvn [1954^-5], 372.) 



liness too stern for their needs, or a father's direction where they 
looked for a lover's softness. 

The early loss of his true mother; the loss, a few years later, of 
his nurse, or foster-mother; and the loss again of his own mother, 
would accustom him, as a natural pattern, to separation from 
women who loved him, to sudden partings with women whom he 
felt drawn to. We may speculate that he would therefore tend to 
forestall the misery of being thrown off by curtailing an attach- 
ment before the woman should do so. His sister's advantage in 
seniority, and in continuously enjoying their mother's presence 
while he lived among cousins or strangers, was complicated by 
the odd design of all their travels : for after a long separation dur- 
ing which he must have been reminded of them by his nurse, he 
rejoined the mother and daughter in circumstances which almost 
surely made an anti-climax; he then was sent away from them to 
the Kilkenny School, and finally found they had gone to England 
together. When he was in England alone, his mother and sister 
were together in Dublin; when he was in Ireland alone, his 
mother and sister were together in Leicester. It would seem nor- 
mal to him (we may further speculate) that a woman he loved 
should move away as he approached, or that he should move 
away from one who approached him; it would seem normal that 
a child should cut himself offfrom his family and choose to live in 
another household. Finally, if we assume that he resented his 
mother's apparent neglect of him, but that he could not admit 
such a complaint openly (even to himself), an obvious question is 
what became of the feeling. Unless Swift was remarkably differ- 
ent from other children, he probably attached the resentment to 
the sister who had enjoyed his mother's attention while he was 
missing it. We should not feel puzzled if as an adult he acted 
coldly towards Jane, especially if his coldness should seem (to 
him) connected with her making a marriage as imprudent as 
their mother's. 


Chapter Five 

Swift's knowing how to read at an early age need not imply 
that he was unduly precocious. The achievement was ap- 
parently a goal not infrequently set by eager parents or 
guardians. Aubrey reports that Katharine Phillips c had read the 
Bible thorough before she was full four yeares old 3 . An early 
seventeenth-century writer on education was annoyed that gram- 
mar schools had to be troubled with the very young in the first 
place, and thought no child should enter until he could read the 
New Testament in English. 1 One may reasonably say that Swift 
would have learned at home how to read and write English, also 
a little arithmetic. He would have become familiar with his 
primer (or some equivalent), the psalms, and of course the New 
Testament. By the school statutes of 1684, the Kilkenny students 
had to know their Latin accidence before entering, and to be 
ready immediately for the study of grammar.'- 5 

At this school a high level of preparation was probably expect- 
ed of the boys. Founded in the sixteenth century by the eighth 
Earl of Ormonde, here was certainly the finest institution of 
its kind in Ireland. The best of the *old English' gentry of the 
Pale sent their sons to it; and among the students had been 
Richard Stanyhurst, the Elizabethan historian and translator 
of Virgil ; Peter Lombard, historian and Roman Catholic Arch- 
bishop of Armagh; and Luke Wadding, a Franciscan scholar 
and a fervent organizer-in-exile of Irish Catholic patriotism. 

1 Foster Watson, 'The Curriculum and Textbooks of English Schools in the First 
Half of the Seventeenth Century*, Transactions of the Bibliographical Society, vr, pt. i 
(Oct. i goo-Mar. 1901), 160. 

3 Edward Ledwich published the statutes in History and Antiquities of Kilkenny, in 
Collectanea de Rebits Hibernicis, ix (Dublin 1781), 509-17. 



It was the normal training ground for the boys of Swift's 

The town itself stretched mostly between the Ormondes' 
castle, high on a cliff overlooking a bend in the river, and the 
thirteenth-century cathedral, on a rise about a thousand yards 
north-west. We know what impression the town made upon a 
visitor in 1709: 

Kilkenny is a large straggling city with Irishtown. I think it has as 
much ground under it as any I have seen except Dublin. The houses 
are but ordinary. There are here several old abbeys and buildings, 
one of them the cathedral of St Kenny's [i.e., St Ganice's], a fine 
old Gothick building. Kilkenny is finely watered by several excel- 
lent springs, by the noble river Nore and two others. From a bridge 
on the former you have a fine prospect of the castle belonging (as 
does most of the town) to the Duke of Ormond. 'Twas built [i.e., 
greatly rebuilt, at the Restoration] by the old duke. It is finely 
situated to the river but in no other respect answerable to the char- 
acter it bears. There is not one handsome or noble apartment. The 
rooms are dark and the stairs mighty ugly. 1 

The school stood in the west of the cathedral churchyard 2 and 
accommodated from fifty to sixty boys, most of them between 
nine and fifteen years old. Some, Swift was to know in later 
life: his cousin Thomas, William Congreve (admitted only 
half a year before Swift left), Francis Stratford, who as mer- 
chant and speculator was to make and lose a hundred thousand 
pounds. 8 

At the head of the school stood men who provided these boys 
not only with a stern set of moral principles but also with immedi- 

1 Thomas Molyneux, in T.G.D. MS. r. I. 3 (Natural History of Ireland, pp. 91-2). 
I have freely made deletions and altered spelling and punctuation. 

2 It is sometimes said that the school in Swift's time was already on its present 
site, across the river from the old town. But all the references to the new site seem 
to date from 1684, except for one dated 1666 which I take to be a misreading of 
1686. The date on the entrance gateway, on the register, and on the statutes is 1684. 
But see John Browne, 'Kilkenny College*, Transactions of the Kilkenny Archaeological 
Society, i (1849-51), 221-9 (Dublin 1853). 

8 See Burtschaell. Stratford is entered at T.C.D. as coming from a *Mr. Wilson', 
and not Ryder; however, Swift twice says they were at 'school and university' 
together (Journal, 14 Sept. 1710, i Mar. 1711/12). 



ate examples of how well-directed talent, elevated by responsive 
patronage, could climb to power and dignity in the established 
church. In addition, one of the men eventually provided a show- 
piece of how little integrity of character might underlie both the 
expressions of morality and the happy career. A prudent use of 
hypocrisy and a wise choice of sponsorship were effective, if not 
essential, factors of ecclesiastical prosperity. 

During Swift's stay, the Kilkenny School was under two suc- 
cessive masters of extraordinary abilities: Edward Jones and 
Henry Ryder. Jones (1641-1703), who came from Wales, had 
been educated at Westminster School and at Trinity College, 
Cambridge, winning scholarships at both. He took his B.A. de- 
gree in 1664 and was elected a fellow of his college three years 
later. Going to Ireland as domestic chaplain, to the Duke of Or- 
monde (then lord lieutenant), Jones became master of the school 
at Kilkenny in 1670. After ten years, during which his church 
preferments steadily mounted, he left; and in 1683 he was made 
Bishop of Cloyne. Returning to England during the troubles of 
1688-90, he never again settled in Ireland. In 1692 he was made 
Bishop of St Asaph, a charge which he distinguished by Corrup- 
tion, negligence, and oppression*. 1 After considerable litigation 
he was suspended for eleven months (1701-2) and died a year 
after resuming his function. 

The nature of Jones's crimes, though not directly relevant to 
Swift's career, is worth noticing as a specimen of what could be 
accomplished by a thoroughly unscrupulous prelate a speci- 
men about which Swift could not have missed hearing : 

By his own confession he had promoted to a canonry one who had 
been accused to him of crimes and excesses; he had permitted lay- 
men to perform the office of curates ; he had been guilty of a simonia- 
cal contract in the disposal of some of his preferments, and had al- 
lowed his wife to receive money, by way of earnest, for certain pro- 
motions. Besides which, he had been in the habit of appropriating 
to himself a year's profits of vacant livings, on the plea of carrying 



on the lawsuit for the recovery of an advowson a plea, it is almost 
needless to add, never put into practice. 1 

Swift's other master enjoyed both a speedier rise and a deeper 
integrity. Born of a Bedfordshire family, Henry Ryder (c. 1646- 
96) had also been a scholar at Westminster School as well as 
Trinity College, Cambridge. He took his B.A. degree in 1666/7 
and went to Ireland in 1672 to be master of the Free School in 
Dublin. About eight years later, Ormonde persuaded him to 
leave that post and come to Kilkenny. Ambitious, like Jones, for 
ecclesiastical preferment, Ryder did not long remain a school- 
master. He managed to accept a prebend of Ossory (1681) with- 
out giving up his work in Kilkenny ; but in 1 683 he was collated to 
a prebend of St Patrick's Cathedral, and he moved back to Dub- 
lin, soon after. His progress in the church lasted until he became 
Bishop of Killaloe, i6g3. 2 

For fourteen years, therefore, the Kilkenny School was headed 
by men from Trinity College, Cambridge. Thomas Cartwright, 
a founder of the Puritan, party in England, had been a fellow of 
Trinity, where his influence was powerful. The tradition of Cal- 
vinism was a Cambridge tradition, and the movement was al- 
most as strong in Trinity as in Emmanuel College, from which 
came the fathers of Puritan New England. In the selection of a 
schoolmaster, theology was a high consideration; and Jones and 
Ryder may well have brought to Kilkenny some of the austerity 
and intensity which they had met as undergraduates. 

In neither the curriculum nor the pupil discipline of Anglican 
schools at this period was there much variety. Clothing, games, 
buildings, the government of masters or ushers might indeed 
differ widely. But the studies and the moral training of the boys 
did not. By combining what we know of Restoration pedagogy in 
general with the details of a Kilkenny School charter given only 

1 Condensed, with omissions, from D. R. Thomas, The History of the Diocese ofSt 
Asaph, 1908, i. 137. 

* Venn; G. F. R. Barker and A. H. Stenning, Record of Old Westminsters, 1928, n, 
811 ; H. J. Lawlor, Fasti of St Patrick's (DundaJk 1930), pp. 139, 254, 



two years after Swift left, we can deduce the elements of his 

We may, on all grounds, assume, therefore, that the pupils at 
Kilkenny underwent a solid course of education. It would prob- 
ably have started in the fifth form with the reading, writing, and 
speaking of Latin. When they reached the fourth form, the child- 
ren would learn syntax, parsing, and construing, while con- 
tinuing to improve their earlier accomplishments. A book com- 
monly read at this point was Aesop's Fables. Greek was often added 
in the third form, where the readings might include the New 
Testament in Latin and Greek; students would learn to write 
letters and to make double translations (English-Latin and 
Latin-English) ; they would be embarking upon such standard 
authors as Caesar, Cicero, and Terence. The second form would 
probably include oratory, the composition of Greek verses, 
Greek-Latin double translation, and more difficult authors. In 
the first form Hebrew might be introduced* The study of rhetoric 
in the upper forms was particularly important because of the 
value, in adult life, of effective writing and public oratory. In 
fact, the highest product of the written exercises was the formal 
theme ; and the goal of the spoken exercises was the oration 
both, of course, in Latin. 

Throughout the curriculum, naturally, a fundamental aim 
was the inculcation of Anglican doctrine and morality, 1 c The 
teaching a little Latin and Greek . . . would not deserve the name 
of Christian education; were not at the same time the greatest 
regard had to the subduing of disorderly passions > the rectifying of 
perverse inclinations , the implanting of virtuous habits, and the secur- 
ing of them by religious principles, which are attainments far be- 
yond all the most curious arts and profoundest sciences in the world/ 2 
The sort of character-training which accompanied formal in- 
struction at home and in school may be gathered from one of the 
numerous "advice* books of the period, implicitly designed 
many of them to instruct those newly come to the heights of 
society in the elegant exercise of their increasing power. It was, 
1 Foster Watson, op. cit., pp. 162-6. * The Royal Grammar (1695) > 



of course, ordinary that students should be punished for bad man- 
ners or indecent deportment; and school statutes can be quoted 
penalizing foul language, cursing, gambling, physical violence, 
infringement of the sabbath, disobedience, and slovenly dress. 1 
For more elusive principles of conduct and morality, we may 
generalize from a work like The Whole Duty of Man. While this 
presents ideals, rather than actual conditions, its popularity indi- 
cates that the doctrines set forth were commonplaces of counsel if 
not of behaviour. 

The Whole Duty, though its authorship is still uncertain, comes 
obviously from a setting of pious, royalist Anglicanism. The pre- 
fatory letter to the first edition (1658) was by the Rev. Dr Henry 
Hammond, brother-in-law to Sir John Temple and favourite 
uncle of Sir William Temple. This gracious gentleman, like 
Swift's grandfather, devoted himself to the cause of Charles I, 
and similarly died just too soon for the Restoration to compen- 
sate him. The book which he introduced was a manual of re- 
ligious conduct for middle-class and humbler families, adapted 
c to the very meanest readers'. 2 If we ponder the number of par- 
sons to whom Swift was related, through both his mother and his 
father, as well as the nature of his own career, we must assume 
that the outlook of such a book was congenial to his family circle. 

Apart from the need of enforcing active piety in each part of the 
daily routine we are recommended to think of Judgment Day 
while dressing 3 the Whole Duty has a specific and significant 
tendency. Overpowering all other teachings is the attack on 
pride. Part of man is spirit, but part is flesh, a lump of corruption, 
subject to illness, pains, and at last to death when it is 'laid to 
rot in the earth'. 4 The soul itself is a battleground where the pas- 
sions and the will have it out with the understanding. Unless God 
should step in, the contest will lean toward evil. Let any man look 
into his heart, says the author, and he will see that his soul is dis- 
eased. The understanding is dark; the will is crooked; and the 
passions are 'disordered and rebellious, even against the voice of 

1 Foster Watson, The English Grammar Schools to 1660 (1908), pp. 135-6. 
a The Whole Duty of 'Man (1738), p. i. * P. 404. 4 P. i. 



his own reason 5 . 1 So treacherous is man's depravity that mar- 
riage has only two purposes : 'the begetting of children^ and the 
avoiding of fornication' . 2 The method of the Whole Duty is to deny : 
it teaches us to shun sin. Nowhere is man safe from the Enemy. 
Sleep itself, if not severely limited, can be e many sins in one' : 
waste of time, injury to body., deprivation of the soul, and viola- 
tion of God's command to labour. 3 The individual is cautioned 
ceaselessly to examine himself, to upbraid himself, to repent. 
Today of course the prayer book still keeps these injunctions; 
but they hardly underprop those manuals of 'life adjustment' 
which seem our nearest equivalent to the Whole Duty. 

Exactly how the school was managed while Swift was there 
we are uncertain. However, the statutes laid down in 1684 have 
been preserved, and they should be in the same spirit as the earlier 
conduct. 4 The pupils' morality was strengthened by attendance 
at the cathedral on Sundays and holy days. Prayers with scrip- 
ture readings were held every morning and evening. For these, 
all the boys assembled at seven in the morning during the autumn 
and winter, six in the spring and summer. Lessons lasted until 
eleven (or ten), when the pupils went through the catechism for 
half an hour. Four more hours of classwork followed in the after- 
noon, beginning at one during the winter, twelve in the spring 
and summer. On Thursday and Saturday, which were half-days, 
the routine ended with the catechism; and the afternoons were 
free for recreation. Otherwise, the twelve-month school year was 
interrupted only by short vacations at Easter, Whitsuntide, and 
Christmas, too short for a journey to England. 

The little time allowed for play outside might be spent along 
the river Nore, where Swift probably suffered the tragedy which 
he was to recall forty or fifty years later : 'When I was a little boy, 
I felt a great fish at the end of my line which I drew up almost on 
the ground, but it dropped in, and the disappointment vexes me 
to this very day.' 5 There is also a credible tradition that the pre- 

1 P. vi. 2 P. 1 73 ; cf. the marriage service in the prayer book, 8 Pp. 204-5. 
4 The outline of the school schedule is borrowed from John G. Hodges, William 
Congreve (New York 1941), p. 18. 
"Ball iv. 76-7. 

[ 4 0] 


occupation which he was later to show with Anglo-Latin word 
games was conceived here, since they were an ancient and uni- 
versal pastime of schoolboys. c He said', Lyon writes, c he first 
learned, soon after he entered the school, these words which he 
termed Latino- Anglice, "Mi dux et amasti cum". This kind of 
writing was afterwards one of those whimsical amusements that 
he sometimes entertained himself with.' 1 Swift's most important 
comrade was his cousin, also fatherless and separated from a 
mother in England. Thomas bore the name of the family hero ; 
he was about the same age as Swift's sister Jane. Although born 
in Oxfordshire, Thomas apparently fell upon his Irish uncles* 
charity when he was five years old, because his father died and the 
widow was left very poor. With such striking parallels to Swift's 
circumstances, he must have been an effective substitute for an 
elder brother; and as they were not only to go on through school 
and university together, but to embark as well upon the same 
career, the relationship could (for a while) only grow deeper. 

What impression did the school make upon Swift ? According 
to his comment more than a quarter-century after he left, it was 
the usual combination of drudgery and release ; certainly these 
memories are less bitter than the reflections on his Trinity Col- 
lege years : 

I formerly used to envy my own happiness when I was a schoolboy, 
the delicious holidays, the Saterday afternoon, and the charming 
custards in a blind alley; I never considered the confinement ten 
hours a day, to nouns and verbs, the terror of the rod, the bloddy 
noses, and broken shins. 2 

I have suggested that Swift's relatives trained him in an austere 
religion and a harsh morality. Against the crazy shifts of material 
circumstance during the Commonwealth and Restoration in 

1 Lyon, pp. 12-13: the translation is, 6 My ducks eat a masticum*. The tradition 
of Swift's carving his name on the sideboard of a seat is too weak to be worth re- 
futing CJohn Browne, e Kilkenny College', Transactions of the Kilkenny Archaeological 
Society 9 1, 1849-51 [1853], P- 229). Yet Lyon says, *In the school-room his name still 
remains, as he cut it on the side board of the seat of his class with his knife, after the 
custom of boys* (p. 13). 

a Ford, 12 Nov. 1708. 



Ireland, one may assume that such rigorous principles came to 
seem the more attractive and absolute to those who held them. 
In a school established on the same principles and managed by 
masters themselves imbued with the spirit of Trinity College, 
Cambridge, we can only suppose that the corruption of the hu- 
man heart became plainer to the young eye. Hard industry was 
necessary to occupy the errant attention; constant self-amend- 
ment was needed to put down pride; and the power of the 
rational will had constantly to be employed against the disorders 
of the passions. If Swift does not mention such lessons, it was 
probably because he took them for granted. Yet to a child born 
fatherless and cut off from his mother, his teachers should have 
seemed parents indeed, especially when they inculcated the 
historical prejudices, the political dogmas, the stern faith, and 
stern morals of his family. Contrariwise, in a more elusive pro- 
cess, through knowing cousin Thomas, rather than sister Jane, 
as an elder comrade, by living among schoolmates and masters 
who were exclusively masculine, and following a regime that 
allowed only a few quick interruptions of the school year, Swift 
would probably have found the remote femininity of his mother 
and sister to be a deepening enigma, with associations of danger, 
allurement and jealousy. 


Chapter Six 

Many but not most of the boys leaving the Kilkenny 
School went on to the university: from April 1685 t 
March 1687 less than a third did so. Almost as man^ 
returned home 'to their father 5 , whatever might be his occupa 
tion. Some were apprenticed to attorneys, surgeons, or merchants 
one to a sadler (in London) . One left e to be a gardiner'. Some be 
came ill; one 'left the 5 class sicke of a consumption". Swift's cou 
sin Godwin left (July 1686) e to be a merchant in Spaine'; hi 
cousin Deane left (August 1688) c to be a merchant in Portugall'. 
Swift and his cousin Thomas proceeded to Trinity College, th< 
University of Dublin. Under the date 24 April 1682 April wa 
the usual month for boys to go from this school to the university 
they were both admitted as pensioners, or paying, boarding un 
dergraduates. 2 

In making this passage. Swift did not leave the province of th< 
Butler family. Ormonde looked after the interests of Trinity Col 
lege as carefully as he guarded those of the lower school which h< 
had refounded. He considered T.G.D. to be first a nursery o 
clergymen, and thought the Church of Ireland should be stafFec 
with its graduates. 3 He saw to it that the College recovered it 
lands in Kerry (which had been claimed by others), and he ob- 
tained additional grants to improve its revenues. 4 Carte's praise 
of Ormonde's devotion to c that learned society' is not excessive 
'He was a vigilant overseer of their discipline and conduct, <. 
powerful encourager of their studies, a generous patron to suet 

1 MS. register of Kilkenny College, in T.G.D. Library. a Burtschaell. 

3 Carte rv. 379-80. 

4 E.g., the *Munster lands* of Bodl. MS. Carte 45, f- 65, and the petition con 
ceraing T.C.D. property, mentioned ibid., f. 77. 



as were educated in that college.' 1 Before entering, Swift may 
possibly have visited his mother in Leicester (or, less probably, he 
may have done so later in his six-year stay at the College) : his 
one voyage to England which we cannot with certainty account 
for would fit most conveniently here. 2 

There were then over three hundred students in the College, 3 
which had been founded at the end of the sixteenth century. The 
buildings, on the grounds of an Augustinian priory, were set 
among green fields east of town, about a mile and a half from the 
castle. Before Swift took his B.A. degree, however, the city, now 
about 60,000 in population, had expanded to include the Col- 
lege itself. 4 

The grounds lay roughly half a mile south of the Liffey, and 
were enclosed by an uneven oblong of a wall, the longer sides 
running parallel with the river. In the west front one of the nar- 
rower sides stood the gate-house. For the college architecture, 
the models had obviously been the English university quad- 
rangles. Centred within the whole walled area, like a picture in a 
large frame, was the original and regular block, erected in 1593. 
Made of thin red Dutch brick, the houses looked more like Cam- 
bridge than Oxford ; they were two full stories high, with large 
dormer windows above in the pointed roofs. The walks edging the 
inner faces of these buildings surrounded a court which was di- 
vided into four plots by a pair of intersecting walks. The space 
was not much larger than e two tennis courts stretching side by 
side'. 6 

In the east and west wings, students were lodged* The south 
side was for the fellows and perhaps some students ; it also housed 
the library. In the centre of the remaining, north side rose the tur- 
reted hall, with the kitchen to the east of it and the chapel to the 
west. Built on to the south-west corner of the quadrangle, the 
provost's residence abutted against the west wing. On the north 
side of the chapel stood a steeple, sole vestige of the Augustinian 

1 Carte iv. 18. * See MLN^ LXV, 256-7. a Maxwell, p. 74 

4 Ibid,, p. 102; cf. Luce, Berkeley > p. 31, and Stubbs, p. 143; also Ogg n. n. 

* Maxwell, p. n. 



priory, kept as a landmark for ships passing up the river. To some 
restive students, the modest, conventual effect of the block could 
not have seemed liberating. But tangible marks of encouraging, 
if erratic, growth stood about. Between the west fagade of the 
quadrangle and the west wall of the entire College, extended a 
'Great Court', bordered on the north-west by three L-shaped 
stories of imposing chambers 'Sir Jerome Alexander's Build- 
ings' erected in the 1 670*8. During Swift's residence, additional 
college chambers were built in the same area by other bene- 
factors. 1 A new hall and chapel, projected by Narcissus Marsh 
provost, 1679-83 were under way (on the same sites as the old) 
when Swift entered. The account of them given by Marsh pic- 
tures scenes which Swift often watched, and reminds one that, 
even in the seventeenth century, scaffolding was a normal feature 
of the academic landscape : 

Whilst I was Provost of the College, both the hall and the chapel 
being too little and straight to receive the number of scholars that 
was then increasing very much each year, I resolved upon building 
a new hall and chapel. I thought it most proper to begin with the 
house of God, and thereupon caused the foundation of a new chapel 
to be laid, and before the structure was half finished I was removed, 
and Dr. Huntingdon compleated the work. In the meantime the 
scholars were forced to attend prayers in the College hall. When the 
chapel was finished, the next work was to build a larger hall, and 
because the old one could not be conveniently enlarged as it stood, 
it was necessary to pull that down and to build a larger in its place, 
both in length and breadth, which was the work of some years. 
Whilst this was doing the scholars having no place to eat in, they 
were forced to make use of the library for that purpose, and because 
the books were not chained, 'twas necessary that they should re- 
move them into some other place. They laid them in heaps in some 
void rooms. 2 

The constitution of the College possessed a history more com- 
plex: than that of its fabric. In its early years, the community had, 

1 Stubbs, p. 125. 

a Condensed from Stubbs, pp. 1 16-17. 1 have not indicated my many omissions 
and changes in spelling and punctuation. 



like most Irish institutions, suffered violent alternations of back- 
sliding and reform. Through the reign of James I, moreover, it 
had retained a close bond with Cambridge ; its teachers had been 
predominantly Puritan; and its religious policy tolerant. When* 
Laud became chancellor of the university (1633)3 a new charter 
was procured from the king, less democratic than the original and 
less tolerant in religion. But a series of ineffectual provosts, the 
earthquake of the 1641 Rebellion, and the alterations of the Com- 
monwealth radically undermined both the discipline and the 
finances. At the Restoration, Ormonde resumed his place as 
chancellor, with Jeremy Taylor (then Bishop of Down and Con- 
nor) as vice-chancellor, and Thomas Seele as provost. The re- 
organization which followed established the College as it stood 
when Swift entered. Although Taylor died the year of Swift's 
birth, Secle continued to be provost, and secured the new order. 
Besides openly grounding him in the fading traditions, hu- 
manistic and scholastic, which made the staple of European uni- 
versity studies, Trinity College re-enforced some moral or social 
doctrines already familiar to Swift and at the same time intro- 
duced him to intellectual currents which were new to him and 
the world. Thus in its own recent history, as well as in changes 
made while Swift was a student, the College illustrated again the 
peculiar mutability of property in Ireland and the dependence of 
Irish policy upon English events. It also revealed once more both 
the difficulty and the overwhelming desirability of preserving 
any stable order in the troubled nation* The heads of the College 
embodied yet again the assortment of contradictory ingredients 
which might contribute to advancement in the church : genuine 
piety and erudition, practised hypocrisy, consistent adherence to 
the schemes of superiors, private asceticism. But through them 
and other senior members of the College, Swift also met for the 
first time the little cloud which was to float eventually over the 
whole world and toward which his mature attitude is among the 
problems that the modern reader of his works has most trouble 
in comprehending : this is the natural philosophy, the new ex- 
perimental philosophy, inspiring the Royal Society and pro- 


ducing in the end our own idea of science. Contrary to the argu- 
ments of many scholars. Swift originally encountered the new 
philosophy not as in any way subversive of the humanistic or re- 
ligious orthodoxy of his elders but as embedded in the same 
groundmass: if he judged by his masters and teachers, it ap- 
peared subversive only of those scholastic speculations which 
came early to disgust Swift himself. Scholars have often connect- 
ed Swift's mockery of virtuoso experiments with men and books 
which he came upon as an adult. Yet we will discover the themes 
of his satire of 'science', including several leit-motivs of both A 
Tale of a Tub and Gulliver 9 s Travels, not merely in the general in- 
tellectual life of Dublin or Trinity College, but above all in the 
opinions and researches of his beloved tutor and lifelong friend, 
Dr Ashe. Finally, if, apart from such factors in the milieu, an- 
other kind of token is wanted of what the College came ulti- 
mately to signify to Swift, we may look ahead into his middle age 
and foresee that one of the methods he would most earnestly con- 
template for the disposition of his estate would be to leave it to the 
institution of Jeremy Taylor and St George Ashe. 1 

Under the statutes, as recast by Taylor, the presiding officer 
of the College still had the title of provost. (As a university con- 
sisting of a single college the corporate body also owned a chan- 
cellor and vice-chancellor.) Besides him, there were seven senior 
fellows with power of election and nine junior fellows. As to 
their pupils, in addition to the two hundred or more pensioners, 
provision was made for seventy scholarship students, of whom 
thirty had to be natives of Ireland 'chosen out of the poorest, if 
they be deserving 9 . 2 In a society so narrow, with rather a small 
gathering of students, the head of the institution must have had a 
direct influence upon the undergraduates. While Swift was in 
residence, there were two provosts, Narcissus Marsh (1679-83) 
and Robert Huntingdon (1683-92). Both men were extraordin- 
ary for their piety and scholarship ; and in their lives will be seen 

1 G. P. Mayhew, 'Swift's First Will and the First Use of the Provost's Negative 
at T.C.D.% Huntington Library Quarterly, xxi (Aug. 1958), 295-322. 

a Robert Bolton, A Translation of the Charter and Statutes of Trinity-College (Dublin 



some of the patterns set most conspicuously before the adolescent 

As an undergraduate at Magdalen Hall, Oxford, Marsh 
( 1 638-1 713) had been used to fasting every week from six o'clock 
Thursday evening until eleven Saturday morning. He was or- 
dained even before the canonical age, having been elected a fel- 
low of Exeter College less than five months after taking his B. A. 
degree. From the principalship of St Alban Hall, Oxford, he 
came to his Trinity College post (1679) through an appointment 
arranged by the great Duke of Ormonde. 

As a provost, Marsh was more conscientious than serene; and 
he welcomed his elevation to the bishopric of Ferns and Leighlin 
(1683). Concerning his resignation from the College, there ap- 
pears in his diary a famous utterance which should speak to the 
heart of a modern university administrator: 

Finding this place very troublesome, partly by reason of the multi- 
tude of business and impertinent visits the Provost is obliged to, and 
partly by reason of the ill education that the young scholars have 
before they come to the College, whereby they are both rude and 
ignorant; I was quickly weary of 340 young men and boys in this 
lewd and debauch'd town; and the more so because I had no time 
to follow my always dearly beloved studies, 1 

Apart from religion, Marsh found his absorbing interests in mu- 
sic, oriental languages, and the new experimental philosophy of 
nature. In the words of his epitaph, he 'dedicated his leisure 
hours to the study of mathematics and natural philosophy; and 
above all was highly skilled in the knowledge of languages, es- 
pecially the Oriental*. 2 

Marsh was associated with William Petty and William Moly- 
neux in the founding of the Dublin Philosophical Society. On be- 
coming Bishop of Ferns (which did not involve abandoning the 
capital) , he by no means gave up this connection; and among his 
letters from the country to his fellow philosophers, one of Marsh's 
characteristic regrets is that he has done little toward his 'history 

1 Maxwell, p. 74. * Biograpkia Britannica^ v (1760), 3050, n. 



of the generation of insects, on which subject I have meditated 
and made observations for many years' : unfortunately, he writes, 
there are few insects in his present neighbourhood (Staplestown) 
'and especially of flies the fewest that I ever saw in any place 
and of those not one rare or unusual'. 1 This gap between hope 
and fruition marks his other contributions to the Society's trans- 
actions a catalogue of experiments to be performed with mag- 
nets, 2 and a critique of a fellow member's hygroscope. 8 

Though Swift's character of Marsh was composed at least two 
decades after their academic relationship ended, the unremitting 
severity of the analysis might suggest a reaction against an adoles- 
cent awe of apparent saintliness and erudition : 

His disposition to study is the very same with that of an usurer to 
hoard up money, or of a vicious young fellow to a wench; nothing 
but avarice and evil concupiscence, to which his constitution has 
fortunately given a more innocent turn. He is sordid and suspicious 
in his domestics, without love or hatred; which is but reasonable, 
since he has neither friend nor enemy. . . That which relishes best 
with him, is mixed liquor and mixed company; and he is seldom 
unprovided with very bad of both. He is so wise to value his own 
health more than other men's noses, so that the most honourable 
place at his table is much the worst, especially in summer. It has 
been affirmed that originally he was not altogether devoid of wit, 
till it was extruded from his head to make room for other men's 
thoughts. He will admit a governor, provided it be one who is very 
officious and diligent, outwardly pious, and one that knows how to 
manage and make the most of his fear. 4 

Nevertheless, Marsh had one abiding influence upon the 
young genius which Swift may never have suspected. For it was 
in the Institutiones logics written by the provost 'in usum iuven- 
tutis academicae Dubliniensis' that he studied those common- 
place examples and ancient truisms which he was to manipulate 
in his most brilliant satires : 'homo est animal rationale' ; e nullus 
equus est rationale'; *si sirnia non sit irrationalis, est homo*; 

1 T.C.D. MS. Molyneux I. i. a, flf. 75-75 v , letter of 30 Jan. 1684. 

2 B.M. MS. Ad. 481 1, ff. i6o v , i2 v -3. Ibid., f. I74 V , 
4 T. Scott an. 189-90. 



'solum animal rationale est disciplinae capax 3 ; and those apos- 
tolic names 'Johannes, Petrus, Thomas' used as examples of 
individual men. 1 

After the usual withdrawal to England during the 'troubles' 
of 1689-90, Marsh returned to stay in Ireland, becoming in turn 
Archbishop of Cashel, of Dublin, and of Armagh (1703), and 
therefore Primate of all Ireland. He died a bachelor and was 
buried in St Patrick's Cathedral. 

Robert Huntingdon (or 'Huntington', 1637-1701), Marsh's 
successor as provost, was the son of a curate. He went up to Mer- 
ton College, Oxford (1652), and was elected a fellow the same 
year he took his B.A. degree (1658). Huntingdon's interest in 
biblical studies and oriental languages was so passionate that he 
accepted an appointment as chaplain to the English 'factory' (or 
trading settlement) at Aleppo in order to satisfy it. Arriving there 
in 1671, he continued to live and travel in the East for ten years. 
During this time his main occupation, in addition to study and 
worship, was collecting manuscripts, for himself as well as others. 
Narcissus Marsh, then for the most part in Oxford, was one of his 
constant correspondents ; others were John Fell, Edward Pocock, 
and Edward Bernard. When Huntingdon came home ( 1 68 1 ) , he 
took up his Merton fellowship again. But when Marsh resigned 
from the provostship of T.C.D., Huntingdon was urged to accept 
it, and finally, reluctantly, did so. To close friends, he complained 
as Swift was to do in middle age that his residence in Ireland 
seemed an exile. 2 

Like Marsh, Huntingdon admired the new natural philosophy, 
and the first formal meeting of what became the Dublin Philoso- 
phical Society was held in his lodgings. The significant aspect of 
his participation seems to have been lending his authority to the 
membership. His more direct contributions are not dazzling: he 

1 Dublin, 1 68 1, pp. 116, 185, sig. AS, pp. 175, 42. The Bodleian Library's copy 
of Marsh's book is endorsed on the title page *donatu a Revf Authore D* Nar- 
cisso Marsh, S.T.P.' (8 A. 138 Art). The important discovery of Swift's allusions 
to the commonplaces of the old logic was made by Professor R. S. Crane; all but 
the third of my Latin sentences are examples suggested by him. 

Stubbs, p. n 8. 



read the group an account of porphyry pillars in Egypt, and gave 
them two bottles of Connaught mineral waters to be experi- 
mented upon. 1 

In 1688 Huntingdon fled from Ireland. Although he returned 
for a short stay after the Battle of the Boyne, he soon gave up the 
provostship (1692) and re-established himself in England. Un- 
tempted by a bishopric at this point, he refused a see and took an 
English benefice instead. But in the year of his death, he did ac- 
cept the bishopric of Raphoe in Ireland (1701). 

If such men guided the young Swift indirectly, the person who 
most intimately affected his evolving character was his tutor and 
lifelong friend, St George Ashe. Twenty students of Swift's class 
were under this one scholar's tuition. 2 Though Ashe's family came 
from Wiltshire, his father's estate lay in county Meath, Ire- 
land, and he himself was born in Roscommon. An older brother, 
Thomas, inherited the family property, worth about a thousand 
pounds a year ; a younger, Dillon, took orders and was made vicar 
of Finglas : Swift went to college with Dillon and made fast 
friends of both. 

Ashe received his B.A. degree from T.C.D. in 1676, and was 
elected a fellow of the College three years before Swift's admis- 
sion. Mathematics and the new, experimental philosophy were 
his peculiar interests : he was chosen professor of mathematics 
(1685) 5 and a fellow of the Royal Society, and he followed Wil- 
liam Molyneux as secretary of the Dublin Philosophical Society. 
To the Philosophical Transactions of the Royal Society, Ashe con- 
tributed a number of papers, some while Swift was an under- 
graduate. One may, alas, estimate the depth and virtuosity of 
these researches from their subjects, which include a solar eclipse, 
a man whose fingertip bled periodically, and "a girl in Ireland, 
who has several horns growing on. her body'. 4 

It was while Swift lived under his guidance that Ashe worked 
most actively for the Dublin Society. During the icy winter of 
1683-4, he compared the effects of freezing upon eggs, urine, and 

1 B.M. MS. Ad* 481 1, ff. 162, i6a v . a Stubbs, p. 143, n. 8 Ibid., p. 1 15. 
4 Philosophical Transactions, abridged ed., 1809, ni, 86, 156-7, 229-30. 


other substances. 1 Soon after, he read a paper arguing that mathe- 
matics can give more certainty than any other form of reasoning; 
in its echoes of Descartes and its contempt for scholasticism, his 
argument seems perhaps more earnest than original. But if we 
think of how such themes were to scatter themselves over Swift's 
fantasy of Laputa (not to mention other works), we shall find the 
reasonings of Ashe to be of peculiar interest. Mathematics, he 
writes, is supreme, 

because quantity, the object about which it is conversant, is a sen- 
sible obvious thing, and consequently the ideas we form thereof are 
clear and distinct and daily represented to us in most familiar in- 
stances; because it makes use of terms which are proper, adequate, 
and unchangeable ; its axioms and postulata also are very few and 
rational ; it assigns such causes and generations of magnitudes as 
are easily apprehended and readily admitted; it rejects all trifling 
in words and rethoricall [sic] schemes, all conjectures, authority, 
prejudices, and passions ; lastly, so exquisite an order and method 
in demonstrating is observed, that no proposition is pretended to 
be proved, which does not plainly follow from what was before 
demonstrated. 2 

In the summer of 1684, Ashe reported his detailed observations 
of a solar eclipse; in the autumn, he tried to demonstrate that it 
was possible to square the circle. 3 He gave regular reports on the 
weather and described a spiral barometric glass which would 
'express all variations minutely enough'. 4 He produced a dis- 
course on air which curiously foreshadows the 'Aeolism' of 
Swift's Tale of a Tub though Boyle's experiments are probably 
Ashe's inspiration: 

No one is ignorant how necessary a part of natural history it is to 
consider the various affections, properties, and alterations of air 
since its influence is so universal, that no portion of matter is exempt 
from this ubiquitarian, like an anima mundi it permeates all; even 
the most solid metals possess it, if their sound may be ascrib'd to 
the motion of the included air; and the hardest gems have a con- 

i B.M. MS. Ad. 481 1, f. i. * JbtiL, ff. ai v -a. 

*Ibut. t ff. 148-9, 154, 156. 4 Ibid., ff. 187-8 and passim. 


siderable proportion thereof; their transparency being allow' d to 
proceed from the abundance of their pores ; plants seem to owe their 
very life and being to the air they include, which does as it were 
supply the defect of a heart, in asmuch as ... it is moderately com- 
prest and dilated reciprocally, and begets a kind of systole and 

When Ashe contemplates the new experimental philosophy and 
judges the traditional forms of natural science against it, his tone 
rises to rhapsody: 

Natural philosophy (as it is generally managed) is little else than a 
learned romance, which may amuse and divert, but can never 
satisfy the mind of man, which is fed only by experiment and 
demonstration, and not with gay empty speculations, or spruce 
hypotheses. . . All the search of inquisitive men for so many ages, 
has not been able to fasten upon this Proteus the air, but these few 
following affections, (viz.) heat and cold, dryness and moisture, 
different degrees of gravity, and its spring or elasticity; for the 
happy discovery of some of which, and the improvement of J em all 
we are obliged to the age we live in, and most of all to our own 
countrymen. 1 

Even when Ashe travelled, he did not forget the Dublin society, 
but sent home such descriptions as the following of Cassini's 
telescope : 

He has a telescope of above two hundred feet long, by which he 
made his late discoveries, this he manages by a tower built on pur- 
pose, and has a contrivance of clock work which governs the tele- 
scope so, that it keeps pace with any star or planet. 2 

Later, when he was a bishop, Ashe preserved his hunger to learn 
about new advances in the arts and sciences : in a characteristic 
message to a prospective visitor, he writes, *Do not forget to bring 
your latest Mercuries both French and English, and other new 
books/ 3 However, the fullest and most naive statement of his 
faith in the experimental philosophy belongs to a speech in which 

1 B.M. MS. Ad. 481 1, ff. i35 v --6. * Ibid., f. 140, 

8 T.C.D. MS. Lyons, no. 749: letter of 18 Jan. 1700/1, to William King. 



he asked Lord Clarendon, as lord lieutenant of Ireland, to be 
patron of the Society (January 1686). Here, although he recalls 
the phrases of Bacon, Hobbes, and Sprat, the young don directly 
expresses his own principles. In a series of historical ironies, the 
man who was to be remembered above all as Swift's teacher, not 
only eulogized the institutions which his pupil would ridicule, 
but also reproached those who were already making fun of them : 

*Tis true, knowledge was of old for the most part only the study of 
the sullen and the poor, who thought it the gravest peice of science 
to contemn the use of man-kind, and to differ in habit and manners 
from all others; It was heretofore condemn 5 d to melancholy re- 
tirements, kept as a minor under the tuition of ambitious and arro- 
gant guardians, buried in cloysters, or the more dark obscurity of 
affected jargon and unintelligable cant; Antiquity too was ador'd 
with such superstitious reverence, as if the beauty of truth, like that 
of a picture, cou'd not be known or perceived but at a distance, as 
if theire eyes, like the praepostrus [sic] animalls, were behind them, 
and their intellectuall motions retrograde; No wonder then that 
knowledge did not outgrow the dwarfishness of its pristine stature, 
and that the intellectuall world did continue such a microcosm: for 
while they were slaves to the dictiates of their forefathers, their dis- 
coveries, like water, cou'd never rise higher then [sic] the fountains, 
from whence they were derived. . . [But when, through the found- 
ing of] the Royall Society; captive truth was rescued from its 
former bondage, and clouded knowledge began to shine more 
bright; when instead of words and empty speculations were in- 
troduc'd things and experiments, and the beautiful bosome of 
nature was exposed to veiw, where we might enter into its guarden, 
tast of its fruits, satisfy our selves with its plenty, instead of idle talk- 
ing and wandering under its fruitless shadows; Then philosophy 
was admitted into our palaces and our courts, began to keep the 
best company, to refine its fashion and appearance, and to become 
the employment of the rich and of the great. 

Having thus exemplified several commonplaces which were to be 
satirized in A Tale of a Tub^ Ashe went on to reproach the mock- 
ers of the great enterprise : 

1 Compare especially the Epistle Dedicatory and the Pfefacc. (Of course, I do not 
believe that Swift was directing His satire against Ashe*) 



Even in our cradle, like Hercules, we have suffer'd persecution, 
and been fircely oppos'd by a loud and numerous; though (God be 
thanked) an impudent sort of adversaries, the railleurs and the 
witts . . . we are told by them that our time is spent in vulgar experi- 
ments, in empty useless speculations, which (suppose true) is it not 
necessary that many loads of unprofitable earth shou'd be thrown 
by, before we come at a vain of gold, yet certainly the contempla- 
tion even of flies and shells and the most trifling works of nature 
(which they so much ridicule) is more manly then downright idle- 
ness and ignorance, and the extreams of raillery are more offensive 
then those of stupidity ; They should reflect that all things are cap- 
able of abuse from the same topicks by which they may be com- 
mended, 1 that (besides the ill manners in discountenancing such 
studies as our great master has declared himself the founder and 
promoter 2 ) burlasque [sic] and laughter is the easiest and the 
slenderest fruit of witt, that it proceeds from the observation of the 
deformity of things, whereas there is a nobler and more masculin 
pleasure which is rais'd from beholding their orders and beauty. 3 

Yet in his boldest praise of mathematics and experimental 
philosophy, Ashe never meant to weaken the reputation of the 
church. Of his orthodox piety and his ecclesiastical energy, there 
is solid evidence. During a parochial visitation as Bishop of Glo- 
gher, for example, he boasted of having c set five or six churches 
a building' 4 ; little over a week later, he had 'confirmed near two 
thousand persons 9 , given orders to repair e all the ruined churches 
in the diocese', and placed curates c wherever they were want- 
ing'. 5 Ashe's letters to a fellow bishop regularly end with a sober 
request for his friend's prayers. He deeply, chronically resented 
the growth of nonconformity, complained of the dissenting 
ministers' 'crowds at communions^ and pressed for an end of 
their freedom to perform marriages. 6 As a token of the steady eye 
which he kept on his own clergy, we may just remark him prose- 

1 The reference is to the practice in manuals of rhetoric of arranging under the 
same headings parallel exercises in praise and blame. 

2 Charles II, in the charter of the Royal Society. 

8 T.C.D. MS. Molyneux i* 4. 17, no. 2. For the complete text, see Appen- 
dix D. 

4 T.C.D. MS. Lyons, no. 714: 31 Aug. 1700. 8 Ibid., no. 717: 30 Aug. 1700. 
6 Loc. cit. said passim. 



cuting a lamentable Mr Kirkwood who, in addition to worse 
vices, had c sat two nights and three days drinking at Eniskilliny, 
without ever going to bed'. 1 

During the Revolution, Ashe went to England and became 
chaplain to Lord Paget, English ambassador at Vienna. After- 
wards he was for three years provost of T.C.D. (1692-5), then 
Bishop of Cloyne, Bishop of Glogher, vice-chancellor of T.C.D., 
and Bishop of Derry. In Swift's circle, we shall find Ashe close to 
several persons : he acted as Congreve's college tutor, he ordained 
George Berkeley, and he became a friend of Swift's dearest com- 
panion, Esther Johnson. (In the fabulous scenario of Swift's 'mar- 
riage' to her, Ashe is commonly named as the priest.) Upon his 
death, Addison wrote, c [He] has scarce left behind him his equal 
in humanity, agreeable conversation, and all kinds of learning.' 2 
The possibility that Swift made Ashe the target of his satires upon 
virtuosi must be explicitly dismissed, not only because the same 
ideas appear so commonly and in so many other places, but as 
well because Swift's tone toward Ashe is so uniformly affectionate 
and respectful: the connection between teacher and pupil only 
reveals how early and in how orthodox a setting Swift first came 
up against the ideas. 

1 T.G.D. MS. Lyons, no. 720: 9 Sept. 1700. a Ball ni. 3. 


Chapter Seven 

So pervasive and ancient are the misrepresentations of Swift's 
career as an undergraduate that the first blame for the er- 
rors must fall upon Swift himself; for he originated the tra- 
dition of his own disgraceful academic record. On the other 
hand, the parallel report that his discipline was as regrettable as 
his scholarship belongs to a later authority. Both these stories are 
misleading. However, to arrive at a more reliable account, one 
must examine in some detail the programme of the college, the 
behaviour of Swift's classmates, and a few statistics relating to 
academic marks and disciplinary fines. Since the evidence re- 
mains fragmentary, it is only by such an examination that one 
can either understand what sort of degree Swift finally took, or 
have a glimpse of the sort of student he actually was. 

Swift's schooling was mapped in well-defined areas. 1 Nearly 
all instruction at the College was verbal, and Latin was generally 
the medium of study. At dinner and supper, if the young men 
conversed at all, they had to use Latin. For the B.A. course the 
statutory subjects comprised little besides languages Latin, 
Greek, and Hebrew and Aristotelian philosophy as inherited 
from the Middle Ages by way of Renaissance commentators. On 
the quarterly examinations the undergraduates received marks 
in Latin and Greek, the 'theme', and the department of Aristotle 
which they were studying. Bene, mediocriter, negligenter, male, pes- 
sime represent the range of marks, and haudor vix was available as 
a kind of minus sign. 2 

1 Most of my material on curriculum and pedagogy is paraphrased from Max- 
well, pp. 4953. * Forster, pp. 3840. See Appendix E. 



Undergraduates were assigned to tutors fellows of the Col- 
lege who saw them frequently (sometimes daily) as individuals 
or in small groups, to test them on what they had learned and to 
assign further work. Earlier in the century, the pupils (who were 
commonly adolescent) had often lived in their tutor's rooms. 
Specialists among the fellows lectured (in Latin) on the separate 
subjects. Normally the lecturer in a course would read to the stu- 
dents and then question them to be sure they understood what 
they had heard. This method did not satisfy everybody. When 
Thomas Molyneux (B.A., T.C.D., 1680) was at the University of 
Leyden in 1683, ^ e wrote home to his brother in Dublin, 

The professor of mathematicks and astronomy reads every Wed- 
nesday in the schools, though none of the rest of the professors do. 
Their lectures are usually three quarters of an hour long; not read 
out of a book, as the way is with us and at Oxford, but spoke off 
hand, on such subjects which suppose they have well read and con- 
sidered before they ascend their pulpits. 1 

The Dublin undergraduates had to take notes as they listened, 
and to expand them every week into a Latin commentary on the 
lectures, which was shown to the lecturer. They had also to pre- 
pare a theme, or version of English into Latin, every week, upon 
special subjects. All undergraduates went to lectures on Greek. 
They all had to declaim, two a week in turn, on a subject taken 
from morality or politics. 2 These declamations were held in the 
hall on Saturday or Friday after morning prayers. 

To each class was given a different name : first year, junior 
freshman; second year, senior freshman; third year, junior 
sophister; fourth year, senior sophister. In the first year the con- 
centration was on logic, probably as expounded in Boethius' 
sixth-century Latin, version with exhaustive commentary of 
Porphyry's already bulky Greek Introduction to the Categories of 
Aristotle. In Swift's time, the undergraduates probably used 
Narcissus Marsh's manual of logic as a guide. 

1 *Sir Thomas Molyneux*, Dublin University Magazine, xvra (1841), 474. 
* Bolton, p. 69. 



Aristotle himself was reached in the second year, when the 
lecturer took up some part of the Organon or body of Aristotle's 
writings on logic c as briefly as possible, not going from, the con- 
text after commentators' - 1 The Physics followed in the third year 
and the Metaphysics and Nichomachean Ethics (during Lent !) in the 
last. For the M.A. course the added subjects were mathematics 
and politics. 

To demonstrate their mastery of the material, the students en- 
gaged in strict syllogistic disputations, lasting no more than an 
hour apiece and involving at least four disputants. These were 
scheduled in the afternoons, between two and four. In each class 
there were three disputations every week on a question drawn 
from the lectures. It was essential, therefore, that the entering 
freshmen learn logic before anything else (a little as elementary 
composition is used in American universities today) . 

The most distinctive feature of the academic course at this period 
was the disputation. This the universities had inherited from, the 
Schoolmen. The mediaeval conception of the degree was that of a 
licence to teach. Hence these trials in argument. At the lectures 
there were disputations upon the present or preceding lecture: 
sometimes the students were ordered to dispute more Socratico. Dis- 
putations were also held in public, especially for candidates for 
degrees. The Respondent opened his thesis in logical form; others 
attacked with the aid of syllogisms; the Moderator presided and 
judged. The topics discussed were usually selected from the classi- 
cal authors, the Scriptures, or the works of the Fathers. And, as 
thought became more secularised, such subjects, as 'The com- 
parative value of hereditary and elective princes', or 'The influence 
of the moon upon the tides', were included. 2 

Of course, an odour of sanctity permeated the curriculum. The 
commonest product of a seventeenth-century university was 
clergymen. At six in the morning, students began their day with 
a short chapel service; the first lectures were at seven. Morning 
prayers followed at ten, evening prayers at four. By the statutes, 
undergraduates were required to attend all three services daily; 

1 Bolton, p, 71. * Maxwell, p. 52. 



but the fine for absence was only a penny, and in Swift's class 
even Thomas Wilson, who was to become the saintly Bishop of 
Sodor and Man, missed chapel 'some fifty or sixty times' in less 
than a year ( 1 685-6) . x 

These rites were the responsibility of the masters of arts in holy 
orders (except for lawyers and physicians, all fellows had to enter 
into holy orders within three years of their commencing M.A.). 
They had also, on Sunday and Friday after prayers, to discuss a 
biblical text e in the manner of a sermon', a practice known as 
e commonplacing'. There were special services for Fridays and 
holy days ; on Sunday mornings a sermon, delivered by a resident 
Master, accompanied the prayers. Undergraduates and bache- 
lors of arts had the exercise of giving the blessing before meals, the 
thanks afterward, and the Bible-reading at dinner. Once a week 
the younger undergraduates were also rehearsed in the catechism. 

Swift's description of how he himself fit into this milieu is well 

At fourteen he was admitted into the university at Dublin, where 
by the ill treatment of his nearest relations, he was so discouraged 
and sunk in his spirits, that he too much neglected some parts of his 
academick studyes, for which he had no great relish by nature, and 
turned himself to reading history and poetry. So that when the time 
came for taking his degree of batchlor, although he had lived with 
great regularity and due observance of the statutes, he was stopped 
of his degree for dullness and insufficiency, and at last hardly ad- 
mitted in a manner little to his credit, which is called in that college 
speciali gratia. And this discreditable mark, as I am told, stands 
upon record in their .college registry. a 

This is the voice of an elderly man, past his three score and ten. 
With the frailties of adolescence he has only a restrained sym- 
pathy. Although fourteen was not an amazing age for one coming 
up to a university, it was recognized to be distinctly early. Rich- 
ard Sherlock regretted having entered Oxford at fourteen, and 
said, *To send raw and green youths thither before the tongues be 
learned and understood . . . proves often such a defect, that will 

1 Keble x. 14. * Autob., fit 7 v -8. 



hardly after be made good without double diligence.' 1 6 Double 
diligence', no doubt, was precisely what Swift lacked. In his col- 
lege class he was either the youngest or very nearly so ; yet if he 
had seemed precocious enough at school to go along with his 
elder cousin when Thomas moved to Dublin, the shock of the new 
standard and the new, adult society of Trinity College may have 
troubled him so that he could not demonstrate his true powers 
until much later. 2 

To qualify for a degree, a candidate for the B.A. performed a 
scholastic disputation. But before he was admitted to this final 
and perfunctory ritual, he had to complete twelve terms of resi- 
dence. At the beginning of each term examinations were held in 
the subjects lectured on the term before ; and no term counted to- 
ward a degree until the candidate passed the corresponding ex- 
aminations hand male or better. 3 Of Swift's academic attainments 
there is useful documentation in a list of the marks given to 1 1 9 
students on the Easter term examinations i685- 4 This was one of 
the last examinations before he took his B.A. degree the following 
February. In 'physics' (i.e., the third-year course of Aristotle), he 
received male, in Greek and Latin bene, in Latin theme negligenter. 
By far the most common mark on this examination was media- 
criter; about as many male's appear as bene's, which is not very- 
many, and about the same number of negligenter' 's. Cousin 
Thomas (Swift's elder and a scholar) received mediocriter in every- 
thing. The rarest and severest mark on the roll,pessime (apparent- 
ly failure), was applied to neither. Besides Swift, only six of the 
students examined received bene in Greek and Latin, and only 
eight received bene in more than one subject. Edward Chandler, 
described by Leslie Stephen as e a man of more learning than 
capacity', 5 received mediocriter in all three divisions. The famous 
Thomas Wilson received male in physics, male in Greek and Latin, 
and mediocriter in theme. 

If the Easter 1685 results were typical of Swift's academic 

1 Keble i. 12 (quoting Sherlock). 

a His age on admission was about 14*5, against an average that year of 16-2 
(Stubbs, p. 143, n.) 
9 JV. & Q., 21 Aug. 1875, p. 151. * See Appendix E. 5 DNB. 



standing, he clearly exaggerated his own 'dullness and insuffi- 
ciency'. While he did take his B.A. degree speciali gratia^ or by 
special grace, this favour was not uncommon : four other students 
graduating with him all older than Swift received the same 
dispensation, out of a total class of thirty-eight. The year before, 
two out of thirty-eight took B.A. degrees ex speciali gratia (Febru- 
ary 1685) ; the year after, seven out of thirty-one (February 1687) ; 
a year later, three out of thirty- three (February 1688) .* It is true 
that the fellows once described speciali gratia as a c mark of un- 
worthiness and disgrace' (February 1688) ; but this was in deni- 
gration of a papist upon whom Tyrconnel had forced them to 
confer the M.A. degree, and whom he now wished them to elect 
as a junior fellow. 2 In the mid-nineteenth century the special 
grace was invoked when a c merely technical or purely formal' 
exercise was dispensed with. 3 

In other words, the young Swift did passably well in college, 
but not so well as the septuagenarian Swift could have wished. 
He excelled in those pursuits in which he was himself to be dis- 
tinguished language and literature; he did poorly in what he 
would always dislike abstract philosophy and formal rhetoric. 
The story is hardly unusual. In Anthony Wood, for example, 
there is a similar note on Samuel Daniel, and Wood obviously 
approves of the poet's taste for e easier and smoother studies' than 
Decking and hewing at logic'. Bacon, Milton, and Hobbes show- 
ed the same resentment of university scholasticism. William 
Wake, Swift's contemporary, who became Archbishop of Canter- 
bury, also gives such an account : 'Running too fast through the 
systems of logic and metaphysics, I took a sort of distaste to both; 
and neglecting those studies, applied myself more to the classic 
authors, which were much more easy to be understood and more 

1 T.G.D. MS. General Registry, pp. 238, 249, 257, 271. 

2 Barrett, pp. 5-6. 

8 A". <2? Q., 21 Aug. 1875, p. 151. George Salmon gives another account in Non- 
Miraculous Christianity, 1881, pp. 224-5, n. He seems -wrong, however, first in 
assuming that male meant failure, and secondly in assuming that the 1685 Easter 
term examinations would have been the only ones on which Swift received such 
a mark* 



pleasing to me.' 1 Neither Swift's lack of zeal for the college exer- 
cises nor the fellows' willingness to excuse him from some of them 
will puzzle anyone acquainted with merely that part of the pro- 
cedure which survived into the nineteenth century : 

The proctor on the candidate's visit hands a paper to him contain- 
ing four quartettes of questions; the first quartette comprising 
questions in ethics, the second in metaphysics, the third in casuistry, 
and the fourth in physics; and [here] is the first question of each of 
these four quartettes : 

An omnia peccata paria sint ? 

An sensibus sit fidendum ? 

An bellum possit esse utrinque justum? 

An terra sit immobilis ? 

Upon questions of which these are fair specimens, each candidate 
is required to write twenty-four syllogisms on the wrong side, and 
twelve on the right. When three candidates are thus prepared, each 
with a batch of syllogisms and two theses, viz., one in Greek, upon 
anything at all, and one in Latin, in laudem philosophiae, they proceed 
to the Examination Hall, accompanied by a proctor and a modera- 
tor, whose presence is rendered necessary, lest the disputants in the 
heat of debate should attempt to convince one another by any less 
harmless method than a syllogism. 2 

As for Swift's attitude toward those relatives who maintained 
him at the university, that censure may be due to several circum- 
stances. His uncle Godwin may have been Swift's main support, 
but it was during the boy's undergraduate years that Godwin's 
fortunes went into a decline. Deane Swift says that 'about two 
years' after Swift had been admitted to Trinity College, God- 
win's prosperity began to fail. 3 He reports that although God- 
win's income during the reign of Charles II rose as high as three 
thousand pounds a year, he ruined himself by foolish schemes 
of investment, the silliest being an iron- works in which he was 
cheated by his three partners. Swift's own disgust with this pro- 
ject happens to be on record; explaining the origin of the name of 

1 Sykes, William Wakei. 10. a JV, & Q. s si Aug. 1875, P- *5* 
8 D. Swift, Appendix, p. 41. 



Swandlingbar, c a famous town, where the worst iron in the king- 
dom is made', he says, 

It was a most witty conceit of four gentlemen, who ruined them- 
selves with this iron project. Sw. stands for Swift, And. for Sanders, 
Ling, for Darling, and Bar. for Barry. Methinks I see the four logger- 
heads sitting in consult, like Smectimnius, each gravely contribut- 
ing a part of his own name, to make up one for their place in the 
iron-work; and could wish they had been hanged, as well as un- 
done, for their wit. 1 

Possibly, the family felt no need to give Swift an account of the 
case in detail; so the reduction in his stipend, which would natur- 
ally have followed Godwin's losses, may have seemed like a slight- 
ing of the poor cousin. Furthermore, Swift could not have helped 
transferring to his other guardians some of the resentment which 
he did not apparently like to admit against his own father and 

Since Godwin married four times, had eight children, and 
from them received at least fifteen grandchildren, his assistance 
to Jonathan must have been an act of disinterested kindness. 
Though the tradition of his nephew's bitterness toward him is 
strong, the evidence is all secondary. In fact, there is no conclusive 
proof that Godwin was the boy's mainstay through school and 
university; the early biographers propagate the story, and no 
argument has been advanced against it. Uncles William and 
Adam may well have shared the expense of both Jonathan and 
their other fatherless nephew, Thomas. The first-hand evidence 
all points at William as the most helpful uncle. It was he who 
helped Mrs Swift collect the arrears due her husband at the 
King's Inns; it was he to whom Swift apologized for having 'al- 
ways been but too troublesome' ; it was he who obtained a testi- 
monium from T.C.D. for Swift when he went up to Oxford. 2 

The source of Swift's low spirits remains complex. Poverty 
seems not enough of a reason. When William King was an under- 

1 Davis rv. 282. It is just possible, alas, that Deanc Swift's own account (pp. 
18-20) is based on this source. 
"Ball i. 8-10. 

[6 4 ] 


graduate, he found himself far worse off than Swift Contending 
with straitened means and almost overwhelmed, relatives and 
friends neglecting me, as they themselves were struggling with 
poverty; so that I had scarcely twenty pounds, through the 
whole space of six years in which I stayed at college, from any 
other source than from the college itself 3 . 1 Nevertheless, King's 
temper was unaffected; and if he had any emotional crises, they 
were due to theological qualms rather than shabby clothes. Cer- 
tainly Swift had respect for the university itself; he considered the 
scholarly discipline there to be c much stricter . . . than either in 
Oxford or Cambridge'. 2 Yet he obviously felt that as a student he 
had been unappreciated: after receiving his Oxford M.A. by 
incorporation, he wrote (1692), C I am ashamed to have been 
more obliged in a few weeks to strangers, than ever I was in 
seven years to Dublin College.' 3 


But while Swift's academic standing too has revealed little cause 
for despair or bitterness, there are signs that his deportment was a 
spot where the harness rubbed. Records are available for only a 
little of the time before he took his B.A., when, in his own words, 
he lived e with great regularity and due observance of the sta- 
tutes'. 4 Later, when we do have records, the case is complicated 
by the upheavals throughout the kingdom as James II's policies 
revived memories of 1641: Protestants were fleeing, and the 
government was tampering with the college constitution as well 
as the municipal corporations and similar bodies. Every level of 
society felt the demoralizing effect of sudden remodelling. In 
January 1687 the appointment of Tyrconnel as lord lieutenant 
meant the 'signal for a general exodus of all such Protestants as 
could by any means remove themselves and their goods to Eng- 
land and Scotland'. 6 In the spring Trinity College men heard 
how the vice-chancellor of Cambridge University was deprived 

1 G. S. King, A Great Archbishop of Dublin, 1906, pp. 7-8. 

* Ball in. 309. * Ibid. i. 10. * Autob,, 8. 6 Dunlop, p. 123. 



of his office for refusing to admit a Benedictine monk as master of 

Still, a connoisseur of undergraduate mischief would hardly be 
shocked by the range of cases which the senior fellows of Trinity 
College handled in these years (i 686-8). We find one Mosse 
ordered e to be admonished for climbing the walls, and tampering 
with the locks'. Two scholars, Kieffand Griffith, were accused tf of 
indecent conversation with two women in Stephen's Green; and 
of unseasonable walking in the night' ; Kieff was found guilty of 
fornication and expelled. Another pair of students were ordered 
to be expelled for fighting in town at night 'having formerly 
been guilty of bilking of taverns, and other irregularities' ; and a 
third, a scholar, who had joined in this fracas was to be whipped 
and admonished. Half a dozen were admonished c for bilking a 
tavern'. Another was expelled e for frequenting the town, and al- 
most totall neglect of duties'. A scholar named Spencer wounded 
the porter; it was ordered that he 'be publickly admonisht and 
make a confession on his knees in the hall . . . and be suspended of 
his scholarship ; the profits of his scholarship during his suspen- 
sion are to be paid the porter'. Another student was expelled 'for 
writing, and publishing a scandalous libell on ladies of quality'. 
Several scholars were punished for *a tumult raisd by them in ye 
town'. Four other students received various penalties whip- 
ping, suspension from degree, admonishment Tor lying out of 
ye college and riotous behaviour at an unseasonable time of 
night'. On all these selected instances, a desperate comment may 
be heard in an order of the house, July 1688, that the walls of the 
College should be built three feet higher. 1 

Of all Swift's contemporaries, the liveliest college history be- 
longs to John Jones, who later took degrees in divinity but did not 
become a parson* Three years older than Swift, Jones came into 
the College only a week after him, sat under the same tutor, and 
was graduated B.A. at the same time. Having started out as a 
sizar, he won a scholarship in his third year; yet he needed a 

1 T.G.D. MS. General Registry, pp. 249, 252, 358, 261, 265, 266, 267, 272, 273, 



special grace to admit him to the degree. This in turn leads us to 
the episode of unique interest here ; for, two and a half years after 
achieving the B.A., Jones was briefly deprived of it for a scandal 
which illustrates in cheerful detail one style of discipline main- 
tained in Swift's class. 

At Trinity College, as at other British universities, it was cus- 
tomary each year for a chosen student to deliver a mocking speech 
called the tripos. The orator bore the title terraejilius and spoke a 
macaronic mixture of English and bad Latin. It was Jones who 
performed the tripos at the commencement of July 1688, ridi- 
culing as was the custom the eccentricities of other members 
of the university, particularly the officers and fellows. The wit of 
the production reveals more ingenuity than refinement; one sec- 
tion, for example, relates to the fleasome breeches of the rake 
whom Tyrconnel tried to foist upon the College as a fellow: 

'Tis almost incredible so many cattle should thrive on so bare a 
pasture. Every night he dares venture them off, he's in danger of 
losing them. Once when he lay without them, they crept from the 
garret to the street-door. 

And so forth. Another scene is a tavern conversation between this 
man and a whore. Jones's special object of attack was the amateur 
scientist, like Ashe who himself speaks in a dialogue on eclipses : 
C I conclude, we are all like to be in the dark.* 

Because Jones went too far in his libels, he lost his degree two 
days after his performance, Tor false and scandalous reflections 
in his tripos'. At the end of the week he was allowed to apply for 
restoration of the degree but was suspended instead from the 
benefits of his scholarship probably for a month. 1 Although 
there have been attempts to father Jones's tripos upon Swift, the 
evidence is insignificant. He later became a famous school- 
master and safely grew into a D,D. Many of Swift's relatives were 
sent to him; and one of his pupils, Thomas Sheridan, was to be 
for many years Swift's closest friend. 

For several kinds of common offences, the college authorities 

1 This was the usual period. See Barrett, pp. 20, 66-9. 
F [6 7 ] 


imposed simple fines : a penny for absence from chapel (6 a.m., 
10 a.m., 4 p.m.) ; half-a-crown for failure to appear at nightly 
roll-call in hall (9 p.m.) ; twopence for absence from chapel when 
a surplice had to be worn; twopence for missing catechism; a 
penny for not attending disputations, declamations, or lectures. 
Missing chapel was Swift's (like most students') commonest of- 
fence ; the lectures which he cut most often the record is only for 
broken stretches over two years, largely after he took his B.A. de- 
gree were on mathematics. 1 

In all categories, however, Swift shows up rather badly, wheth- 
er we compare him with Congreve, three years his junior ; Thom- 
as Swift, a scholar; Dillon Ashe, the brother of Swift's tutor; or 
with the two bishops-to-be of the English establishment, Thomas 
Wilson and Edward Chandler. Except for Gongreve and Ashe, 
these men all graduated with Swift. Cousin Thomas's fines are, 
with fair (but not perfect) consistency, lighter than Jonathan's. 
For example, during eight weeks in the summer and autumn of 
1687, Swift paid 45. gd. (not counting iss. sd. cancelled) to 
Thomas's 2S. ?d. (not counting QS. cancelled). During one week 
in the summer of 1686, only a single other student was fined as 
heavily as Swift for missing chapel. Still, we may find some com- 
fort in Swift's friend Stratford (the future merchant), who some- 
times matched and sometimes outdid him. Over three weeks in 
the autumn of 1 686, for example, Stratford was fined 43. i od. (not 
counting 53. cancelled, for missing roll-call) against Swift's 
2s. i id. (not counting is. cancelled, for missing services requiring 
a surplice) ; while Thomas Swift paid merely is. 3d. (not count- 
ing half-a-crown cancelled, for missing roll-call) . 2 

Such comparative figures can be fairly reckoned for only limit- 
ed periods; so the evidence remains inconclusive. Over a three- 
week space soon after he was made B.A., Swift was fined 6s. (not 
counting 4d* cancelled) to Stratford's as. (not counting zs. 6d. 
cancelled). During a similar period in the spring, Swift's fines 
were 33. iod.; his scholar cousin's 35. 8d.; Congreve's is. gd.; 

1 For a description of the available records, see Barrett, p. 8. 
* 13 Aug.-7 Oct. 1687; 10-16 Jul. 1686; 16 Oct.-5 Nov. 1686. 



and Thomas Wilson's is. i id. During two weeks in the summer. 
Swift was fined is. id. (not counting 73. 6d. cancelled, for missing 
roll-call) ; Thomas Swift, gd. (not counting ys. 6d. similarly can- 
celled) ; Gongreve, nd.; Thomas Wilson, yd. (not counting 53. 
cancelled, for missing roll-call) . Compared with Swift's accounts, 
the fines of Dillon Ashe, Thomas Wilson, and Edward Chandler 
seem negligible. 1 

On even this fragmentary record, we should not be surprised 
by outbursts of grosser misconduct, especially with the thunders 
of Revolution approaching. In March 1687 Swift was admonish- 
ed for 'neglect of duties and frequenting the town' ; however, six 
others, including his cousin (and possibly his room-mate) were 
censured with him. This was in a period when the government 
was pressing the fellows to admit as one of their number the un- 
qualified Bernard Doyle, who would not take the oaths. Soon 
came word that the fellows of Magdalen College, Oxford, were 
under pressure to elect a papist as president the beginning of a 
process which ended, early the next year, with the turning out of 
all but one of the fellows of Magdalen and their replacement by 
Roman Catholics. Meanwhile, Tyrconnel went so far as to trans- 
mit to England a bill for repealing the whole Act of Settlement in 
Ireland; but the Privy Council rejected this in Westminster. At 
the end of June 1688 news arrived of the trial and acquittal of the 
Seven Bishops at Westminster Hall. 

If Swift chose this season to break discipline again, he was suit- 
ing his mood to the times. On his birthday, over three weeks after 
the Prince of Orange landed at Torbay, he was found guilty of 
new crimes : he had, with five others, started tumults in the Col- 
lege and insulted the junior dean. The whole half-dozen were sus- 
pended from their degrees. Swift and another boy were singled 
out for carrying themselves even more unbearably e adhuc in- 
tolerabilius' than the rest; and both were ordered to beg the 
junior dean's pardon publicly, on bended knees. 2 After a month, 

1 6-26 Mar. 1686; i-2t May; 19 June-2 Jul. For further comparisons, see 
Appendix F. My dating may be a week out, since it is not clear, in the MSS., 
whether the date which appears on each recto applies to the leaf or to the opening. 

a Barrett, pp. 14-15. While only Swift's surname is mentioned in the sentence, I 



the degrees were restored, though by then all the culprits may not 
have been in residence. 

We may notice here that if Deane Swift was right. Uncle God- 
win suffered a stroke, or c a lethargy', about 1688, and lost his 
speech and memory, 1 The repercussions of such an event could 
only have added to a dependent nephew's anxieties. But with the 
national crisis moving so fast and so close, undergraduate unrest 
would inevitably mount. Among the senior members of the Col- 
lege, too, there was trouble, as Tyrconnel continued to intimidate 
them and tried to win possession of the college silver. 

Not that escapades bolder than Swift's would be out of char- 
acter for a twenty-year-old university student of any century. 
They suggest that he was more than a drudge and less than an 
angel. If he felt 'discouraged and sunk in his spirits', he did not 
show it by hangdog behaviour. Thomas Wilson, who was to be 
practically canonized by John Keble, was not only fined about 
once a week for missing chapel; but for other misconduct, 
'chiefly non-attendance at disputations and lectures', he was 
fined more than seventy times in less than a year; and within the 
same period he was on four occasions charged five shillings for 
being out after hours: yet Wilson, a remarkably pious young 
man s was on the verge of ordination Keble's apologies for him 
make a pretty exercise in casuistry. 2 On Swift's side, we should re- 
flect that it would be an ordinary human pattern for a boy who 

agree with Barrett that the reference is to him, and not to his cousin Thomas. This 
cannot be proved, as Barrett supposes, by the omission of *Se r> (for 'senior') from 
Thomas Swift's name and the retention of *D* (used in the buttery books for 
'dominus* or graduate) during the period of Jonathan's suspension from his degree; 
first, because *Se r ' does not appear consistently, and secondly because another 
student who was once suspended -John Jones kept his title of *D* even during his 
suspension. (In September 1684 a student was similarly suspended from his degree 
for a month because he had affronted the junior dean [T.CJD. MS. Central 
Registry, p. 241]*) However, Barrett's third argument is convincing that one of the 
admittedly less culpable boys, holding a scholarship, was suspended from it; so if 
'Swift*, judged more guilty, received no such suspension, we may assume he held 
no scholarship. In the Senior Buttery Books for these weeks, Thomas appears to have 
paid almost no fines, and one assumes therefore that he conducted himself ex- 
emplarily. Sometimes disciplinary judgments were noted in, the buttery books; no 
such note appears beside Thomas's name. 
1 D. Swift, p. 36. * Keble x. 13-14- 



felt resentful against his family to spite them (not of course de- 
liberately) by making a poor showing at the university. Perhaps 
the old Dean Swift of 1 738 felt embarrassed by his college re- 
cord but remembered the dreariness of his student budget and 
justified one fact by the other. 


Since the administrative class, to which Swift belonged, was to be 
numbered in the hundreds, there could be few of them whom he 
would not meet sooner or later; and since a university degree 
appears among the normal accomplishments of this class, he was 
bound to find at Trinity College many people who would remain 
his acquaintances in later years. It could hardly be otherwise, for 
in Ireland any Protestant of Swift's generation who did take a 
university degree was more likely to choose T.C.D. than Oxford 
or Cambridge. Among the names which reappear, most became 
clergymen, several rising to the episcopal bench; but other car- 
eers are also found : soldier, merchant, author, politician, school- 
master. A sampling is worth tracing two or three in detail ; for 
from a prospect of the fates of Swift's acquaintances, we may 
judge what paths lay open to the young student. Though the 
church did absorb the bulk of the manufacture of the college, and 
other learned professions took most of the remainder, Swift need 
not have considered his ultimate fate to be predetermined. 

The non-cleric whose early career shone most brilliantly in 
Swift's eyes was probably Congreve. But he left the Kilkenny 
School only in April 1686,* moving on to Dublin four years later 
than Swift. There, however, he shared the same tutor and bought 
six times as much beer and wine, at least in the College. After the 
first days of 1687 Congreve is not listed on the books (though he 
may have returned at the end of the year) ; and he left without 
collecting a degree. 2 

Henry Tenison went into parliament. He was the eldest son of 

1 MS. register of the Kilkenny College, in the T.C.D. library. 
z Hodges, pp. aa-6. 



the Bishop of Killala, entered Trinity College two and a half 
months after Swift, and took his B.A. degree in 1 687. Afterwards, 
Tenison moved on to complete his legal training in London, at 
the Middle Temple. In the Irish House of Commons his career 
began when he accepted the election for county Monaghan 1 695. 
Later, he was also chosen M.P. for Louth; and there Swift to 
whom he was 'Harry' paid him a visit. 1 Soon, he obtained an 
appointment as commissioner of the revenue, and seemed hand- 
somely launched on the way of public life. But the tale was cut 
short in 1 709, when he died. For the sake of 'Harry's' memory, 
Swift once gave assistance to a jobless servant of Tenison's. 2 

Francis Stratford, though five years Swift's elder, was admitted 
less than a fortnight before him, came under the same tutor, and 
graduated B.A. at the same time. A quarter-century later, when 
Stratford was the wealthy merchant, he and Swift were, accord- 
ing to dubious tradition, coupled in a Spectator story about two 
schoolmates, one c an impenetrable block-head' and the other 
'the pride of his master' : 'The man of genius is at present buried 
in a country parsonage of eight-score pounds a year; while the 
other, with the bare abilities of a common scrivener, has got an 
estate of above an hundred thousand pounds.' 3 Although the 
theme may indeed be based on the contrast of Swift with Strat- 
ford, it gives an unfair impression of their performances, since 
Swift was undistinguished at T.C.D., and Stratford at least 
graduated without the need of a special grace. On the quarterly 
examination for the Easter term 1685, their marks averaged out 
to something very similar. 4 

The clerics of course came nearer to Swift's road; and we shall 
see that while family influence could do much for them, it could 
not insure success. Dillon Ashe, for example, though addicted to 
puns, claret, and philandering, was brother to a bishop; and his 
brother made Dilly the archdeacon of his diocese. 

Yet Jeremiah Marsh, the son of an archbishop, is a surprising 
case of mediocre achievement. The father was Francis (not Nar- 

1 Ball rv. 454. * Jowrnal, 28 May 1711. 

* 15 Apr. 1712, by Eustace Budgell. * See Appendix E. 



cissus), Archbishop of Dublin. The son, about Swift's age, came 
up to Trinity College (from St Paul's, London), a few months 
after him; but they took their B.A. degrees at the same time. 
Though Marsh followed his father as treasurer of St Patrick's 
Cathedral, and also became Dean of Kilmore, he rose no higher. 
Swift, in December 1 7 1 3 when he still had some power wrote 
that Marsh was "one I always loved, and have shown it lately, by 
doing everything he could desire from a brother' 1 (meaning to 
suggest, not very truthfully, that he was trying to help Marsh to 
the bishopric of Kilmore). Archbishop King in 1714 described 
him to the Archbishop of Canterbury as c a grave, sober, discreet 
man, and would make a very honest bishop'. 2 But Marsh re- 
mained a dean. 

Talent too could not guarantee distinction. John Richardson 
(B.A., 1686), though a gifted scholar, deeply religious, and ener- 
getic in the pursuit of his ends, chose to devote himself to the con- 
version of the Irish-speaking natives. His cause found small fa- 
vour with those irresistible powers who desired no weakening of 
the 'English interest' ; and the summit of his preferments was an 
impoverished deanship. 3 

Neither, as another classmate shows, were political ties in- 
fallible. John Travers, though four years Swift's senior, entered 
Trinity College less than a year ahead of him. In his adult years, 
he apparently made friends both with Swift's beloved Esther 
Johnson and with the family of Bartholomew Vanhomrigh, 
whose daughter was to provoke Swift's gravest indiscretions. For 
many years, Travers was chaplain to the Irish House of Com- 
mons, who successfully demanded a D.D. for him. Nevertheless, 
his highest preferments were a city vicarage and the chancellor- 
ship of Christ Church Cathedral. 

None of these classmates enjoyed the career to which Swift 
eventually aspired. Among those who did were Peter Browne, 
Ralph Lambert, Thomas Wilson, and Edward Chandler. 
Browne became Bishop of Cork; Lambert, Bishop of Meath; 

1 Ball u. 103. * G. S. King, A Great Archbishop of Dublin, 1906, p. 173. 



Wilson, Bishop of Sodor and Man; and Chandler, Bishop of Dur- 
ham. In this quartet of fortunates, the one whose eminence Swift 
found hardest to digest was the Bishop of Meath; for Lambert 
pulled himself up by unswerving submission to a Whig junto 
whose programme depended upon weakening the church : climb- 
ing, Swift once said, is performed in the same posture as creeping. 

Wilson and Chandler pursued their hopes in the English es- 
tablishment, outside Swift's orbit. It is Browne whose story ran 
closest to Swift's ; and by peering ahead at his utterly convention- 
al success, we may judge the depth of Swift's unorthodoxy. 
Admitted to Trinity College less than two months after Swift, 
Browne took his B.A. degree at the same time. In 1692 he was 
elected a fellow and seven years later provost of the College. Here 
the 'austere, retired, and mortified man' 1 enforced discipline 
upon students and fellows alike. When Browne met one of his 
subordinates gownless in London, he did not conceal his dis- 
pleasure; and he was further displeased when the same gentle- 
man (the bon vivant Benjamin Pratt, a classmate of Swift's and 
Browne's, who later became provost himself) failed to raise his 
cap to Browne in the college hall. Browne executed the T,C.D. 
statute obliging celibacy, and deprived a man of his fellowship for 
getting married. Once a pair of undergraduates tried to stop the 
dean of the College from pronouncing judgment on a friend, by 
snatching the paper from the dean's hand; Browne immediately 
expelled them. Nevertheless, he gained something of the reputa- 
tion Swift was to acquire, for 'austere and simple habits, lavish 
and secret . . . charities'. 2 Patrick Delany (who in Swift's middle 
age became his friend and later his biographer) was a disciple of 
Browne's, but apparently emulated his master's quirks with his 

Browne's success was due to a course which Swift was to hear 
ineffectually recommended by his own archbishop the com- 
position of substantial works upon respectable, theological sub- 
jects. 'Say not, that most subjects in divinity are exhausted/ 
William King helpfully advised Swift (1711); 'for, if you look into 

1 Ware n. 296. a DN8. 



Dr. Wilkins's heads of matters you will be surprised to find so 
many necessary and useful heads, that no authors have meddled 
with.' 1 While Swift never so wildly mistook his talents as to at- 
tempt such a project, Browne (who judged King's De origine mail 
to be fundamentally misconceived) first established himself as a 
prospect for ecclesiastical elevation by the systematic rebuttal 
which he wrote of Toland's deism. 

Browne's book, A Letter in Answer to . . . Christianity Mot Mysteri- 
ous, was published in 1697, two years before he became provost. 
Narcissus Marsh (then Archbishop of Dublin) supposedly asked 
him to write it, and valued the finished work so highly that he 
recommended the author's promotion. Toland allegedly boast- 
ed, thereupon, that it was he who had made Browne a bishop. 
Browne crusaded against toasts to the dead; and Swift once 
warned a parson of the Cork diocese 'not to drink or pledge any 
health in his company, for you know his weak side in that mat- 
ter'. 2 Even while counting him as a friend, Swift did not spare his 
character : 

He is a capricious gentleman ; but you must flatter him monstrously 
upon his learning and his writings; that you have read his book 
against Toland a hundred times, and his sermons, if he has printed 
any, have been always your model, etc, 3 

Nevertheless, though Swift ridiculed his productions, Browne's 
achievement in metaphysics is genuine ; A. A. Luce calls him c a 
subtle thinker with decided views of his own'. 4 Sober volumes of 
clerical erudition continued to flow from him throughout his life, 
and he also won a shining reputation as a preacher. "The most 
speculative writer of his age', Swift was to label him *and as 
scholars tell me, excellent in his way, but I never read much of 
his works' 5 : yet in 1710 it would be he a not Swift, who became 
Bishop of Cork. 

From such a prospect of Swift's elders and contemporaries, one 
need not have had his genius to realize that no way to preferment 

1 Ball i. 286. * Ibid, m, 246. 8 Ibid., p. 1244. * Luce, p. 31. 
* Shcrburn v. 1 7. 



was secure. Yet there were honourable steps that aspirants nor- 
mally took. For a person comparatively unsponsored, as Swift 
was to be, a studious college record, marked by a scholarship, 
would be followed by holy orders, a fellowship, and industrious 
publications on aspects of divinity. Within the church, he would 
support the labours of those superiors to whose principles he felt 
he could adhere. He would also tie himself to the policies of those 
statesmen who seemed aligned with his superiors. If he could de- 
liver a sermon or write a pamphlet to advance the aims of his 
patrons, he would carefully do so. 

If one did not make an early choice of an ecclesiastical calling, 
or if the gradual route of modest industry disgusted him, he 
might risk the heavier gamble of a tutorship, chaplaincy, or 
secretary's post in a great house, with the hope that the head of 
the family would graciously reward him for services rendered, by 
fixing up a comfortable benefice after a reasonable term of years. 
Since the church was almost an administrative department of the 
government, not only were bishops more regularly chosen on 
political than spiritual grounds (especially in Ireland, where the 
lay peers who troubled to attend the House of Lords were some- 
times outnumbered by the more sedulous bench of bishops), but 
cathedral stalls and good livings generally were conferred for se- 
cular reasons. The number of chaplains was great, of bishoprics 
few; the supply of parsons manufactured at universities *seems 
consistently to have exceeded the demand'. 'If you expect pre- 
ferment', a prudent adviser told a harness-weary curate, e y u 
must bustle and try to peep after it, as most of the profession do in 
these days,* 1 

Now, consistently throughout his life, Swift was to execute 
some of the prescribed manoeuvres. By pursuing them quite sys- 
tematically, he could have risen, as Atterbury did, in the face of 
an alert opposition. But even in his college days, he seems to have 
split differences. Apparently, he exerted himself to keep the 
respect of St George Ashe and other distinguished men whom he 
admired; but he let his studies slide, even though scholarly oc- 

1 Dorothy Marshall, English People in the Eighteenth Century, 1956, p. 101. 



cupations always interested and entertained him, and though he 
reverenced men of learning. Likewise, he could violate the col- 
lege discipline even while his relatives, not his own parents, were 
maintaining him at T.C.D. If this impression is sound, he had 
already entered upon his policy of flirting with disaster while 
longing for success, of pretending to be indifferent to what he 
wanted most. 

Nevertheless, the people whom Swift met during this period 
were a kind of social capital, on which he drew interest for the 
rest of his life. They were a second set of acquaintances, to be 
added to those relations among whom he had passed his infancy 
and childhood. Decade after decade, their names were to recur 
in his correspondence. Through William Waring (B.A., 1685) 
Swift probably met the only woman to whom we know he pro- 
posed marriage. Captain John Pratt (B.A., 1689) became Swift's 
financial guide and invested large sums for him. Pratt's older 
brother, Benjamin (B.A., 1689), another of Ashe's pupils, became 
Swift's friend and collaborated with him in church politics. The 
university studies which Swift followed were mostly traditional 
and hardly peculiar to Trinity College; basically, the prescribed 
work did not differ much at Oxford or Cambridge. However, the 
individuals after whose lives he was to shape his first hopes and 
ideals belonged to Dublin in the ninth decade of the seventeenth 

Even so, for a serious, bright, ambitious young man of Swift's 
circumstances with a modest immediate family, little financial 
underpinning, and few powerful friends apart from those on 
whom he had himself made an unusual impression no career 
was more likely, in England or Ireland, than the ladder of prefer- 
ment of the Established Church. To gain any distinction, if he 
should select this career, he would have to assert himself either 
by extraordinary services to the distributors of good things, or 
else by unusual genius. 


Chapter Eight 


It was while Swift was at Trinity College that the most fruitful 
intellectual tradition of his time found an establishment in 
Dublin a natural philosophy built upon observation, ex- 
periment, and mathematics. Although Swift was not to give a 
deep and positive response to this development, it was to be his 
repeated theme; and his treatment of it is one focus of his work. 
Perhaps the first source of what will appear as his indifference or 
hostility to the modern tradition of natural science was his meet- 
ing it under the misleading though attractive auspices of the 
members of the Dublin Philosophical Society. 1 

To this group of learned men, like their equivalents in Oxford, 
London, or Amsterdam, divisions of knowledge which have since 
grown mutually exclusive seemed either indistinguishable or else 
mutually strengthening. Only by reading the problems of the 
nineteenth century into the events of the seventeenth century 
have historians been led to think otherwise. 2 During Swift's 

1 A careful history of the Dublin Philosophical Society is needed. In my notes be- 
low I use the following sources, giving the references in the square brackets: B.M. 
MS. Ad. 481113, of which I quote from only the first [Ad. 4811]; *Sir Thomas 
Molyneux*, four articles in the Dublin University Magazine, xvni (1841) \D.U. M.~\\ 
Correspondence of the Philosophical Society of Dublin with the Philosophical Society of Oxford, 
in R. T. Gunther, Early Science in Oxford, xn (1939), 128-208 [Gunther]. Further 
materials will be found in the following: T.G.D., Molyneux papers (MS. 1.4. 17, 
1 8); Southampton Givic Centre, Archives, Molyneux papers; Thomas Birch, 
History of the Royal Society, 1756-7. 

9 Ogg ii. 52330 is thoroughly misleading, though many scholars share his con- 
fusions. In England during the seventeenth century, there was no essential con- 
flict between religion and what we now call science, or between humanistic studies 
and experimental research, or between the universities and the new philosophy of 
nature; there was no essential tie between Puritanism and experimental science; 



youth, keen philologists were eager experimenters; pious divines 
practised mathematics. The c illuminati 5 of Dublin possessed what 
today would be a rare combination of aesthetic, scholarly, and 
scientific energies ; and in the service of the state they might be 
high commissioners, privy councillors, or, as bishops, members 
of the House of Lords : too many jobs were chasing too few people. 
These men saw no conflict between physics and faith, imagina- 
tive writing and scholarship, music and mathematics, or classical 
poetry and experimental science. Humanistic values, as well as 
the Established Church, they took for granted. To them, as- 
tronomy was a way to demonstrate morality, and Christianity 
was an inspiration for algebra. 'After great study', Narcissus 
Marsh wrote in his diary, *I had a good invention in conical 
sections, for which God's holy name be praised' (7 February 

Surveying the composition of the Society, we shall first observe 
what a variety of occupations were pursued by the comparatively 
few members of the governing or intellectual class of the king- 
dom, and secondly what an intimate alliance this particular body 
of them had with Swift's university. Surveying the operations of 
the Society, we shall discover a remarkable range of subjects fore- 
shadowing episodes or images in Swift's writing: astronomical 
researches which anticipate Laputa, excremental or animal ex- 
periments like those at the Academy of Lagado, a toy fleet like the 
one Gulliver would steal from Blefuscu, a study of spittle such as 
Swift was to deride in A Tale of a Tub, and even a flying island. 
These and other instruments of Swift's ridicule were already 
commonplaces though not yet of belles-lettres and they exis- 
ted in what to him would seem a framework of absurdity. That he 
need not have learned of them in Dublin, we must insist ; but that 
they were available to him many years before he embarked upon 
his great literary enterprises is certain. 

Although Sir William Petty and William Molyneux were the 

and seventeenth-century science had little influence upon technology: see A. R. 
Hall, Ballistics in the Seventeenth Century, Cambridge 1952; B-- L. Colic, Light and 
Enlightenment (Cambridge 1957), p. 4 and passim. 



most knowledgeable researchers of the group, it was centred in 
the dons of Trinity College and supported by clergymen con- 
nected with them. Besides Ashe, Marsh, and Huntingdon, the 
earliest members included Edward Smyth, a fellow of the Col- 
lege ; Samuel Foley, still a fellow but later to be Bishop of Connor 
and Down ; and William King, formerly a fellow but now rector 
of St Werburgh's and eventually to become Archbishop of Dub- 
lin. Behind the conception of the Society, however, the most im- 
portant agent was of course Molyneux. 

A Dubliner by birth (1656), he had been educated at Trinity 
College and in the Middle Temple, London. But having no taste 
for the law, he devoted himself to mathematics, architecture, 
engineering, optics, and other scientific or philosophical pur- 
suits. His translation of Descartes's Meditations was published in 
1680, when he also began a correspondence with the great as- 
tronomer Flamsteed. He became a friend of Locke's as well; in 
fact, both William Molyneux and his younger brother Thomas 
seem to have sought out or written to as many famous scientists 
as they could. In Dublin he promoted regular gatherings of men 
who shared his interests. 

During Swift's second year at Trinity College^the earliest steps 
were taken toward setting up a 'philosophical' society like the one 
at Oxford. Soon Huntingdon offered the hospitality of the pro- 
vost's residence to the group which Molyneux assembled; and 
the first formal meeting was held there in October 1683. Marsh, 
who was still Bishop of Ferns, read a paper; so did St George 
Ashe and Molyneux himself. At the end of the month Molyneux 
wrote to his brother (who was studying medicine in Leyden) that 
he had, in imitation of the Royal Society, 

promoted the rudiments of a society, for which I have drawn up 
rules, and called it Conmntw Philosophies About half a score or a 
dozen of us have met about twelve or fifteen times, and we have 
very regular discourses concerning philosophical, methodical, and 
mathematical matters. Our convention is regulated by one chief, 
who is chosen by the votes of the rest, and is called Arbiter Con- 
ventionis, at present Dr. [Charles] Willoughby (the name president 



being yet a little too great for us). . . Sir [William] Petty and all the 
virtuosi of this place favour it much; and have at some times given 
us their company. 1 

Petty was one of the most extraordinary Englishmen of the 
seventeenth century. Born in 1623, son of a London clothier or 
weaver, he claimed to have mastered, before he was sixteen, 
arithmetic, navigation, French, Latin, and Greek. After study- 
ing medicine in Holland and France, and being a fellow-student 
of Hobbes at Paris, he returned to England. In 1650 he was made 
a fellow of Brasenose College, Oxford. Petty became famous the 
following year by an achievement of which Swift must have 
heard: a hanged murderess was cut down too soon; and Petty, 
instead of dissecting her body as planned, brought her back to 
life with the help of another doctor. Shortly afterwards, he was 
made professor of anatomy in Oxford and vice-principal of 
Brasenose College. But in 1652, when he was appointed phy- 
sician-general to the Parliamentary army in Ireland, he entered 
upon a thirty-five-year period of alternating his residence be- 
tween England and Ireland. 

His friend, the great Robert Boyle, was already in that king- 
dom, on a visit (1652-3) to his Irish estates. Boyle described it as 
c a barbarous country, where chemical spirits were so misunder- 
stood and chemical instruments so unprocurable, that it was 
hard to have any Hermetic thoughts in it'. 2 Together, he and 
Petty carried on experiments in anatomy. However, the scheme 
which established both Petty's main reputation and his private 
fortune was a survey of the lands, forfeited by the Roman Catho- 
lics, which were to be divided between the Adventurers* and the 
Parliamentary soldiers* Not only did he carry this out by an in- 
genious method which was remarkably efficient ; but through the 
opportunities afforded him by the project, he managed to obtain 
a great deal of land for himself. After the Restoration, confirmed 
in the estates which he had acquired, Petty gave up much of his 
administrative work and extended his scientific pursuits. He was 

1 >. U.M., p. 472. * DNS. 



a charter member of the Royal Society, and received his knight- 
hood on the occasion of its establishment. When Molyneux was 
planning the Philosophical Society, Petty was in Dublin. He 
helped draft the original rules and constitution of the Society, 
and compiled for it a list of simple experiments which could be 
easily performed. For the year beginning November 1 684, he was 
elected president. 

Meanwhile, the Society had been active from its inception. 
Even the hard winter of 1683-4 did not impede events. Ashe and 
Marsh used the intense cold for experiments in freezing liquids; 
and though the surface of the LifFey was solid for more than six 
weeks, the philosophers met every Monday. Encouraged by the 
Oxford and Royal Societies, they soon decided to give them- 
selves a 'better form'. Marsh, Petty, Willoughby, and Molyneux 
'were pitched upon to draw up rules 5 , and quickly presented a 
set modelled upon those of the Royal Society. These were accept- 
ed of course, and at the end of January the first officers were 
elected Willoughby as president and Molyneux as secretary- 
treasurer. In April 1684 they began meeting in a room of their 
own, ( in [the] Crows-nest'. By May of the following year, Moly- 
neux could tell his brother that they had a fixed place in a house 
in Dame Street, 'where we have a fair garden for plants, and a 
laboratory erected for us'. 1 When Clarendon arrived as lord 
lieutenant (1685-6), they tried to interest him in their work; he 
encouraged them to seek a charter, and he came to a meeting. 2 

In the researches carried on by the members, no man showed 
more energy than Swift's tutor. Ashe's devotion to geometry and 
meteorology has already been noticed. He read papers on mathe- 
matical problems ; gave accounts of the barometric pressures, 
winds, rainfall, etc., at Trinity College; brought in curiously 
shaped stones and fossils (including one 'taken out of the bladder 
of a noble man's cook in the country who lately dyed of it' 8 ) ; and 
described a horse 'whose yard was fix'd about two inches below 
the anus', with e two teats under the belly*. 4 John Jones's ridicule 

*7itW.,f. i70 v . 



of Ashe as a virtuoso obsessed with eclipses had some justification. 
To calculate the eclipse of the sun in July 1684, he and Moly- 
neux drew up tables; but the observations were almost com- 
pletely obscured by poor weather. The following December, an 
eclipse of the moon was due, and again Ashe laid plans. Again the 
sky was overcast, and nothing could be seen. In November 1685 
the moon obliged him with another eclipse ; Ashe reported that 
a cloudy sky had made precise observations impossible. 1 

Molyneux discussed telescopes and dioptrics. He showed how 
a telescope could be used to examine a miniature portrait in de- 
tail: it "makes all the features appear most strong', he said, c and 
likewise hereby the faults of such a peice of painting are most 
apparent, which otherwise are apt to escape the unarmed eye 5 ; 
it represents 'the small face of a curious peice of miniature with 
all the advantage imaginable, and as big if not bigger then the 
life', 2 He dissected a newt and showed the circulation of its blood 
with a microscope. 3 He talked on the petrifying properties of 
Lough Neagh and on hydrostatics. 4 

Allen Mullen, M.D. of Trinity College, was famous for his 
Anatomical Account (1682) of an elephant which had been acci- 
dentally burned in Dublin. He performed experiments on dogs, 
usually filling them with odd substances. He pumped eighteen 
ounces of water into the thorax of one, and found that it grew 
short-winded. Into the jugular vein of another, he injected an in- 
fusion of opium mixed with brandy and water; the dog died. He 
cut off a piece of another's lungs, and tied up the surrounding 
tissue to prevent bleeding; the dog lived. 5 Mullen fed a die to a 
dog : 'he kept it in his body twenty-four hours, when it came out 
it had lost half its weight, but retain'd its cubical figure most 
accurately, and every point on each side.' 6 

Petty was good at giving advice. He proposed experiments on 
land carriages and on mineral waters, made a list of instruments 
needed by the Society, and suggested general rules of procedure, 7 


He liked to design models. For the road-haulage experiments, he 
displayed a land carriage' of Lilliputian dimensions, a solid 
parallelepiped of five inches thick, and ten inches long, weighing 
ninety-nine ounces: being so ordered that it may be put on 
wheels, either one sett or two setts, of equal or unequal diameters, 
or it may be made a sled, or to be drawn on four or two dragging 
wheels, or on the ful flat'. 1 Even before the Society had been 
formed, in May 1683, he had demonstrated a toy navy : 'He has 
a fleet of ships in little modells of about one and a half and two 
foot long, of all kinds, more than the King of France, and with 
these in a great broad trough of water he performs wonders.' 2 

Other topics came up often and from many sources: mag- 
netism, the calculation of the longitude, methods of softening 
wood or of petrifying it, the 'formation of letters and an universall 
character', 3 water pumps and other devices to raise water. Often 
the research is plainly misguided, such as an almost Laputan 
attempt to show that water does not move 'naturally' : The 
figure of the particles that do compose water are the most unfit 
for motion; for it has been long demonstrated that round bodys 
are the most fit; but the parts of water are generally agreed to be 
oblong.' Hence, argues the philosopher, 'its first and chief mo- 
tion proceeds from the elastick power of the air containd in it', 
and it moves fast or slow Trom the different pressure of the ex- 
ternall air upon it'. 4 Some of the research is excremental, such as 
a report on, the worms found in a dog's intestines, 5 or details of 
the spherical little "bladders' voided by a sick man instead of 
faeces: 'Some containd a hard substance of a consistence like a 
greene plum and some a harder: they were of very different 
colors and bignesses, some nigh as big as a pidgeons egg*' 6 

Many of the experiments were copied from those of the Oxford 
and Royal Societies; and what the members did not perform 
themselves, they might learn about through correspondence 
with these or with various philosophically disposed individuals. 
Their London correspondent could tell them, The spot in the 

* Ad. 4811, f. r6i v . D.l7.Af.,p. 482. Ad. 481 1, f. 175. 

* Ibid., f. 41 v . * Ibid.) ff. I77~7 V . Gtmther, p. 197. 



sun is disappeard. Monsr. Marietta at Paris is dead.' 1 Their Ox- 
ford correspondent could report, c Human spittle, clarified by 
standing, being mixt with syrup of violets, turn'd to a delicate 
green color.' 2 Petty might read them the account sent to him of a 
new satellite around Saturn. 3 Similarly, they learned at second 
hand about an 'odd meteor' (obviously a mirage) occurring in 
hot weather over Rhegium, Italy, which made cities appear 
among the clouds, and men walking. 4 

As the troubles closing James II's reign upset any sort of 
orderly business, the meetings of the Society grew irregular; and 
the group was finally dispersed. When it came to be re-formed, 
in the spring of 1 693, 5 Swift was living in England. 

Meanwhile, without estimating how much he heard of the 
Society's business, we can be sure that he had an acquaintance 
with most if not all of the members. In Jones's tripos alone, is 
sufficient evidence that the students were aware of their seniors* 
researches ; and both there and in Ashe's address to Clarendon, 
is evidence that outsiders used to ridicule these occupations. For 
personal and historical reasons, though Swift might respect or 
even admire the men themselves, he would be likely to treat their 
scientific pursuits as outmoded hobby-horses, no more worth 
while than the logic and metaphysics which the same men hand- 
led at the university. In humanistic studies, the College pro- 
gramme had a rational articulation; the purpose of each exer- 
cise seemed clear. But to a nature as order-loving as Swift's, the 
miscellaneous enthusiasms of half-baked experimenters would 
by contrast seem a jungle. 

Furthermore, in defending the philosophy of systematic ex- 
periment and observation, it had been common for zealous pro- 
pagandists (not usually the real scientists themselves) to appeal 
to the usefulness of the inventions which would come out of the 
research : there was a note of panacea in most of the recom- 
mendations. By the end of the seventeenth century this note was 
turning sour. To many younger men, the Royal Society appeared 

1 Ad. 481 1, f. 27 V . a Ibid., f. 48. * Ibid., f. i65 v . 
4 Ibid,, ff. I70 v -i. 5 Ibid., f. 182. 



to have belied Sprat's promises and to represent a decaying fad, 
as specious as alchemy or astrology. It was one thing to believe 
that scholasticism had been overthrown by Bacon, or to see Des- 
cartes's mechanical philosophy defeated (as it was soon to be) by 
Locke: these could be interpreted as victories of empiricists over 
system-builders, of simple facts over wild speculation. It was an- 
other thing to espouse yet another universal system just as its sun 
seemed to be put out. 

The question at issue is not whether there was a feeling in society 
that philosophers should study c things of use' instead of weighing 
the air and magnifying the flea, but whether the philosopher him- 
self yielded to such an opinion. The evidence seems to show that 
he did not ; he clung to his old name and status and resisted the 
attempt to make natural philosophy a sort of superior technology 
restricted to the problems of cider-making and navigation. Theo- 
retical studies held the field the scientist could be a Gimcrack if 
he chose so that by the end of the century the interest aroused by 
publicists like Samuel Hartlib had waned because the promised 
miracles had not occurred. 1 

Most of the discussions held by the Society were ill informed or 
amateurish when they were not absurd. Though a sympathetic 
historian can recognize the deep issues underlying the grotesque 
experiments, an undergraduate would feel less patient. He would 
not know that behind the collecting of petrifications, for example, 
lay problems like the origin of life, the difference between ani- 
mate and inanimate, and the validity of the biblical story of crea- 
tion. Or if he did, he would assume not that these grand inquiries 
were premature, and the investigators wrestling with tasks for 
which their instruments were not yet adequate, but that the 
mysteries were essentially unfathomable and the labour spent on 
them wasted. 

Did he also think this labour tended directly to undermine the 
Established Church ? Hardly since no one ever accused Ashe, 
King, or Marsh of impiety; and since the first became Bishop of 
Clogher, the second, Archbishop of Dublin, and the third, Pri- 

1 A. R. Hall, Ballistic f in the Seventeenth Century (Cambridge 1952), pp. 163-4. 



mate of all Ireland, Swift would hardly have associated their in- 
terests with Puritanism. If he ever considered, as a group, the 
savants who dominated the Society, his judgment must have been 
one of unusual esteem, however sadly he may have regretted 
their foibles. Indeed, it would be just because he thought he 
understood these men, or appreciated their virtues, that he could 
freely dismiss as folly the labours which absorbed them. 1 

Naturally, the forces which broke up the Society also shook 
the College; for by the time William of Orange landed in 
England, Tyrconnel had done his work in Ireland. He had 
swamped the corporations with Roman Catholics, and his 
threatening attitude toward the Act of Settlement was well 
known. Prudent families of Protestants began to dispose of their 
property and to leave the country. In some districts the Irish 
Catholics, believing their sun had risen at last, simply seized 
various belongings of the Protestants. 

The idea of 1641 was sharp in every Protestant imagination. 
Those who had no fear for themselves were anxious about their 
families, and the departures grew into an exodus. 'All combine 
to leave the state,/Who hate the tyrant, or who fear his hate.* 2 
Robert Huntingdon and Edward Jones fled to England in 1688. 
Narcissus Marsh was driven from his bishopric early in 1689, and 
after a brief stay in Dublin went to England. William Molyneux 
moved to Chester at the end of January 1689. 

By this time Trinity College was in an impossible situation. In 
the summer of 1687 Tyrconnel had tried to intimidate the pro- 
vost and senior fellows into appointing a Roman Catholic to a 
lectureship for which there was no proper endowment. They 
managed to defeat him; but early the following year the king 
ordered them to admit a Roman Catholic to a fellowship. They 

1 1 do not accept the theses argued by either Robert Merton, in 'Puritanism, 
Pietism, and Science', in his Social Theory and Social Structure (Glencoe, Illinois, 
1957), pp. 574-606; or R. F. Jones, in 'The Background of the Attack on Science*, 
in Pope and His Contemporaries, ed. J* L. Clifford and Louis Landa (Oxford 1949), 
pp. 96-113. See Colie, pp. 49-65, said passim. 

2 Dryden, Aeneid I. 498* 



showed evidence that the candidate had committed thefts and 
had fathered bastards. The king did not press them further, but 
Tyrconnel withdrew the annual grant of money which the Col- 
lege had been accustomed to receive from the Exchequer. Mean- 
while, one of TyrconnePs judges was trying to 'discover' a plot by 
some of the students to murder the lord deputy. His honour at 
last admitted that this was c a very ridiculous business' ; but when 
the provost and fellows tried to sell some of the college silver (a 
conventional method of raising money), Tyrconnel intervened 
and would not let them. 1 

By the beginning of i68g a none of the tenants of the college 
properties were paying rent, and supplies fell so low that c but one 
meal a-day' was ordered to be provided in hall, c and that a din- 
ner, because supper is the more expensive meal'. Most of the 
students must have left about this stage. Early in March ten fel- 
lows, including Ashe, embarked for England ; and another died. 
In September the four who remained were simply turned out, 
and a regiment of foot was quartered in the College. Though the 
refugee fellows only began to return after the Battle of the Boyne, 
a tiny skeleton crew of senior members stayed in Dublin all 
through the troubles. 2 

Swift and his cousin Thomas were probably gone by the end of 
January i68g. 3 Although they had arrived together, worked 
under the same tutor, and no doubt left together, they had begun 
already to move apart. The elder boy had won his scholarship 
after two years as an undergraduate, had behaved himself rather 
more decorously than his cousin, and now paused a while before 
the correct next step. This was to resume his studies for the M.A. 
at his father's college, Balliol, Oxford, which he entered in 
November 1690. (Dillon Ashe had gone up to Magdalen Hall in 
May.) But Jonathan took another route, starting from Leicester. 

1 Stubbs, pp. 119-27. * Ibid., pp. 127-33. 

9 Barrett, p. 13. In a letter of 1692, Swift said he had spent seven years at Trinity 
College, a figure which would set his departure in the spring of 1689, or probably 
earlier since his emphasis is on the length, rather than brevity, of the period 
(Ball i, 10); Temple, in May 1690, says Swift *was near seven years in the College 
of Dublin, and ready to take his degree of master of arts, when he was forced away 
by the desertion of that College upon the calamities of the country* (Ball i. 2). 


Part 1 1 

Chapter One 

The troubles then, breaking out', Swift wrote of himself, 
*he went to his mother, who lived in Leicester, and 
after continuing there some months, he was received 
by Sir William Temple, whose father had been a great friend 
to the family and who was now retired to his house called 
Moorpark near Farnham in Surry.* 1 Leicester, of course, he 
never meant to be more than a halting-place. Swift and his 
mother must immediately have considered what career he might 
follow. With his ancestry, his schooling, and his poverty, the 
most sensible direction was ecclesiastical, except that he had, in 
his own words, c a scruple of entring into the church meerly for 
support*. 2 

While the future remained unsettled, Swift did not limit his 
conversation to his mother; for his conduct with one young 
woman stirred up unnecessary anxieties. The girl was Elizabeth 
Jones, daughter of the Rev. John Jones, vicar of Wanlip, Lei- 
cestershire 3 ; and it was Mrs Swift herself who brought him c to a 
knowledge* of the family, for Miss Jones was her cousin. 4 The 
friendship reached a stage where his *prudent mother' began to 
feel alarmed that he might be in love with her. However, the end 
of Swift's visit was the end of the affair, unless the ^letters to 
Eliza* which he later asked a friend to burn, were written to her. 5 
A few years after he left, she married an innkeeper 6 ; and in an- 

1 Autob., f. 8. a Ibid., f. 9. 8 Near Belgrave. * Ball rv. 55-6. 
6 Ibid. I. 2930. 

8 Theophilus Perkins of Loughborough, married at Thurcaston 1692; Josiah 
Birkhead was the surety. 



other trip to Leicester, Jonathan was soon involved with another 
young woman. 

From his mother's house he went to Surrey as a sort of secre- 
tary, writing, reading aloud, and keeping accounts for the son 
of the Swifts' benefactor in Ireland. 'His whole family 5 , Temple 
wrote, "having been long known to me obliged me ... to take care 
of him.' 1 Of all Swift's conscious decisions, this was to be the most 
far-reaching; and of all Sir William Temple's decisions, none was 
to give him a larger share of immortality. It was too late for the 
host to be much altered in character by his protege, but Swift's 
nature underwent, from Temple's example, the deepest changes 
it could suffer after adolescence. From an appreciation of the 
older man's attitudes and values, we can come to understand 
what the younger man purposed and worked for. 

The profundity of Temple's effect on Swift has always been 
underestimated. During a period of twenty years Swift was to 
transcribe, edit, and see through the press almost the whole body 
of Temple's literary works, including three volumes of corres- 
pondence and two of memoirs. Having come straight from the 
university, he was to live for a decade (with two long breaks) in 
the Temple family. From that household he was to take the 
woman who as a child of eight met him on his arrival, as a girl of 
eighteen saw him finally depart, and as a woman of twenty fol- 
lowed him to his native city, there to remain his most intimate 
companion until she died. Swift obtained a most minute know- 
ledge of Temple's career, drew parallels between it and his own, 
and sometimes acted on these parallels. He felt directly the 
effects of Temple's experiences as a son and a father. He admired 
Temple's character and his mind. Temple's literary style, politi- 
cal philosophy, moral outlook, and aesthetic judgment became 
either models or points of departure for Swift's own. 

It was not alone the man Temple had become that formed 
Swift: it was also the man he had been. If Swift had not pos- 
sessed a comprehensive familiarity with this history and back- 
ground, he could neither have bathed Temple's memory in the 

1 Ball i. 2, 



superlatives which he devoted to it, nor have identified himself, 
again and again, with the patterns of his master's life. The migra- 
tions of various Temples between England and Ireland ; the 
domestic arrangement of a single Sir William usually attended 
by two ladies (his sister and his wife) but often cut off from his be- 
loved Dorothy; Temple's repeated efforts to reform national 
policy, which were repeatedly and treacherously frustrated; his 
utter lack of material reward for a career of distinguished public 
service these were some of the points at which Swift was either 
to find or to make his own life run parallel with Temple's. 


Temple was the same age as Swift's uncle Godwin, He had been 
half-orphaned at eleven, when his mother died; and much of his 
early education was supervised by her brother, Dr Henry Ham- 
mond, author of the introduction to The Whole Duty of Man. 
Afterwards, as dean of Christ Church, Oxford, Hammond had a 
reputation for 'shaming the vicious to sobriety, encouraging the 
ingenuous to diligence, and finding stratagems to ensnare the 
idle to a love of study'. 1 When the saintly man died, Temple 
showed the power of his attachment by falling 'quite sick 5 at the 
news. 2 Hammond was the main instilier of those virtues which 
were to make Temple appear to Swift as a Christian hero. 

The college selected for the boy by his father was not, how- 
ever, royalist and Anglican, like his uncle's, but markedly Puri- 
tan: Emmanuel, Cambridge. 3 Here, though his tutor was the 
great Platonist and liberal thinker, Ralph Cudworth, the effect 
of the two opposing Christianities was to give Temple a cool dis- 
position toward all churches. Even at twenty-four he could write, 
Taith must be purely an inspiration of heaven or an operation of 
custom, not a work either offeree or reason* 4 ; neither Hammond 
nor Cudworth would have smiled at this maxim. Like Swift, 

1 John Fell, L\fe offfammond, 1661, p. 49. 

fl Lady GifFard, in Temple's Early Essays, p. 9. 

3 Lemuel Gulliver's college. * Woodbridge, p. 27. 



Temple found the exercises of scholasticism unappetizing; and 
although Cudworth pressed him to study logic and philosophy, 
his 'humour 5 was too 'lively' to pursue them. 

Entertainments which agreed better with that [humour] and his 
age, especially tennis, passed most of his time there [i.e., at the 
university], so that he used to say, if it had been possible to forget 
all he had learned there, he must certainly have done it. 1 

The fact that he was remarkably handsome probably contribut- 
ed to his delinquency. 

After Cambridge there were the conventional (but unusually 
extensive) travels on the Continent, and the very unconventional 
six-year courtship of Dorothy Osborne happily concluded 
by their marriage. The tenacious opposition of Temple's father 
to his choice of wife did not weaken the intimacy of parent and 
child. Temple set great store by filial obedience, and at later cri- 
tical points in his career, he submitted to his father's judgment. 2 

Five months after the birth (1655) of their first child, John, the 
Temples established themselves in Ireland, where they lived 
quietly until the eve of the Restoration. But Temple was ob- 
viously waiting to be called into the service of the state. Mean- 
while, though they divided their time pleasantly enough between 
a place built for them in the country and old Sir John's house in 
Dublin, they were made regularly miserable by the successive 
deaths of their numerous children ; for during this period only 
John survived. 


With Temple's election to the Irish convention of 1660, his pub- 
lic life got under way; and he was a member of the parliament at 
Dublin the next year. Working on important committees, he 
journeyed between England and Ireland, and met the king for 
the first time. His moderateness and his powers of conciliation 

1 Lady Giffard, quoted by Woodbridge, pp. 10-1 r. 

* E.g., in his refusal to be ambassador to Spain and his refusal to buy the office of 
secretary of state, 



were so successful that later, when he sat in the English parlia- 
ment, another member said, 'he was glad [Temple] was not a 
woman, because he was sure he might have perswaded him to 
any thing 9 . 1 This tribute is reported by Temple's sister, Martha, 
whose distinction it was to have been widowed almost immedi- 
ately after her marriage to Sir Thomas GifFard (i662). 2 Lady 
Giffard stuck closer to her brother than even Lady Temple, 
whose authority in the family seems at times to have been weaker 
than that of her sister-in-law. In 1665, the whole family Sir 
William, Lady Temple, Lady Giffard, and John Temple 
moved to England, where they went to live in Sheen, Surrey 
(near London), while Temple pursued his ambitions among the 
court circle. 

At last, in the middle of the year, the court sent him to Mun- 
ster ; his charge was to keep the prince-bishop to a promise of in- 
vading the United Provinces, with whom England was at war. 
Although the larger purposes of this project were shattered. 
Temple saved the pieces so well that in 1666 he was made a 
baronet. In his next, and most spectacular, feat, Temple revealed 
a candour, and a faith in the efficacy of candour, rarely found 
among the huddled notions of diplomacy. (To Swift, this was to 
seem one of Temple's most admirable traits, and one of those 
most worthy of emulation.) From observations during his travels, 
from recent events on the Continent, and from talks with John 
De Witt, Grand Pensionary of Holland, he had formulated the 
central principle of his diplomacy: that the English must join 
with the Dutch to prevent Louis XIV from overwhelming Eu- 
rope. Temple's optimism about the possibility of consummating 
this wish is as characteristic as his misguided confidence in 
Arlington the secretary of state and Charles II, both of whom 
were busy undoing his web quite as fast as he pulled the threads 
into place. Nevertheless, the Triple Alliance (of England, 
Sweden, and the United Provinces) was arranged, and arranged 

1 Temple, Early Essays t p. 9. 

2 In a letter to Ormonde 13 May 1662 Sir George Wentworth says his nephew, 
Sir Thomas *Giflbrd J , died when he had "not beene aboue a weeke a married man* 
(Bodl. MS. Carte 31, f. 517). 



so quickly that Temple received amazed congratulations on hav- 
ing c in five days' concluded a treaty which the French ambas- 
sador had thought would be still under discussion after six weeks. 1 
Swift of dispatch and easy of access. Temple disposed of all the 
ramifications by June 1668. He now innocently trusted that 'the 
general interests of Christendom [were] secured against the 
power and attempts of France'. 2 Macaulay, echoing Pepys, calls 
the Triple Alliance 'the single eminently good act performed by 
the government during the interval between the Restoration and 
the Revolution'. 3 Even Burnet admitted it was c the masterpiece 
of king Charles's life' and said that if the king had stuck to the 
treaty, it would have been c both the strength and the glory of his 
reign'. 4 Thanks, however, to the tortuous manoeuvres of Arling- 
ton and the king, the Alliance though popular enough in both 
Holland and England achieved practically nothing. 

Nevertheless, when Temple arrived home, he seemed to have 
killed the minotaur, and might have claimed a peerage or a pen- 
sion. That he chose to do neither, and that he hesitated several 
weeks before accepting a new post as ambassador to Holland are 
signs of another trait which impressed Swift : Temple's deliberate 
indifference to visible rewards. Temple apparently was guileless 
enough to believe that by requesting no preferment he would 
deepen Charles's reliance upon him. He also credited himself 
with sufficient magnanimity to enjoy the pure kudos of patriotic 
service for its own sake without regretting the lack of negotiable 
benefits. But he was more transparent and less saintly than he 
imagined. In politics as in crime nothing binds men closer than 
reciprocal profits* Charles sold his powers to whoever would give 
the services he required, and Temple, by not pressing his own ad- 
vancement, lost both the ability to enforce his policies and the one 
consolation which might have softened his failure; for he found 
himself ruefully out of pocket through the demands of his public 

1 Woodbridge, p. 85. a Ibid., p. 91. 

8 *Sir William Temple*, in Constitutional Essays, World's Classics edition, p. 346. 

4 Burnet i. 440. 




During his first service in Holland, 1668-71, Temple learned to 
allay a bit 'that frankness of my heart which made me think 
everybody meant as well as I did*. 1 Although he conceived of 
his function as to shore up c the Triple Bond 9 , he was actually a 
blind of integrity behind which a new war with the Dutch was 
projected. While Charles and Arlington secretly arranged the 
Treaty of Dover (by which England joined with France against 
the Dutch and against Protestantism), Temple was assuring the 
anxious Pensionary that no crown would 'enter into counsels so 
destructive to their honour and safety as those he suspected'. 2 In 
September 1670 he was recalled, ostensibly for brief consultation, 
but really to keep him out of the way without alarming the 
Dutch. Not until the following August, after almost a year's 
loitering in a chilly court, was he released from his mission and 
permitted to rejoin Lady Temple, John, and Diana a daughter 
born in 1665 all of whom the king had forced him to leave in 

Even now he had not learned, Lady GifFard writes, 'the les- 
sen . . . one should alwayes be perfect in, before one comes to court, 
to swallow every thing'. 3 For a few years he took care of his family 
and improved the estate at Sheen, without, however, ceasing to 
watch closely the developments in foreign and domestic policy. 
He refused to accept the high offices which were offered to him. 
until he could take up the negotiations to end the very unpopular 

After concluding a separate peace with the Dutch representa- 
tives in England, he went to Holland as 'mediator* between 
France and her opponents, the United Provinces, Spain, and the 
Empire. Since Charles II was secretly co-operating with the 
French, Temple had to endure four years' worth of mock bar- 
gaining which took him back and forth, from England to Holland 
to Flanders; and although nobody got out of the miry slough less 
bedaubed than he., the treaty finally signed at Nimeguen (Feb- 

1 Woodbridge, p. 106. a Ibid. 9 p. 107. 8 Temple, Early Essays, p. 16. 



ruary 1679) was a joke, which only postponed the judgment to 

In explaining why Charles did at last turn against Louis XIV 
and allow peace to be made. Temple nicely illustrates another of 
his elementary principles which Swift was to adopt. This is the 
assumption that the ultimate spring of national policy is the 
character of the executive or ruler who enunciates it : 

Though the wise reflections of the best historians, as well as the 
common reasonings of private men, are apt to ascribe the actions 
and councils of princes to interests or reasons of state, yet whoever 
can trace them to their true spring, will often be forced to derive 
them from the same passions and personal dispositions which 
govern the affairs of private lives. 1 

Not that Swift had to learn the principle from Temple ; it was 
a commonplace, and it fitted in with Swift's intuitive conception 
of society. In the reign of Louis XI V, moreover, there seemed to 
be irresistible proof of it, if anyone should raise a doubt. But 
Temple's authority, and the many examples he supplied, fixed 
the doctrine so that even when the evidence led a different way, 
Swift would account for political decisions by the character of 
the man making them. In the following anecdote, retailed by 
Swift himself as a precious revelation from Temple, no mention 
is made of the widespread popular discontent with the court's 
pro-French policy, but all the weight is laid upon the private 
feelings of the king. Yet the 'article' which Temple and Swift 
treat as crucial has no basis in fact. 2 

The secret of the king and the dukes [i.e., the Duke of York's] being 
so eager and hearty in their resolution to break with France at this 
juncture, was as follows. 

France in order to break the force of the confederacy, and elude 
all just conditions of a general peace, resolved by any means to enter 
into separate measures with Holland ; to which end it was absolutely 
necessary to engage the good offices of the King of England, who 
was look'd upon to be master of the peace whenever he pleas'd. The 

1 Introduction to the History of England, p. 284. * Cf. Ogg i. 556-8. 



bargain was struck for either 3 or 400 thousand pounds. But when 
all was agreed. Monsieur Barillon the French ambassador, told the 
king; That he had orders from his master, before payment, to add a 
private article; by which his majesty should be engaged, never to 
keep above eight thousand men of standing troops in his three king- 
doms. This unexpected proposal put the king in a rage, and made 
him say, 'd's fish, does my brother of France think to serve me 

thus? Are all his promises to make me absolute master of my 

come to this ? Or, does he think that a thing to be done with eight 
thousand men. 

9 Tis possible I may be a little mistaken as to the sums of money, 
and number of men; but the main of the story is exactly as I had it 
from the author. 1 

While the close of the histrionic enterprise at Nimeguen hardly 
purchased Temple everlasting honour and renown, he did get to 
know William of Orange so intimately that the young prince 
often dined with the Temples and came to place complete con- 
fidence in Sir William. In April 1676 the prince asked his opin- 
ion about a marriage between himself and Princess Mary, 
daughter of the Duke of York a match which Temple wished to 
encourage. After hearing the hopeful answer, the prince decided 
to 'enter upon this pursuit 3 . 2 The wedding, which took place a 
year and a half later, tilted the balance of European power : with- 
out it, the Glorious Revolution and the War of the Spanish 
Succession might not have occurred. 

The reversals and irresolutions of Charles II began now to tell 
upon Temple, giving him as his sister says e a distast to the 
thoughts off all publick imployments'. 3 He returned to England 
in March 1679 and never went abroad again. For almost two 
years more he did carry on a political career of sorts. Though un- 
willing to reach for the office of secretary of state (which several 
times was dangled before him), he directed the unsuccessful re- 

1 Swift's note to pp. 355-6 of Temple's Letters in. 

2 Temple, Memoirs n. 155. a Temple, Early Essays, p. sto. 



organization of the privy council, represented Cambridge Uni- 
versity in parliament, opposed the Exclusion Bill, and dodged the 
Popish Plot. This was, however, no time for moderation; and his 
attempts to keep peace between the king and the commons were 
as futile as the pilgrims' words at Vanity Fair. Temple withdrew 
to Sheen in January 1681, and (through his son John) promised 
Charles that he 'would never meddle any more with any publick 
affairs'. 1 

The life of action which he renounced for himself, Temple felt 
eager to open to his son. When John was only nineteen, Sir 
William wrote from The Hague to Lord Arlington, 

I hope [after this mission], to go and sleep at home, and leave my 
son in the busy world, which requires men spirited with some other 
heats, than I have about me. If upon his coming over you can find 
any thing you would have said to me, though he be young, yet I am 
pretty confident he may be trusted with it; for he has a plain steddy 
head, and is desirous to do well. 2 

When John did deliver a letter from Danby, the lord treasurer, 
Temple told Danby he hoped the boy would 'live to deserve some 
place in your service'. 3 Temple seemed as well to wish to involve 
his son in his literary occupations ; for when he dedicated a vol- 
ume of memoirs to John, he wrote, 'Whatever I leave of this or 
any other kind, will be in your disposal.' 4 In general, he seems to 
have felt the same close identification with John's destiny that 
his own father had felt with Temple's. That attitude was intensi- 
fied by the fact that no other child of his was living. The charming 
Diana had died of smallpox in 1679 'a child he was infinitely 
fond off; & none ever deserv'd it more from a father'. 6 Temple 
had not taken the event easily. e My heart is so broken', he wrote, 
'that I have done nothing since as I should do, and I fear never 
shall again.' 6 

But not even John was to outlive his father. He had been mar- 
ried in 1685, and a year later Sir William gave him the Sheen 

1 Temple, Memoirs m. 139. 3 Temple, Letters ra. 39. s Ibid., p. 62. 

* Memoirs U 9 sig. A4 V . 5 Lady Giffard, in Temple, Early Essays, p. ax. 

6 Woodbridge, p. 207. 



property. The parents moved with Lady Giffard farther into the 
country, near Farnham., Surrey, where they took over a small 
estate which Temple named Moor Park, after the place in Hert- 
fordshire where he had spent his honeymoon. As this neighbour- 
hood threatened to become a battleground for the Jacobite and 
Williamite troops, the elder Temples moved back to Sheen, and 
there both families remained until the end of 1689. When Wil- 
liam of Orange came over in 1688, Temple exercising the sort 
of prerogative which his own father had been used to would not 
let his son go to meet the prince at the landing. The day after the 
coronation, however, the king made John Temple secretary at 
war (12 April 1689). Advising William on the Irish situation, the 
new minister apparently persuaded him to set free Richard 
Hamilton (a brigadier-general in the Irish army, imprisoned in 
the Tower) so he might go to Ireland and argue Tyrconnel into 
giving up. Instead of doing what he had promised, Hamilton 
joined the rebels himself. Out of shame for his costly mistake, 
John Temple drowned himself. Within a week of his appoint- 
ment, he took a boat on the Thames, and leaped out near Lon- 
don Bridge. 1 

His son's death at the age of thirty- two stunned Temple and 
added a final darkness to his retirement. e [It] brought a cloud 
upon ye remainder of his life & a damp upon ye good humor so 
natural to him & so often observ'd yt nothing could ever re- 
cover.' 2 

When Swift arrived as a fatherless refugee, it was this childless 
great gentleman that he met, one who, after the deepest experi- 
ence of court life and diplomacy, had retired to the compensa- 
tions of a disillusioned domesticity, 

1 See Ogg n. 247; AfZJV., oai (Mar, 1947), 145-54- 

2 Lady Giffard, in Temple, Early Essays, p. xii. 


Chapter Two 

f "^he troubles which dispossessed the Temples were of course 
I the same that drove Swift to England. He was in the house- 
JL hold at Sheen or Moor Park at least by the close of the year, 
and probably before the summer. 1 Swift came to his new home as 
to delectable mountains 3 although the excitement of the young 
student hardly shines through the words of the old dean record- 
ing only that he 'continued for about two years* there. Temple 
had not lost touch with kings and states. He was host and guest to 
William III and to the Duke and Duchess of Somerset. The clos- 
est of his friends, Henry Sidney, was still busied with the c airy 
visions' which he himself had left behind. To Swift's public 
career, however, both Sidney and the Duchess of Somerset were 
to make extraordinary, if negative, contributions. 

All the clan of Sidneys were familiar to Temple from Penshurst 
and Dr Hammond, As a boy he had adored Henry's beautiful 
elder sister, Dorothy the fair Sacharissa of Waller's lyrics. Her 
son, the second Earl of Sunderland, was (unfortunately) a trusted 
old acquaintance until Temple achieved enough sagacity to 

1 In the autobiographical fragment Swift says first that he \vas received by 
Temple 'who was now retired to his house called Moorpark near Farnham* (f. 8) ; 
but farther on, that Temple, getting tired of Sheen, 'bought an estate near Farn- 
ham . . where Mr Swift accompanied him* (f. 9). In a letter of 1727 he speaks of 
having eaten too many apples at Richmond (i.e., Sheen) when he was about 
twenty, and then mentions Moor Park as belonging to a later era (Ball zu. 413 14). 
It is possible to explain away the implicit references to Sheen as the place where he 
originally joined the Temples, but easier I think to assume that he did begin 
to live with them there, toward the middle of 1689, after William III was secure 
on his throne. Of course, Swift's memory was remarkably fallible, especially as to 
dates: the Pindaric ode to Temple which Swift headed 'June 1689* was certainly 
written later, Woodbridge (p. 219, n.) and Pons (p. 143, n.) notwithstanding; but 
the date he gave it may still be significant. Lyon, supporting the notion that Swift 
lived with Temple at Sheen, writes, < The Dean often said so himself* (p. 18). 



realize that the earl then secretary of state was feeding him 
lies and using him as a tool. With Algernon, a brother of Henry, 
Temple had a friendly correspondence which came to an end 
when the republican views of the exile made letters to him 
dangerous. At Sheen another brother, Robert, later Earl of 
Leicester, was a much-liked neighbour. 

Henry Sidney had tremendously good looks and winning 
manners to recommend him, also a shrewdness for pushing his 
own chance. As a politician he could not, said Swift, *turn a 
wheel for a mouse' ; but he had the invaluable foresight to realize, 
almost before everybody else, how probable it was that William 
of Orange would succeed James II. While living in Holland, Sid- 
ney made so much a friend of the prince that the scheme of in- 
viting William to England in 1688 might be said to owe its suc- 
cess more to him than to any other person. Apart from a trium- 
phant career of philandering, this was his most illustrious 
achievement. Sidney's term as secretary of state was an avowed 
receivership ; his brief lord lieutenancy of Ireland was remark- 
able for the chaos it engendered; and his title of Earl of Romney 
(1694) was less a token of the king's esteem than of his affec- 

If proof were lacking of Temple's inability to judge character, 
his respect for Sidney would be enough. Nevertheless, at one time 
or another during Temple's eighteen years of retirement, Sidney 
not only took the titles already mentioned but served as a mem- 
ber of parliament and as a privy councillor, held high military 
offices, and acted as a lord justice both of Ireland and of the 
realm. He kept up influential connections in France, Holland, 
and Ireland ; he was with King William at the Boyne. Into Moor 
Park's quiet garden-state he brought constant reminders of 
the renowned metropolis, with glistering spires and pinnacles 
adorned. The young Swift could not foresee that a crisis in his 
own development would some day evince the earl's inadequa- 

Other members of the household were to count for much more 
in Swift's life than either Temple's relatives or his guests. Re- 



becca Dingley, a spinster of the Hammond connection, stayed on 
as a waiting woman to Lady GifFard. 1 Temple's steward was 
Ralph Mose. Bridget Johnson, a widow, held the post of house- 
keeper. 2 She had been married to Edward Johnson, who had 
been Temple's steward in his time; and to him she had borne 
three children. Esther, 3 the eldest of these, was a sickly girl of 
eight when Swift arrived. This first-born, frail, and now father- 
less child, who was to inspire Swift with his deepest love, learned 
penmanship from him, and her writing as an adult was some- 
times taken for his. In addition, he recommended books for her 
to read, and 'perpetually instructed] her in the principles of 
honour and virtue'. 4 He found himself, no doubt, more comfort- 
able with Dingley, Mrs Johnson, and little Hetty than with the 
two old ladies and the baronet. 

At the moment, however, Swift's chief of men was Temple, 
whose way, though it seemed glorious, had apparently been 
ploughed by patience, temperance, and deeds of peace. Lady 
Giflfard's picture of him in 1 690 brings out the ageing statesman's 
affability and love of intimate society. His mood, she said, was 
naturally gay, but occasionally soured by fits of gloom. c His con- 
versation was easy and familiar with all people, from the greatest 
princes to the meanest servant.' His keenest pleasures were do- 
mestic: talking with friends and enjoying the house and gar- 
den. 5 An intelligent Swiss visitor, seeing Temple about four 
years later, was struck by his health and serenity: e He is free 
from business, and to all appearances free from ambition . . . 
and though he is gouty and getting on in years, he tired me in 
walking/ 6 

Lady Temple was no passionate, moody girl with an episto- 
lary flair. Smallpox had long since spoiled her beauty; nine 
children born and buried had darkened her temperament. She 

1 Her grandmother was a sister of Temple's mother. 

* The first wife of Ralph Mose was still alive; there is a bill for her clothes among 
Mr James Osborn's collection of Temple family papers (no. 21). According to the 
Farnham parish register, Bridget Johnson did not marry Mose until 1711. 

* Though christened Hester, she used the other form. * T. Scott xi. 127. 

* Temple, Early Essays, pp. 28-9. 6 Woodbridge, p. 232. 



had learned in her own words upon her son's suicide Svhat 
this world is' : 

It seems it was necessary that I should have a near example of the 
uncertainty of all human blessings, that so having no tie to the 
world I may the better prepare myself to leave it; and that this cor- 
rection may suffice to teach me my duty must be [my] prayer. 1 

For Lady Giffard, on the other hand, the family woes were 
vicarious; and she far outlived both her sister-in-law and her 
brother. Widowed at twenty-four, within a fortnight of her mar- 
riage, but ten years younger than Temple, she relished the high 
and secure station made for her at his side. During his travels she 
was with him more often than his wife; she copied her studies 
from his and became something of a bluestocking. While gracious 
and cordial to Lady Temple or others of her own rank, she does 
not seem to have unbent much toward inferiors. Swift shows no 
sign of having felt or returned any great fondness on her part. She 
was no doubt happier endearing herself to her magnificent young 
friend at Petworth, the Duchess of Somerset. 

Not that the intimacy between Moor Park and Petworth was 
unnatural. The duchess, Elizabeth Percy, daughter of the fifth 
Earl of Northumberland, was a cousin of the Sidneys. Temple 
had corresponded with her father and had admired her grand- 
father. Before she was fifteen, her avaricious guardian (the dow- 
ager countess) had pushed her into two disgraceful marriages. 
But she was released from the one by widowhood and from the 
other by a combination of flight and murder; for a jealous rival 
of her second husband had him killed after she had already run 
away from home. According to a strong tradition, her flight, 
which ended in Holland, was assisted by the Temples. In the 
spring of 1 682, about four months after the murder, she married 
Charles Seymour, the shallow, arrogant sixth Duke of Somerset* 
Red-headed, stubborn, and shrewd, she still continued to be 
notorious, less however for private than for public intrigue. 
Whatever Swift's first impressions of the duchess may have been, 

1 Woodbridge, p. 218. 


they did not teach him to avoid the grossest slanders when he 
came to attack her character in print two decades after his 
earliest opportunities to observe her. 

Swift mingled eventually with even the most exalted visitors; 
but his duties that first year were defined rather modestly by 
Temple : 'He has lived in my house, read to me, writ for me, and 
kept all accounts as far as my small occasions required.' 1 The 
work seems light enough, and the milieu afforded advantages 
that were hardly conceivable in Dublin or Leicester. Unfortu- 
nately, however, Swift now suffered for the first time the illness 
which never left him and which grew irregularly in severity: 
Meniere's disease, a disturbance of the inner ear, causing vertigo, 
deafness, or both. Swift mistakenly distinguished between the 
effects of the disorder upon his hearing and upon his equilibrium, 
not understanding that the various symptoms were all due to the 
same evil. His giddiness and 'coldness of stomach' he traced to 
an over-indulgence in apples, and therefore worried all his life 
about eating fruit, which he loved; the deafness, he traced to a 

I got my giddiness, by eating a hundred golden pippins at a time at 
Richmond [i.e., Sheen] . , . four years and a quarter [later], having 
made a fine seat about twenty miles farther in Surrey [i.e., at Moor 
Park], where I used to read and sleep, there I got my deafness. 2 

When Swift consulted physicians, they suggested the climate 
might be to blame, and 'weakly imagined that his native air 
might be of some use to recover his health' 3 ; so he returned to 
Ireland. Temple wrote him a letter of recommendation to Sir 
Robert Southwell, who was going with the king on the Irish ex- 
pedition.; Southwell was to be principal secretary of state for 
Ireland, and would have abundant patronage to dispense : 

1 Ball i. 2. 2 Ibid. m. 413-14. 

3 Autob., ff, 8-8 v , where he gives a slightly variant account of the disease : 'For 
he happened before twenty years old, by a surfeit of fruit to contract a giddyness 
and coldness of stomach, that almost brought him to his grave; and this disorder 
pursued him with intermissions of two or [three] years to the end of his life.* 


He has Latin and Greek, some French, writes a very good and cur- 
rent hand, is very honest and diligent, and has good friends, though 
they have for the present lost their fortunes. 1 

Temple suggested that Southwell find a place for Swift either in 
his own service or as a fellow of Trinity College, Dublin. 

If Swift travelled by his usual route, he stopped off at Leicester 
on the way and saw his mother. But whether his patron's recom- 
mendation did him any good or not has never been discovered, 
except that he certainly received no fellowship. This was an un- 
happy moment for such an aspiration, because the College was 
still trying to repair the injuries of the Revolution, and no new 
fellows were appointed until the spring of 1692. As for the 
Hibernian air, his health actually grew worse in it; and at 
Temple's invitation he decided to come back. Returning to 
England about August 169 1, he again visited his mother, staying 
with her through the autumn. 2 

A new amorous adventure gave rise now to fresh scandal; we 
have an account of the episode in the answer which Swift de- 
livered to an inquiry from the Rev. John Kendall, who had mar- 
ried Swift's cousin, Jane Errick. Though Swift told Kendall, 'I 
could remember twenty women in my life, to whom I behaved 
myself just the same way', this case must have had a special glow 
about it. Again the family feared marriage. Jonathan's excellent 
prospects with Sir William Temple should be cultivated by pru- 
dence, not spoiled through impetuosity. No, answered Swift, if 
anything was going to block his advancement, it was only his 
'own cold temper, and unconfined humor*. Even at twenty-four 
he was alert to the perils of carnality: e a thousand household 
thoughts . . . always drive matrimony out of my mind whenever 
it chances to come there', he said; 'besides that I am naturally 
temperate, and never engaged in the contrary, which usually 
produces those effects. 3 Through the dubious syntax the truth of 

1 Ball r. 2. 

8 On 14 Feb. 1692 he says lie returned from Ireland 'about half a year ago* 



his character emerges : Swift was so sure of his own caution, 
so untempted by misconduct, that he could sneer at gossip- 
mongers : 

I should not have behaved myself after the manner I did in Leices- 
ter, if I had not valued my own entertainment beyond the obloquy 
of a parcel of very wretched fools, which I solemnly pronounce the 
inhabitants of Leicester to be. 1 

And so, quite unmarried, he went from Leicester to Oxford, 
where he visited his cousin. Dillon Ashe had taken his Oxford 
M.A., ad eundem, in December 1690. Thomas Swift was still in 
Balliol College, having been incorporated in November 1690. 
Jonathan saw him toward the middle of December and reached 
Moor Park about Christmas i6gi. 2 With Thomas he must have 
discussed his own intention of entering Oxford University ; for 
on June 14 following, he was incorporated from Hart Hall. Both 
cousins now petitioned to be admitted as candidates for the M. A. 
degree with a dispensation from one of the exercises, all others 
having been performed. 3 The favour was granted; and on July 5 
(two days before Thomas), Jonathan received the degree. 

1 Ball i. 4-6. 

3 On 1 1 Feb. 1692 he says he has been back at Moor Park 'seven weeks* (ibid., 

P- 4). 

8 Nichols n. 108. 


Chapter Three 

Esrary aspirations were beginning to haunt Swift's breast. 
Soon after returning to Moor Park, he said, e ln these 
seven weeks I have been here, I have writ, and burnt and 
writ again, upon almost all manner of subjects, more perhaps 
than any man in England.' 1 The renewed contact with Temple 
must have strengthened these ambitions, since the diplomatist 
turned in retirement to the writing of essays and verses, the pre- 
paration of memoirs, and the editing of his own letters for publi- 
cation. Copying out, during more than a decade, his patron's 
various works, set Swift a model of the highest value. Yet his own 
earliest known compositions are not prose but six poems, pro- 
duced over four years (1690 to 1694) ; and their manner is not 
Temple's. The first four are c pindaric j odes like those of Abraham 
Cowley; and the remaining two are in the newer fashion of 
Waller's or Dryden's heroic couplets. 

If Swift's later and enduring style is to be considered 'char- 
acteristic' of him as a poet, these pieces suggest that he began by 
working against the grain. Eulogy was always to be his talent, but 
it would be best expressed ironically, through mock-insults, as 
when he would sneer at Lord Garteret for e adher[ing] so obstin- 
ately to his old unfashionable academick education', 2 In the early 
pindarics and couplets, however, he uses straightforward pane- 
gyric. Since his themes and values are blamelessly conventional, 
he is, in his search for freshness of effect, flung upon ingenious 
hyperbole; and since his language is too weak for the extrava- 
gance of his feelings, the outcome is bathos. 
1 Ball i. 4. a Davis XEC. 160. 



We may assume that he was sent in this false direction by 
Temple. When Swift as a young man said that he found a 'like- 
ness of humours' 1 between himself and his patron, he was less 
stating a fact than voicing a hope. If Temple derided Halifax 
for possessing a 'humour ; which he own'd always must have busi- 
ness to employ it, or would else be uneasy', 2 Ke could hardly have 
admired what Swift found in a famous piece of self-analysis : that 
(according to a e person of honour in Ireland 5 ) his own mind was 
like c a conjured spirit, that would do mischief if I would not give 
it employment'. 3 

Certainly Temple's literary preferences were opposed to those 
of Swift's maturity. Swift never felt Temple's distrust of the 
comic and satirical; and although Temple made a shining ex- 
ception of Don Quixote (which he valued quite as highly as Swift 
did), his judgment of Swift's adored Rabelais was harsh and nar- 
row. 4 Temple mentions La Rochefoucauld to oppose him; 5 
Swift, to praise him. Mock-epic, of which Swift was to write the 
finest example in English prose, Temple judged as pernicious to 
poetry and virtue alike. 6 Though both Temple and Swift es- 
teemed conversation as the most deeply satisfying kind of recrea- 
tion, Temple detested the raillery and witticisms in which Swift 
rejoiced. 'Those squeez'd or forc'd strains of wit', Temple called 
them, 'that are in some places so much in request, tho' I think 
commonly men that affect them are themselves much fonder of 
them than any of the company.' 7 'Raillery is the finest part of 
conversation', said Swift 8 ; wit and ridicule are 'the meaner parts' 
of conversation, said Temple. 9 

Similarly, the puns and word games which Swift loved held 
small appeal for Temple; in fact, his difficulties as a diplomatist 

1 Ball I. 365. 2 Memoirs in. 67. 

3 Ball i. 4. (Cf. Swift's 'fatal bent of mind, / Still to unhappy restless thoughts 
inclined' Poems I. 55.) The similarity between Temple's and Swift's expressions 
suggests that the 'person* may have been Temple; there are several methods of ex- 
plaining away *in Ireland*. In his letter to the Athenian Society the 'person of ... 
honour* is probably Temple (Ball i. 7) . 

* Of Poetry, pp. 328-9. * Miscellanea m. 292. * Of Poetry, pp. 329-30. 

7 Memoirs n. 358; cf. Miscellanea m. 325-6, 335-6. 8 T. Scott xi. 71. 

9 Miscellanea in. 326. 



were sometimes aggravated by his distaste for codes : c God al- 
mighty has given it to other men to make cyphers and to flie, but 
to me only to walk upon plain ground and to read plain hands.' 1 
Yet while Temple's tolerance for coarse words and coarse images 
falls well below Swift's, his relish for sexual innuendo appears far 
stronger. When Swift, for example, wrote biographies of the 
Norman kings, he suppressed most of the references to their 
illicit passions ; Temple, however, devoted almost two pages of 
his Introduction to the History of England to the probable superiority 
of fornication over lawful procreation as a means of conceiving 
heroes 2 ; and Temple's letters not uncommonly include risqu6 
gallantries of a sort that Swift never permitted himself. 

These contrarieties do not affect the strength of Swift's early 
reverence for Temple and his desire to imitate him. e l never read 
his writings', Swift said at twenty-five, 'but I prefer him to all 
others at present in England.' 8 Eight years later, he could still 
declare that Temple had 'advanced our English tongue, to as 
great a perfection as it can well bear'. 4 We may then suppose that 
Swift tried at first to meet the requirements which Temple set up 
for poets i.e., that they should 

raise up the esteem of some qualities, above their real value, rather 
than bring every thing to burlesque, which if it be allowed at all, 
should be so only to wise men in their closets, and not to witts, in 
their common mirth and company. . . All the wit, [which Waller] 
and his company spent, in heightning love and friendship, was 
better employ' d, than what is laid out so prodigally, by the modern 
wits, in the mockery of all sorts of religion and government. 5 

That Swift's dear teacher, St George Ashe, had made similar 
pronouncements, should have heightened his will to reach a 
standard so remote from the gifts we remember him for. 

While such tendencies undermine the literary interest of the 
juvenile poems, they establish their documentary importance. 
As expressions of Swift's positive values, they are unique. Never 

1 Select Letters, 1 701, p. 1 1 ; see also Woodbridge, pp. 87-8; Letters m. 305. 

2 Pp. 85-7. 8 Ball i. 365. * Davis i. 258. 

5 Letter of August 1667 to Lord Lisle (Works, 1720, n. 40). 



was he to give such unequivocal, simple, direct embodiment to 
his ideals. And as one might expect from a man who made satire 
his employment, these ideals are more remarkable for their in- 
tensity than their originality. One effect of setting up a dead 
grandfather or an absent mother as models of conduct, was evi- 
dently to release Swift's notions of virtue from even the limits that 
a naive child with visible parents can impose upon such notions. 
A man already in his middle twenties might well be excused from 
the celebration of such incandescent perfections as Swift lent to 
his subjects which included the king, the Athenian Society, 
Archbishop Bancroft, and Temple. 


Charity ("doing good 3 ), fortitude, and justice are the main- 
springs of the ode c To the King', which Swift wrote while he was 
in Ireland, between July 1690 and August iGgi. 1 To set off the 
transcendent merits of William, Swift opposes him both to James 
II and Louis XIV; it is obvious that now, as throughout his life, 
the Revolution appeared to Swift a blessing. The seven stanzas of 
the poem are focused upon the benevolence of William III and 
his courageous generalship at the Battle of the Boyne. Swift de- 
nounces the sloth of the English people and the bigotry of the 
Scottish, who opposed William. He makes much play with the 
stereotype antithesis between c great' and e good 9 : Louis XIV may 
seem great, but is not, because he is bad and tyrannical; William 
is truly great because he uses his power for good. 

In this poem, Swift takes himself seriously, and is rather a 
moralist than an entertainer. While his conceits are not impene- 
trable, several of the allusions were difficult even at the time of 
composition, and a couple are annotated by Swift. The manner is 
pompous and rhetorical, with an excess of parentheses, apos- 
trophes, exclamations, and other inflated figures; yet it is under- 
pinned by an indignation which carries conviction. As a sibyl, the 

1 John Dunton quotes stanzas 1-2 in The Dublin Scuffle, 1699, p. 379. Swift may 
have attempted to publish the poem through Dunton, 



young poet fails, but as a Juvenal he has some moments of success. 

However, he also provides himself with a means of shifting the 
point of view from his own person, by dramatizing the constant 
presence of the muse, or spirit of poetry. It is significant that this 
device accords with the doctrine and practice of Cowley. 1 For 
Swift not without irony traced back his high estimation of 
Cowley as far as his own sixteenth year. 2 The much-praised poet, 
who had died about the time Swift was born, was now honoured 
chiefly for his odes and his love poetry ( The Mistress) ; the un- 
finished epic, the Davideis, was no longer read ; his ingratiating 
prose style had yet to be reassessed. Not only the metrical struc- 
ture of Swift's ode but also its details echo Cowley. 3 

Certain themes which are common to all Swift's early poems 
appear first in this ode to the king. One is the representation of 
the poet as a seer, with powers of vision denied to common hu- 
mans. Another is Swift's angry contempt for the mob, the vulgar 
mass of unthinking people. Another is the polar distinction be- 
tween illusion and reality, also expressed as the contrast between 
true and meretricious virtue. Finally, there is Swift's patent 
eagerness to find and to acknowledge a genuinely heroic figure. 

Certain recurrent images and oppositions also appear. The 
celestial is coupled with vision, reality, and virtue, and opposed 
to the earthly; similarly, the solid is opposed to the vaporous, the 
bright to the murky. William III, establishing himself over the 
giddy British populace, is like a bright-faced patrician god visit- 
ing Hades. In a parallel way, stars are opposed to earthly stenches, 
sudden flight upwards to sudden falls. William is like a true sun; 
Louis is like a false and falling meteor. With such motifs are as- 
sociated violent, characteristic feelings, like profound awe of the 
hero, or revulsion from the crowd. But though there is a plethora 
of epithets, there is no relaxation, of pose, no sign of humour. 

Swift's actual development is foreshadowed, as one might ex- 
pect, less in the praising than in the blaming. Here are some 

1 See Gowley's introduction to his Second Olympic Ode of Pindar. 

2 Davis n. 1 14. 

3 Cf. the last stanza of Swift's ode, and vi and x of Cowley *s Ode upon His Majesty's 


points where the weak or erratic rhythms grow firm, and where 
the language grows expressive : the lines on James IPs fate 

The remnant of a falling snuff. 
Which hardly wants another puff , 
And needs must stink when e're it dies. 

Or the lines on Louis XIV's false 'meteor' (the parting attack on 
the Sun King is itself echoed in a paragraph of A Tale of a Tub 1 } : 

Stay but a little while and down again * twill come, 
And end as it began, in vapour, stink, and scum. 

But these, alas, are probably the verses of which Temple would 
have least approved. 


Swift's ethical fervour is more outspoken in the ode to the 
Athenian Society. While revisiting Ireland and again at Oxford, 
he had heard about an extraordinary publication called the 
Athenian Mercury. It was professedly the journal of an anonymous 
group of learned men, comparable to the Royal Society. The 
members functioned mainly as a public information bureau, an- 
swering miscellaneous inquiries on a weird variety of topics. The 
ordinary run of questions was not absurd, but problems like * Why 
are eunuchs never afflicted with the gout ?' and 'How can witches 
contract their bodies into so narrow a compass, as to convey 
themselves through a key-hole?' were common. 2 

At Moor Park, Swift found that Temple himself approved of 
this periodical and had participated in its programme. The great 
man's naivete and Swift's innocent trust in his opinion are mea- 
sures of the muddle which characterized even the educated 
gentleman's attitude toward experimental science at the end of 
the seventeenth century. Out of timidity and diffidence, Swift 
dared not follow his own literary schemes but looked to Temple 
and Lady GifTard for hints. Yet he must have felt hot indeed for 

1 Tale, pp. 165-6. * Vol. n, nos. ao, 28. 


fame if, even though assured of their respect for the Athenian 
Society, he bound one of his first literary ventures to the wheels of 
that e crazy chariot'. 1 

The period of about a year and a half which had passed since 
Swift wrote his ode to the king had given him time to overcome 
his leanings toward clarity of expression; but while the new work 
is a distinct retreat from intelligibility, its argument can, with 
patience, be fathomed. Through twelve diffuse stanzas, com- 
posed in nine days, 2 Swift congratulates the nameless subjects of 
his adoration (actually a bookseller and some hack assistants) 
upon their great part in reviving learning after the wars of 
168891, which are hopefully treated as though ended with a 
peace. He apologizes for the effusiveness of his own tribute, de- 
scribing it as impetuous but well intentioned ; and he urges them 
not to mind the e sect' who cry them down. He praises their in- 
difference to fame and the steadiness with which they maintain 
orthodox religion in the face of sniping doubters. But while he is 
delighted by their skill in refining philosophy and in removing its 
growths of pedantry or scholasticism, he reproves them for their 
compliments to women, which have turned the heads of the fe- 
male sex. At the end he voices his anxiety about the future of the 
Athenian project, for he is afraid that e censure and pedantry and 
pride 5 will break it down. 

There is, as in the ode to the king, some conflict between the 
ostensible and the underlying themes of the poem, as Swift's own 
concerns shoulder aside the figures which he starts from. He can- 
not sustain an initial attempt at objectivity., and shifts awk- 
wardly (though in the pindaric tradition) from the third or 
second person to the first. Instead of separating himself from the 
figure of the muse, he confuses himself with her; not only does the 
rhetorical value of the device thereby largely evaporate, but a 
grotesque, unnecessary ambivalence is introduced between her 
gender and his. She first appears as a dove; but that is displaced 

1 Graik i. 44. Yet Defoe, Motteux, Tate, and Richardson also contributed 
panegyrical odes to the Athenian Mercury. 

2 Ball i. 363. 


by the spirit of philosophy, represented as a woman; the woman 
becomes suddenly a new embodiment of the muse; however, be- 
fore one can take in the change of role, Swift makes her stand for 
womankind, and not poetry or philosophy. He now makes a 
comparison between the two sexes, and gives the advantage of 
course to 'nobler man*, endowed with learning and wit. Through 
all these metamorphoses it grows obvious that Swift has not lost 
the preoccupation with young ladies that he revealed in his 
Leicester flirtation; and the freshest lines of the poem, as well as 
the freshest feelings, are those in ridicule of the picture which he 
finds so provocative : 

With a huge fardingal to swell her fustian stuff, 
A new commode, a top-knot, and a ruff, 
Her face patch? t o'er with modern pedantry. 

Another honest impulse breaks through the surface in Swift's 
longing for literary fame. Just as in his treatment of women, he 
here ridicules the object which allures him. Illusion versus reality, 
vapour versus substance these motifs (employed with other 
functions in the ode to the king) are used to oppose apparent, 
popular fame to real virtue and piety. But the strength of his de- 
sire is evident from the intensity and minuteness with which he 
explores the hackneyed theme. 

Just as, in the earlier ode, he had lavished celestial and divine 
qualities upon William, so here, though with even less propriety, 
he bestows the same attributes upon those invisible members of 
the Athenian Society who have replaced the king as protagonist 
and hero. Images which were stale enough before are reheated 
for the new poem: e.g., flies buzzing about the king reappear as 
flies buzzing about wit. The critics of the Society replace the 
rebels against the king. To abase himself before the honoured 
object, Swift associates himself with the despised c blind and 
thoughtless croud*. 

It is easy to see that Swift's point of departure has been Cow- 
ley's Ode to the Royal Society* The quibbling over the sex of 
*Pons,p. 177, 



'Philosophy', however (in Cowley's first stanza and Swift's ninth 
and tenth), is a token of the degree to which Swift diverges from, 
his predecessor. Where Gowley makes the word masculine. Swift 
makes it feminine, and even dwells on the resulting imagery. He 
also makes a more meaningful departure, in opposition to Cow- 
ley's praise of the natural philosophers' minute and experimental 
precision; for Swift welcomes the Athenians as being Christian 
humanists like himself, and resisting the Hobbesian 'new modish 
system of reducing all to sense 9 - 1 He shows his religious orthodoxy 
through a significant attack upon Epicurean atomism, a fashion 
which was now declining as it became associated with atheism. 
Richard Bentley, in the first Boyle lectures, ripped up the Epi- 
cureans by using the mathematical and experimental demonstra- 
tions of Newton's cosmology to disprove their mechanical de- 
terminism and their explanation of the origin of the universe by 
chance as c but a crowd of atoms justling in a heap% in Swift's 
phrase. Cowley had been eager to prove that Authority was a 
ghost dispelled by Bacon; Swift jeers at the wits who make Provi- 
dence out to be an illusion, and he boasts, C I believe in much, 
I ne're can hope to see.' Thus the ode to the Athenian Society, 
without parodying the ode to the Royal Society, does in some 
ways reply to it. Though Cowley died the year Swift was born, 
his philosophical outlook was less conservative : he hoped for an 
end of some mysteries which it was Swift's instinct to preserve. 
One cannot, after all, discover anything for science unless one 
believes it is 'there*, ready to be found, 2 whereas for Swift the 
sublimest truths reach us neither through discovery nor demon- 
stration but rather intuition or revelation. 


In another ode written about the same time, the old furniture re- 
appears but with fresh elements and with renewed transparency 
of expression. Both changes are probably due to the same source, 

1 Cf. Hobbes, English Works, 1839, 1. 389. 

2 Cf. Michael Polanyi, The Study qf Man (Chicago 1959), p. 35. 



the poem's subject; for it is addressed To the Hon ble Sir William 
Temple. This time, Swift has a hero both whom he sincerely ad- 
mires and of whom he possesses direct knowledge: these facts no 
doubt account for his dropping the veil of obscurity. But further- 
more, the imagery and the central ideas of the poem are largely 
derived from two of Temple's essays. Upon Ancient and Modern 
Learning and Upon the Gardens of Epicurus. 

This is not to say that Cowley's influence has declined : it is 
pervasive still, but belongs rather to the framework of the poem 
than the core. The manner, diction, and imagery of Swift's 
opening recall Cowley's To Mr. Hobbes and his e Leaving Me' 
(The Mistress, poem 1 1, lines 17-22). Swift's lines 50 and follow- 
ing echo the first lines of two odes by Cowley 1 ; his eleventh stanza 
is close in part to the end of 'The Thraldom' ( The Mistress, poem 
2) and in part to the seventh stanza of The Complaint. Swift's con- 
cluding thirty lines are similar to the last two stanzas of Cowley's 
Destiny i the end of his last stanza echoes stanza 6 of The Com- 

The typical aspects of Gowley's style, however, are least like 
that of Swift in maturity. Where one is expansive, the other is 
concise; where Cowley repeats an expression, Swift varies it; 
where Cowley loves to portray his own nature, Swift puts on a 
mask; where Cowley is a rhetorician, Swift is a moralist. Remove 
the posings from the ode to Temple, and what remains is Swift 
under Temple's instruction: the distrust of school logic, of specu- 
lation, and of pedantry ; the delight in looking behind-stage and 
finding the trivial motives of great events; the eagerness to burn 
incense before true greatness; and a bitter, almost Hebraic anger 
at the jungle wrought by corrupt men in the lord's garden: 

How plain I see thro* the deceit! 

How shallow! and how gross the cheat! 

Temple's values and moral ideals are in the poem. The 
'pleasures of retreat' sung by Swift, though a common theme in 

1 Ode v (commending the reign of Charles II) and the second ode in the essay 
on Cromwell's government. 



Gowley and in Restoration verse generally, meant more than a 
play on Horace's secretum iter et fallentis semita vitae. It is from 
Temple that Swift was learning to praise the 'private path of 
stealing life' ; and when in later years the Dean bypassed his ca- 
thedral city to cultivate the gardens of his rural vicarage, he was 
indulging a taste fostered at Moor Park. The preferences of sin- 
cerity to dissimulation in political affairs, of peaceful to martial 
victories, of dilettantism to specialization, are other lasting ideals 
encouraged if not implanted in Swift by his master, and animated 
in this ode. 

These public, explicit themes have a private, biographical sig- 
nificance. But before examining that in detail, we may briefly 
notice the formal structure of the poem. It is a series of rather dis- 
connected stanzas, each of them a meditation upon some facet of 
Temple's character. In imagery and rhetoric this poem shares a 
great deal with the other odes. Its first stanza, for example, is a 
reworking of a motif from the ode to the Athenian Society (elev- 
enth stanza). Similarly, the muse appears here as before, never 
acting as a genuine alternative to the poet's self-regarding point 
of view, but again dissolving, from the emblem of poetic inspira- 
tion into a rustic nymph, and then becoming a task-mistress. 
Swift's new subject, Temple, bows under as many superlatives 
as his old. This protagonist is not only 'good and great 3 but learn- 
ed as well. The mob is now used to mark pedantry : the grubby 
accumulations of arrogant and vulgar university dons (Swift's 
residence in Oxford probably revived earlier feelings) are con- 
trasted with the genteel learning of the hero. 

A vital change is the casual treatment of the villain a 'ser- 
pent' embodying the treachery which destroyed Temple's suc- 
cess at court. Swift seems so powerfully dazzled by his master's 
virtues that he merely glances at the enemy. Instead, the theme 
of his own comparative unworthiness (touched on in the ode to 
the Athenian Society, third stanza) grows loud; he ties himself to 
the 'barren earth', the antithesis of Temple's 'spirit so divine'. 

The ode to Temple may even contain (perhaps not quite in- 
tentionally) some ironic squints at the ode to the king. In that 


earlier poem, Swift says that William is remarkable for at once 
'doing good' and 'being great'. His bravery is of the order which 
is usually met 6 only in romance 3 . In the later poem, Temple is 
learned, good, and great 5 , a combination which 'we ne'er join'd 
before, but in romances meet 9 . Martial courage, however, is not 
Temple's distinction; he is crested with the superior 'laurel got 
by peace*. Instead of flaunting 'scepter, crown and ball 5 , he has 
the wisdom to expose statecraft and court intrigues as 'juggler's 

If we now examine carefully the personal, non-literary aspects of 
the poem, it will be found to mirror the whole relationship be- 
tween patron and pupil at the time when it was written. Temple's 
attitude toward Swift receives a special illumination from a 
younger diplomat's report of a session with Temple in 1677 : 'He 
held me in discourse a great long hour, of things most relating to 
himself, which are never without vanity; but this most especially 
full of it, and some stories of his amours, and extraordinary abili- 
ties that way, which had once upon a time very nearly killed 
him. 31 A man who enjoys such an indulgence at fifty does not 
sacrifice it at sixty-five. Since Temple had lost the most suitable 
vessel for his confidences when his son committed suicide, he 
could not have stood out long against the eager attentiveness of 
the impressionable young man who arrived so soon after that 
catastrophe. To the lonely refugee, who had never known his own 
father, and whose nearest substitute for one (Uncle Godwin) was 
dying or dead, the role of a son must have been all too easy to 

The consequent relationship probably developed the more 
painlessly as Swift's diffidence was matched by Temple's one- 
sided conception of paternal dignity: for with both parties so re- 
served, there would be few overt gestures to make them conscious 
of what was happening: 'Constrained as we are in our demeanor 
toward [parents] ', Temple had written, 

1 Woodbridge, p. 190. 



by our respect and an awful sense of their arbitrary power over us, 
which though first printed in us in our childish age, yet years of 
discretion seldom wholly wear out. Besides, a certain strangeness 
wrought between those relations by the disagreement of age and 
consequently of customs, which is hardly so far wrought out by the 
greatest kindness as to admit such a freedom and confidence as is 
common between friends of our own choice. 1 

For all Temple's boasts of simplicity, he displays, throughout his 
memoirs, a visible pride in the grace with which he managed 
points of ceremony. 2 He had a deep respect for his own presence 
and could hardly have avoided acting toward Swift with a con- 
descension which at times injured his protege's self-esteem. 

By encouraging Swift's poetical beginnings, however. Temple 
further enriched the newcomer's filial role, since though Temple 
had intended his own son to act as his literary executor, 3 it was to 
Swift at last that he would do 'the honour, to leave and recom- 
mend . . . the care of his writings'. 4 Swift's longing for an author's 
fame, c a mad reversion after death', is not only the theme of his 
final stanza but also as he intimates the element through 
which he establishes a real kinship with his master. 

The ode to Temple proves that Swift in turn not only shared 
but correctly understood the ideals of the foster-parent from 
whom he differed so profoundly in character. Above all, the ode 
dramatizes the complexities of Temple's passion to retreat from 
the public life. For while the disenchanted diplomatist truly 
relished the garden retirement of his middle and old age, he 
wished also to enjoy in full the credit of his renunciations. Horace 
was not the only man Temple meant to praise when he wrote, *It 
was no mean strain of his philosophy, to refuse being secretary to 
Augustus, when so great an emperor so much desired it.' 5 It is 
obvious who Temple thought had suffered the greater loss when 
he and the court parted ; and it is obvious that he had not always 
felt indifferent to the prospect of becoming secretary of state. 6 

1 Woodbridge, p. 21. a E.g., Letters m. 270-6. 8 Memoirs n, sig. A4 V . 

4 Davis i. 259. Of. Ball m. 301. 

5 Miscellanea n. 93 (Upon the Gardens of Epicurus) . 

6 Of. Letters ir. 308, in. 334-7. 



When he described in the Memoirs his refusals to take that post, he 
dwelt fulsomely upon the directness of the offers which had been 
made to him and the aplomb with which he had declined them. 1 
He must have cultivated similar allusions in conversation : 'What 
a splutter 9 . Swift later said, 'Sir William Temple [made] about 
being secretary of state.' 2 Nevertheless, although Swift was to 
hymn the charms of seclusion as loudly as his master, he would 
yield to them less single-mindedly at last always, under a screen 
of indifference, informing himself of events, and usually trying, 
when he could, to have a hand in them. 

Swift's attack in both this ode and that to the Athenian 
Society upon 'philosophy ! the lumber of the schools' reflects 
the estimate which Temple had formed of natural, as distinct 
from moral, philosophy. For in this judgment he belonged to the 
opposite side from St George Ashe: C I know no end', Temple 

[natural philosophy] can have, but that of either busying a man's 
brains to no purpose, or satisfying the vanity, so natural to most men, 
of distinguishing themselves by some way or other, from those that 
seem their equals in birth. . . I know no advantage mankind has 
gained by the progress of natural philosophy, during so many ages 
it has had vogue in the world, excepting always, and very justly, 
what we owe to the mathematicks. 

While moral philosophers have always agreed (sic), Temple said, 
natural philosophers have always quarrelled : 

As, whether the world were eternal, or produced at some certain 
time ? Whether if produced, it was by some eternal mind, and to 
some end, or by the fortuitous concourse of atoms, or some par- 
ticles of eternal matter ? Whether there was one world or many ?. . . 
There were the same contentions about the motions of the heavens, 
the magnitude of the celestial bodies, the faculties of the mind, and 
the judgment of the senses. But all the different schemes of nature 
that have been drawn of old or of late by Plato, Aristotle, Epicurus, 
Des-Cartes, Hobs, or any other that I know of, seem to agree but 

1 Memoirs n. 14-15, 272-6, 385-7; Memoirs m. 1-8, 29, 57-8, 98-100. 

2 Journal 3 Nov. 1 71 1 ; c 1 1 Nov. 1710. 



in one thing, which is, the want of demonstration or satisfaction, 
to any thinking and unpossessed man. 1 

This is the doctrine which Swift commends in his ode, which he 
was to paraphrase in A Tale of a Tub* and which would remain 
his considered judgment throughout his life. 

In the ode to Temple also appears the epitome of the grand 
view of history and politics which Swift inherited from his master. 
Like most historians and political theorists writing in the seven- 
teenth and eighteenth centuries, Temple wished to trace public 
events not to their divine but to their human origins. He assumed, 
however, that the men to be examined were the heads of states, 
and that their separate, essential natures, rather than the cir- 
cumstances of their people, determined the central line of his- 
tory. In this assumption, to which Gibbon has given its most 
glorious illustration, Swift would be at one with him ; and Swift 
would have repeated with perfect assurance Temple's declara- 
tion that c all great actions in the world, and revolutions of states 
may be truly derived, from the genius of the persons, that conduct 
and govern them.' 3 

From such a standpoint, any political philosopher, but es- 
pecially one with Temple's career behind him, would be bound 
to exaggerate the effect of intrigues upon history, and to be im- 
pressed by how much could proceed from how little the doc- 
trine of maxima e minimis once more. Swift's expression many 
years later need not be considered a paraphrase of Temple's, 
since the generalization was a natural outcome of both men's pre- 
suppositions, and it was also a commonplace of historians well 
into the era of Bolingbroke and Voltaire. But the parallel be- 
tween teacher and pupil is worth a glance. "Upon how small acci- 
dents the greatest councels and revolutions turn', is Temple's 
language ; and again, 'How small shadows and accidents some- 
times give a rise to great actions'. 4 Swift in maturity would put it 
as 'the greatest events depending upon slight and mean causes*. 5 

1 Miscellanea n. 84-5 (Upon the Gardens of Epicurus). * Pp. 166-7. 

8 Introduction to the History of England, p. 301. 

4 Memoirs n. 334; Letters i. n. e Davis vm. 172, as varied on p. 229. 


In his ode to Temple, Swift's tone has the excitement which one 
attaches to a discovery : 

MethinkS) when you expose the scene > 

Down the ill-organ* d engines fall; 
Off fly the vizards and discover all . . . 

Great God! (said 7) what have I seen! 

On what poor engines move 
The thoughts ofmonarchs, and designs of states. 

On the contrary, Temple as Swift after him prided himself 
upon an aversion to secret political machinations and upon a de- 
votion to simple, honest goodwill. They both had a way of re- 
ducing public affairs to questions of personal honour. Temple's 
success as an ambassador had hinged upon his knack of treating 
other ministers as private acquaintances whose faith he never 
doubted and whom he expected to trust him completely. Swift, 
in his later career, could not admire a statesman without believ- 
ing that he told the truth to Dr Swift; and the great pamphleteer 
would rarely attack a statesman whom he did not call both a liar 
and an intriguer. The purity of their own motives, neither 
Temple nor Swift stopped to suspect: *I can truly say', wrote 
Temple, 'that of all the paper I have blotted, which has been a 
great deal in my time ; I have never written any thing for the pub- 
lick without the intention of some publick good' 1 ; and he would 
have made the same declaration with regard to his political ac- 
tivities. As Swift sings in the ode, Temple's armour was within 
himself, 'made up of virtue and transparent innocence' ! 

They could the more readily agree in this analysis as Temple's 
public career had been largely taken up with efforts to end or 
prevent wars, and Swift's first major campaign as a polemicist 
(1710-13) would be spent on similar aims. The supreme im- 
portance of civil and international peace has always been pre- 
sent to reflecting men, but in the seventeenth century it was an 
obsession of philosophers; and Hobbes had made it the over- 
riding principle of his political and psychological systems. So it is 

1 Miscellanea m. 99. 


not surprising that among the highest desiderata of human life, 
Temple placed peace second only to health. 1 On mercenary 
standing armies, both he and Swift were also agreed, for both 
connected them with violence and tyranny. 'Standing-forces or 
guards in constant pay 3 , wrote Temple, 'were no where used by 
lawful princes in their native or hereditary countrys, but only by 
conquerors in subdued provinces, or usurpers at home ; and were 
a defence only against subjects, not enemies.' 2 However, they 
differed in their estimates of generalship. Though Temple de- 
tested war, he could respect and admire military genius as em- 
bodied for example in Prince Maurice of Nassau, or Turenne ; 
he even admitted the glory of war. 3 Swift might admire the flair 
of a commander like Peterborough, or the courage displayed by 
Ormonde at Landen; but after writing his ode to the king, he 
would not again praise generalship. In the ode to Temple, of 
course, these ramifications do not appear; and the poet, with 
more paradox than coherence, can acclaim his hero's 'laurel got 
by peace' because 

It melts the sword of war > yet keeps it in the sheath. 

One final ideal which Temple incarnated for Swift, as only 
Bolingbroke would ever do again, was the vision of polite learn- 
ing. After Trinity College and after Oxford, the great man's ap- 
parent familiarity with the languages, the arts, and the sciences 
not only awed but delighted Swift. Temple was, as a matter of 
fact, more 'modern' and comprehensive in his range than Swift; 
for he approached a unified conception of aesthetic taste such as 
was realized only in the nineteenth century. He displays a feeling 
for music and the plastic arts which Swift (for whom the sole 
interesting art besides literature was architecture) never dimly 
adumbrated.* Temple's contempt for specialized, speculative, or 
truly technical inquiries, and for those who pursued them, awoke 
Swift's eager sympathy. C A11 that which we call scholastick or 
polemick', said Temple, with an eye on the academic, 'serves, I 

1 Miscellanea m. 1045. 2 Observations upon the C/.P., p. 248. 
8 Letters n. 27. 4 E.g., Miscellanea m. 249-50* 


fear, among us, for little more, than to raise doubts and disputes, 
heats and feuds, animosities and factions, in all controversies of 
religion or government/ 1 We have too long been led astray/ 
wrote the author-to-be of the Battle of the Books, addressing its 
inspirer : 

9 Tisyou must put us in the way; 
Let us (for shame) no more be fed 
With antique reliques of the dead. 


Swift did not agree with Temple in everything. When he failed to 
do so, instead of criticizing him, he found an embodiment else- 
where of the harmony which he sought. The severest lack in 
Temple, as Swift saw him, was probably religion. It is not that, 
like most men of the Restoration, Temple expressed an angry 
distrust of elaborate theological arguments 2 ; for on this Swift 
was bound to agree with him. But Temple, in spite of his loyalty 
to the Established Church, approved the sort of religious tolera- 
tion recommended by his tutor Gudworth and practised by the 
Dutch. While the young Swift may just possibly not have con- 
demned this, his own tutor, Ashe, would certainly have taught 
him to do so; and the mature Swift would have abominated 
Temple's plan to encourage Protestant immigration into Ireland 
e by some large degree of liberty in matters of religion'. 3 This is 
just the sort of proposal that Swift was to denounce (in 1710) as a 
contribution to atheism and a deliberate attack upon the national 
church : 

These men take it into their imagination, that trade can never 
flourish unless the country becomes a common receptacle for all 
nations, religions, and languages; a system only proper for small 
popular states, but altogether unworthy and below the dignity of 
an imperial crown. . . This pedantry of republican politicks hath 
done infinite mischief among us: To this we owe those noble 

1 Miscellanea n. 179. * Miscellanea m. 260-6. 

8 Essay upon the Present State of Ireland, published only in Select Letters (i 701), p. 214. 


schemes of treating Christianity as a system of speculative opinions^ 
which no man should be bound to believe; of making the being and 
the worship of God, a creature of the state. 1 

While Temple was a faithful communicant of the Church of 
England, he distrusted the material wealth of the church, and 
doubted the purity of ecclesiastical motivations, in a manner that 
was repugnant to Swift. Temple frowned on the granting of civil 
power to clergymen; Swift (in 1711) recommended it. 2 Swift, 
when he came to write on tithes, defended them as e if not of divine 
original, yet at least of great antiquity' 3 ; Temple said they were 
of late growth. 4 In the Introduction to the History of England, Temple 
went out of his way to argue that the medieval clergy, though 
pretending to pursue the 'greatness of the holy church', sought in 
fact the 'honours, power, and riches of the church-men 5 , 5 This 
sort of pronouncement, made in a particular digression, Swift 
must have contradicted. The rift which lay between the two men 
glares at us from the ode written by Swift upon William San- 

One rarely mentions this saintly Archbishop of Canterbury 
without quoting Dryden's eulogy of him as the priest 

whom, shunning power and place , 
His lowly mind advanced to David? s grace. 

Consecrated primate in 1678, he was one of the seven bishops 
who, ten years later, went to the Tower sooner than obey those 
edicts of James II which seemed to undermine the Established 
Church. However, he refused either to ally himself, during the 
Revolution, with William of Orange, or, afterwards, to accept 
the new king. In February 1690/1, Sancroft along with five 
other nonjuring bishops and about four hundred clergymen who 
lost their livings was deprived of the primacy; in June he was 
ejected from the palace at Lambeth. A year after the deprivation, 

1 Davis m. 48-9; cf. also p. 169. * Journal i. 347-8- 

8 Landa, p. 130; also pp. 123-35. * Observations upon the U.P., pp. lo-ii. 

B Pp. 130-1. 



he sponsored and helped to arrange a succession of episcopal 
authority in the nonjurors; in November 1693 he died. He 
was, Swift wrote a year and a half before Bancroft's death 'a 
gentleman I admire at a degree more than I can express 3 

partly by some experience of him, but more by an unhappy rever- 
end gentleman my Lord the Bishop of Ely with whom I used to con- 
verse about two or three years ago, and very often upon that sub- 
ject. 1 

While Sancroft was no Jacobite, Francis Turner, deprived, 
nonjuring Bishop of Ely, corresponded with James II not only in 
1690 but later, and probably worked for his restoration. In 1694 
he was almost invited to attend James at St Germains ; in 1 696 he 
was twice arrested. That Swift, so late as May 1 692, should have 
boasted of Turner's acquaintance, both suggests something of the 
young man's ecclesiastical conservatism and indicates that he de- 
fied scandal in religious and political controversy as freely as 
he did in his private conduct. As Scott said, 'Whatever were 
[Swift's] principles in civil politics, he was uniformly a staunch 
high-churchman. 32 Swift's attitude toward the nonjurors was 
probably like William Wake's judgment in 1 689 ; though he him- 
self had taken the oaths, Wake wrote to an acquaintance with 
other leanings, C I am sure such as you cannot [refuse the oaths] 
for interest, and where men act and believe according to their 
conscience, though apparently against their interests, though I 
may differ from them in any opinion, I cannot but applaud their 
honesty, and wish them a satisfaction.' 3 The spectacle of the 
modest Sancroft martyring himself for the sake of noble but lost 
causes excited in Swift the special homage which he reserved for 
such defeated heroes as Charles XII after Poltava and the Earl of 
Oxford in the Tower. Unfortunately, Swift's movements from 
1689 to 1692 are too intricate for one to be sure where he had 
occasion to meet either of the two bishops. 

Partly to please Turner, Swift undertook to produce a large 
pindaric ode upon Sancroft. Beginning to compose it around 

1 Ball i. 363-4. * W. Scott xrv. 3. * Sykcs, William Wake x. 45. 



January 1 Sgs, 1 he confessed, five months later, that it was still on 
his hands : 

I cannot finish it for my life, and I have done nine stanzas and do not 
like half of them, nor am nigh finished a but there it lies and I some- 
times add to it, and would wish it were done to my desire, I would 
send it to my bookseller and make him print it with my name and 
all ; to show my respect and gratitude to that excellent person, and 
to perform half a promise I made his lordship of Ely upon it. 2 

Eleven whole stanzas and a twelfth incomplete are preserved. 
In the Miltonic opening lines, 3 Swift praises truth, of which he 
says Sancroft is the earthly image. Through the old man's sta- 
bility Swift contrasts the wavering motion of the state and the 
mob with the firmness of the church. Drawing a parallel between 
the repudiation of Sancroft by the English and the treatment of 
Christ by the Jews, he condemns the viciousness of his own times. 
Above the impermanence of the many, Sancroft, a star of regu- 
larity in his retirement, is like the sun of Paradise Lost, Book m. 4 
Swift asks for his guidance and complains about the hardships of 
the church, perhaps alluding to William IIFs indulgence of the 
dissenters. Treating the archbishop's withdrawal from office as 
an exaltation, Swift foresees his glory in heaven and hopes he will 
support his faltering brethren and help reform the nation. 

The praise of Sancroft is, as one might expect, exaggerated ; 
nevertheless, it is precise : he stands for religious truth, heaven- 
sent to show men the only rule of life, c the way which leads to 
Christ'. The Anglican Church too receives more than perfunc- 
tory oblations, and Swift unequivocally gives it precedence over 
the state : 

Why should the first be ruirfd . . . 
To mend dilapidations in the last? 

Similar exclamations toward the end, against those who c tear 

* Ball i. 364. * Ibid. 

8 Cf. stanza i and Paradise Lost m. 1-12 ; stanza ii and v. 95-1 13. 
4 Cf. 11. 149-53 and Paradise Lost m. 576-86. L. 116 is derived from Cowley's 
"The Passions' (The Mistress, poem 16), 11. i~a, 32-3. 


religion's lovely face', while not trenchant, are not hollow. At 
twenty-four or five, Swift already possessed the temperament of a 
priest. His next step would be to mount the pulpit. 

Apart from its value as a biographical document, the ode 
shows another stage in Swift's poetical evolution. Bancroft, 
though less remote than the king, was far less familiar than 
Temple; the style of the ode is correspondingly involved, tor- 
tured by conceits, and almost as dense as the ode to the Athenian 
Society. Although a new influence, that of Milton, has been add- 
ed to that of Gowley, the rhetorical scheme has shifted only a 
little. A static hero, a rapt poet, and a mobile muse act out their 
usual confrontations ; old phrases are reshuffled. 1 Just as William 
III had been opposed by a mob of English and Scots, so Bancroft 
is opposed to a herd of weak-faithed plebians. The archbishop 
takes his heavenly place among the stars ; he is eulogized as both 
great and good like the king but his added virtues are rather 
holiness and justice than the learning attributed to Temple. 
Swift's scientific imagery is fuller here than before, and drawn 
more from Copernican astronomy and Cartesian optics than 
Epicurean physics (Hobbes could have supplied all the ma- 
terials) . 

It would not be merely hindsight to suggest that Swift's in- 
ability to finish the poem was natural. He brings both Christ and 
the king in, and tries to exalt them both without reducing the in- 
tensity of his reverence for Bancroft. But one cannot treat a man 
as perfect who has refused to acknowledge a monarch, and then 
treat as perfect the monarch whom he has rejected. Contrariwise, 
Swift draws a lengthy but awkward analogy between Bancroft's 
ordeal and that of Christ : while the conceit is less grotesque than 
Origen's parallel of Christ on the cross and Ulysses tied to the 
mast, it is almost as distracting. Swift has been praising a man for 
divine virtues, and then brings in divinity itself; the effect can 
only be to dwarf Bancroft. The villains of the piece are the Puri- 
tans ; yet one can hardly damn them for insubordinative religious 
zeal and in the next breath bless the founder of the nonjurors for 

1 E.g., cf. 11. 124-5 of the ode to Bancroft and L 28 of the ode to Temple. 


barely distinguishable conduct. Through a diffuse and vague 
employment of lofty epithets, Swift tries to blur the inconsisten- 
cies which he has created; but the longer the argument lasted, 
the deeper they were bound to sink. After a formal, convoluted 
introduction, however, he does speak briefly in his own voice, the 
effusion of an outraged prophet denouncing a poisoned age : 
'Each line shall stab, shall blast, like daggers and like fire.' The 
poet's tone of invective was never to sound more violent; irony 
and humour were never to be more remote. Perhaps the source 
of the fury was exasperation at a poetical impasse. 

Chapter Four 

Further motives for Swift's becoming a priest begin to appear 
in his poem to William Congreve. In a letter of December 
1693., evidently a long time after Swift originally decided to 
compose such verses, he described the finished work to his cousin 
Thomas and offered to submit it for criticism : "They are almost 
two hundred and fifty lines not Pindaric. 31 The poem was in- 
tended to go with any play by Congreve, and Swift hoped it 
would accompany the current one. The Double-Dealer, if that 
should prosper on the stage before going into print. 

While Swift was only two years older than Congreve and had 
been to the same school and university, there is no other record of 
their intimacy before this poem. Now almost twenty-four, Con- 
greve had been in England since 1689. After a spring and sum- 
mer in Staffordshire, he had come up to London, where many of 
his relatives lived. In the spring of 1691 he began to study law in 
the Middle Temple, at a time when its routine "was not one to 
disturb the pursuit of belles-lettres 9 . 2 At the inns of court, Ether- 
ege, Wycherley, and Shadwell had recently preceded Congreve, 
and he himself made little progress toward the bar. Conveni- 
ently situated for the Theatre Royal in Drury Lane and for Will's 
Coffeehouse near Covent Garden, the Middle Temple gave a 
student easy access to the world of the wits. Congreve's short 
novel, Incognita, appeared early in 1692, and he was one of Dry- 
den's collaborators on the Juvenal andPersius of October the same 
year. When Dryden included several other poems and transla- 

1 Ball i. 368. 

2 Hodges, p, 35. My information on Congreve normally comes from Hodges. 


tions by Congreve in Examen Poeticum, 16935 he singled out the 
young beginner as e more capable than any man I know* of trans- 
lating Homer. 'I am Mr. Congreve's true lover% he wrote to 
Jacob Tonson, and 6 [I] hope I shall never loose his affection. 31 
With this great patron's assistance, Congreve prepared his first 
play, The OldBatchelour, for the stage. Produced in January 1693, 
it had what was then the astonishing run of fourteen days; as a 
book it went into a third printing before April; and most con- 
noisseurs clothed the young author in Dryden's mantle. 

The Double-Dealer, which followed in October, had not the 
same reception. While judges of taste, led by Dryden, recognized 
the superiority of the new comedy over the earlier, the run of 
theatre-goers were displeased, perhaps because the satire was too 
harsh. In the dedication to the first edition (December 1693), 
Congreve put his 'illiterate criticks' in their place : 

The ignorance and malice of the greater part of the audience is such, 
that they would make a man turn herauld to his own play, and 
blazon every character. . . Some little snarling and barking there 
has been, but I don't know one well-mouth'd curr that has opened 
at all. 2 

This imprudent riposte was omitted from later editions. How- 
ever, Swift, though writing his poem before he knew how the play 
was attacked or defended, hit upon the same theme : 

What northern hive pour'd out these foes to wit? 
Whence came these Goths to overrun the pit? 

Though Swift adopted both the title and the heroic couplets of 
the complimentary poems prefacing The Old Batchelour, it is in 
fact he himself, and not either Congreve or the critics, that makes 
the underlying business of the poem. What spoils the shape of a 
panegyric is a covert struggle between Swift's eagerness to air 
his own feelings and his sense that the ostensible subject ought 
to be respected. Of the four irregular divisions into which the 

1 Dryden, Letters, cd. C. E. Ward (Durham, North Carolina, 1942), p. 59. 
a Works > ed. M. Summers, 1923, n. 9-10. 



whole poem falls, only the third is a convenient focus for both 

The beginning is a scene between the muse and Swift. She 
scolds him for treating her like a cast mistress when he sends 
her to praise Congreve. The present inspiration nevertheless be- 
speaks her divinity, since only a goddess could bridge the gulf 
between the dramatist and himself; and only Congreve's talent 
could extract such a tribute from the insulted spirit. 'Godlike'., 
therefore, is 

the force of my young Congreve's bays, 
Softening the muse's thunder into praise. 

The figure and the ironies are anticipations of Swift's later style. 
By packing the conceits too closely, however, and changing his 
point of view faster than the reader can follow, he makes the 
composition dark and awkward. Swift half-boasted to his cousin, 
C I cannot write anything easy to be understood though it were 
but in the praise of an old shoe' 1 ; and here is sufficient proof. 

In the next part (lines 49-108), he takes up the mob of bad 
poets and critics, who are separated from him as sharply as he 
from Congreve (though this parallel may be unconscious) . The 
playwright's genius excites Swift's mercy toward them somewhat 
as the muse had established a link between Congreve and Swift. 
Two angles of attack are then used : first, that they pick up bits of 
Congreve's wit to supply their own defects; then that they find 
fault with Congreve in order to hide their own inferiority : the 
critic is thus a poet manque. Who are the calumniators? asks 
Swift. Not London fops, as one might expect, but barely literate 
clods from the provinces, anxious to seem sophisticated. 

Illustrating what he means, Swift tells, in his third part, of a 
Farnham boy who went from school to London and soon came 
back with all the manners of the town, showing off his familiarity 
with Dryden, Wycherley, and Congreve. At this point Swift is 
free not only to indulge a private discontent but at the same time 
to express admiration for a friend. As he describes and then flays 

1 Ball i. 366. 



the boy, one cannot help suspecting that his irritation is stronger 
than the occasion calls for. The fury with the callow beardling for 
pretending to know men who don't know him may be genuine. 
Aggravating it, though, there must be almost a resentment 
against Congreve for seeing more of his new great friends than of 
his schoolmate, Swift. Still deeper, most readers would assume, 
is an involuntary comparison between his own position so in- 
commensurate with his ambitions and his felt gifts and that of 
the younger Congreve. The bitterness, invidious or not, which 
seasons the poem, is more likely to come from these instincts than 
from fury over a stranger's boastfulness. 

In the sixty lines remaining, Swift's tone is more evasive. On 
the surface he seems to tell how much luckier he is than Congreve. 
At Moor Park, in sight of Mother Ludwell's cave 1 and the stream 
which springs from it, Swift needs please only his Apollo, Sir Wil- 
liam Temple. Congreve, in courting the multitude, exposes him- 
self to arrogance and slander. This sentiment seems hollow. 
'What's that . . . if mankind be a fool ?' Swift asks. If it is nothing, 
the reader must wonder, why the fury on Swift's part ? The praise 
of the Temple household as a bard's haven is neither intense nor 
concrete : 

Happy beyond a private muse's fate, 
In pleasing all that's good among the great , 
Where tho* her elder sisters crowding throng, 
She [i.e.. Swift's muse] still is welcome with her inn* cent song. 

Swift's following image of a country muse insulted by beaux, 
collides with the preceding story of the boor who went to town. 
In describing the suave aesthetes, moreover, Swift incongru- 
ously uses rustic metaphors : 'cattle . . . odious smell . . . offensive 
herd'. The four last lines are an unfortunate reference to the 
vision which persuaded St Peter to bring the gospel to the geix- 
tiles (Acts x. n). Either Swift is unwarily striking a false note, 

1 G. G. Moore Smith would like to attribute a poem on this subject to Swift, but 
his argument, even when enlarged by Middleton Murry and Joseph Horrell, is 
to me unconvincing; see his edition of Temple's Early Essays, 1930, pp. xxvi- 
xxviii, 1 86-8, 206-7; also Z^wu m. 1068-9. 



since he has just advised Congreve to despise the many; or he is 
again stating his hope that 'Congreve will reform the stage 9 , i.e., 
preach to gentiles. 

Swift, one infers, would cheerfully sustain the toils facing his 
friend; at least, more experience of the perils which he describes 
would give force to his contempt for them. Nevertheless, there is 
marked individuality in the poem. Gowley's influence is hardly 
absent, 1 but Swift also quotes from one of his own poems, now 
lost. 2 Some strokes, indeed, are as effective as a phrase from the 
pamphlets of his prime: 

My hate, whose lash just heaven has long decreed 
Shall on a day make sin and folly bleed. 3 

It seems that the feelings and topics toward which Swift turned in 
his first flights were to become in later years the staple not of his 
poetry but of his prose. 


With the poem to Congreve, Swift's versification undergoes its 
first important structural change in the direction of his eventual 
style: this change is the replacement of pindarics by couplets. 
While the metre is still pentameter and the rhythms convention- 
al, Swift can begin to feel what his native gifts amount to. Al- 
ready the rhymes, though not obtrusive, betray some of the in- 
genuity and the preoccupation with sound effects which were to 
mark his best-known manner : e.g., Svit counterfeit* (lines 556), 
'since impertinence' (lines 878). In tone a profound change is 
the almost (but not quite) wholehearted acceptance by Swift of 
the office of satirist; he refers to his 'old unvanquish'd pride' as 
what alone 

suspends poor mortals fate, 
Gets between them and my resentments weight^ 
Stands in the gap 'twixt me and wretched men, 
T* avert itt impending judgments of my pen. 

1 Cf. 11. 63-4 here and U. 15-16 of Gowley's 'The Soule* (The Mistress, poem 15). 
8 LI. 205-12. 

8 Cf., from The Publick Spirit of the Whigs, *I will upon occasion, strip some of his 
insinuations . . . and drag them into the light* (Davis vm. 39). 


Rhetorically, there is an advance in that the muse draws apart 
from the poet, and either scolds or flirts with him. A wholly fresh 
element is the use of a fable (lines 1 15-46) and its exegesis (lines 
147-74) to illustrate the central theme of the poem; the elabor- 
ate employment of animals as satirical parallels to humans is a 
similar first appearance of what was to become a permanent re- 
source. (Temple also draws continual, conventional parallels be- 
tween kinds of men and species of animals; but he means to be 
philosophical or 'scientific' in this habit, while Swift's way is 
emblematical. 1 ) 

Nevertheless, the usual devices and motifs persist : mob versus 
hero, illusion versus reality, courtly vice versus country virtue, 
etc. There is imagery drawn from optics (lines 61-4), from 
Temple's writings, 2 and most indecorously from the Bible. 3 
There are digs at amateurs of the new experimental philosophy 
(lines 93-4, 207-12). The mob in this poem is of Congreve's 
critics, who are also swarms of gnats (to match the flies of the odes 
to the king and to the Royal Society), or else who have (like 
Louis XIV) sprung from dung. The illusion is false wit, which is 
opposed to Congreve's true. Swift identifies himself and his rnuse 
with pastoral solitude; the critics and epigones of Congreve, with 
the urban society : unfortunately, the position of the urbane play- 
wright himself is thus made equivocal. 

Again Swift has the advantage of dealing with a hero whom he 
knows directly. However, there seems to be some conflict be- 
tween the poet's conscious knowledge, his unconscious intentions, 
and his deliberate, rhetorical aims. Swift cannot apply as many 
superlatives to a junior school chum as to a fatherly king, patron, 
or archbishop. Surprisingly little of the poem therefore is devoted 
to direct eulogy. More than ever does the poet force his subject 
into the wings and manoeuvre himself upstage. Swift's own 
yearning for reputation takes over ; and since he can hardly claim 
that his own genius is superior to Congreve's, he praises instead 

1 E.g., Temple, Miscellanea 1. 61-2. 

a E.g., 11. 83-4 axe drawn from the Introduction to the History of England, pp. 44, 70, 
or from Miscellanea n. 224; cf. also the ode to the Athenian Society t 11. 298-9. 
8 Gf. 1. 32 and Luke xvi. 26; 1. 231 and Acts x. 1 1 . 



the moral elevation of his rural seclusion as contrasted with the 
corruption of the playwright's milieu; he even brings in the 
figure of Temple all that's good among the great* as guaran- 
tor of such elevation. The three figures clash grotesquely : the 
attributes which Swift praises in himself and Temple seem to 
negate the values represented by his friend. 

To give power to his stifled admiration for Congreve, Swift 
must exalt its price : so he bestows upon himself the highest moral 
standards, the most exacting canons of taste ; gone is the self- 
abasement of the earlier odes. Furthermore, to demonstrate his 
possession of such faculties, he must exercise them upon proper 
targets ; so he denounces his friend's critics, plagiarists, and boast- 
ed, though fraudulent, intimates. While the resulting occasions 
do not in fact enhance the compliments, they perform the more 
useful function of giving adequate employment to Swift's spleen. 
Although, moreover, the victims of Swift's attacks had generally 
been, like the topics of his praise, outside his common experience, 
in this they are men of his own generation and environment. His 
satirical details are therefore vivid, if extravagant. 

The muse has a full-bodied part in this poem; but as in the ode 
to the Athenian Society, she is made so womanly as to suggest 
that the poet is once more preoccupied with his own attitudes to- 
ward the opposite sex. This conjecture seems borne out by the 
wealth of clumsy similes drawn from courtship and seduction. 
Swift compares himself complimenting Congreve (instead of 
composing original works) with a rake sending to a comrade a 
cast mistress whom he has debauched. He compares fashionable, 
arty slang with bastards 'born between whores and fops'. He 
compares Congreve's name, bantered by pretended friends, with 
a 'fresh miss' whose favours c the meanest coxcomb' makes believe 
he has enjoyed. He compares the muse among the critics with 
'some bright country virgin' amid a chattering horde of beaux. 



The last of the early poems. Occasioned by Sir W T *s Late 

Illness and Recovery, is still incompetent; but it is the shortest and 
the best. Though this piece is a fairly close imitation of Cowley's 
The Complaint, it is based upon an idea from Temple's essay Of 
Poetry: 'True poetry being dead, an apparition of it walked 
about.' 1 The form is again heroic couplets, and some of the lines 
open a new vein of earnest, powerful conceits: 

Whether in time, deductions broken chain 
Meets ', and salutes her sister link again; 
Or hunted fancy 9 by a circling flight. 
Comes back with joy to its own seat at night; 
Or whether dead imagination's ghost 
Oft hovers where alive it haunted most; 
Or if thought's rolling globe her circle run, 
Turns up old objects to the soul her sun. 21 

The design of the poem is for Swift rather simple, since it em- 
ploys only two (though incompatible) arguments; and rhetoric- 
ally it achieves remarkably sharp focus. The main reason for the 
improvement in definition and coherence is that the muse here 
not only stands quite apart from the poet but also takes on a func- 
tion formerly assumed by an additional character ; for she is the 
villain of the melodrama : she is the evil vapour, the false meteor, 
the optical illusion. 

The poem is an odd sort of dialogue, made up of two speeches, 
each with its own framework. In the first half the muse appears 
and scolds Swift for not celebrating (in verse) Temple's return to 
health after a dangerous illness. Her speech rather ingeniously 
contains what might be called a poem-within-a-poem, or the true 
discussion of the avowed subject; for she recites the only lines 
directly relating to Temple's disease, which occupy no more than 
a fifth of the whole poem; and even these verses really describe 

1 Miscellanea n. 312; cf. also p. 313. 

a To see how this style, when transfigured by irony and humour, becomes the 
characteristic style of A Tale of a Tub, cf. the parallel passage in the Tale, p. 158; 
cfl also K. WilUams's contrast of Burton (pp. 15-16). 



neither the siege nor the recovery, but the fears of Lady Temple, 
Lady Giffard, and the inferior members of the household (lines 
3766). The muse reproaches the poet. He is sad when he ought 
to sing. The time of fear is past. The grief feelingly displayed by 
'Dorothea' and 'Dorinda' is no longer necessary. Temple has 
recovered. Swift should pile blessings on the occasion. Thus far 
the first half. 

The poet's answer makes the rest of the poem. Rather than 
respond to her appeal, he confounds the muse. In his extraordin- 
ary reply he does not even mention Temple's illness ; instead, he 
attacks the goddess for encouraging him to hope for fame and es- 
teem where there was no chance of either. He lists the rules which 
she has urged him to use, and labels them madness. She has 
tricked him by pretending that virtue and talent will find their 
reward. A 'right woman 9 would waste no time upon one so 
doomed as he. Her advice has led to frustration; her hopes were 
delusive ; she is herself a 'walking vapour'. All his efforts to please 
(i.e., one assumes, to please Temple) have met with 'contempt 
where thou hast hop'd esteem'. Finally, he points out that she 
herself is only an illusion; by renouncing her, therefore, he de- 
stroys her : 

And since thy essence on my breath depends - 9 
Thus with a puff the whole delusion ends. 

Abandoning poetry, he therefore dismisses and annihilates the 

Here at last the figure of the muse has a significant relationship 
to the design of the poem. In fact, she foreshadows the brilliant 
effects which Swift was soon to achieve through the donning and 
doffing of masks or ironical poses. As one aspect of the poet's 
character, his literary ambitions, the muse confronts another, his 
fears of failure. By splitting himself in this way, Swift can pump 
his lines full of hot emotion without threatening the reader. In- 
stead of denouncing the Jacobites, the Puritans, the university 
dons, or the critics, he points his rage at strata of his own being. 
However unimpressive the result may seem as verbal poetry, it 



does possess the crackling intensity and integrity of feeling which 
the name of Swift has come to stand for, and which makes a deep- 
er 'poetry* than any aural polish or rational structure can supply 
without it. True 5 until irony is employed to resolve them, Swift's 
clashing attitudes seem incoherent; but the violence of their clash 
is realized by the poet and directly felt by the reader. 

Temple's figure merges with the muse. Her rules are undoubt- 
edly his at least, as Swift conceives them. The closing lines of 
Swift's accusations against her are ultimately addressed to his 
patron. As father, as inspirer, as censor, Temple had inevitably 
fallen short of the excessive hopes which Swift had placed in him. 
But the young idealist dare not admit either that the original 
standards were impossible or that his chosen hero has sunk below 
them; so he scourges himself for failing to please Temple. This is 
what Swift means when he describes the rules as c madness 9 ; since 
hope, though false, is the only benefit of delusion, nobody will 
persist in a madness which destroys hope : the impossibly high- 
minded code preached by the muse (i.e., by Temple) permits one 
to win greatness only through such means as destroy the rewards 
of greatness. 

This poem is almost magical in the degree to which it fits its 
place as Swift's farewell to a long misconception. Biographically 
and structurally, it embodies just the features one would seek 
in such a performance. The paradoxical reasoning behind it, 
the little drama in which the paradox is acted out, the feverish 
changes of tone, the foreshortened ending, the dissatisfaction 
and restlessness openly admitted, all join to make it a remark- 
ably appropriate tailpiece to a series of false starts. 

Chapter Five 

Behind the poem on Temple's illness lay a situation which 
can be reconstructed with moderate accuracy. Ten years 
after his master's death. Swift wrote to Lady GifFard, 6 I 
pretend not to have had the least share in Sir William Temple's 
confidence above his relatives or his commonest friends ' and 
added bitterly, 6 I have but too good reason to think otherwise.' 1 
The case is put differently, more than fifteen years still later, in 
a sarcastic message to Temple's nephew, with whom Swift had 
picked a quarrel : 

I own myself indebted to Sir William Temple, for recommending 
me to the late king, although without success, and for his choice of 
me to take care of his posthumous writings. But I hope you will not 
charge my living in his family as an obligation, for I was educated 
to little purpose, if I retired to his house, on any other motives than 
the benefit of his conversation and advice, and the opportunity of 
pursuing my studies. For, being born to no fortune, I was at his 
death as far to seek as ever, and perhaps you will allow that I was of 
some use to him. 2 

Swift's reasoning may seem dark to a modern reader. What 
he means to acquit himself of is not ingratitude for favours re- 
ceived, but self-seeking motives in cultivating Temple's patron- 
age. As he saw it, the relationship was of mutual benefit. The im- 
mediate advantages to a young man of learning political wis- 
dom, morality, and literary art from a master of those sciences, 
and of living elegantly at the same time were balanced by the 
secretarial duties which Swift must have fulfilled scrupulously. 

1 Ball i. 172. 2 Ibid. m. 301. 



For 'gratitude' as resulting from special indebtedness, there was 
no place in the relationship ; each party received his due and dis- 
charged his functions. And no one could, in fact, argue that 
Temple 'deserved' more than he "got" from Swift, in labour or in 

But this was after the event. In Temple's lifetime, it must have 
been obvious that the master so enjoyed the pupil's company as 
to tilt the balance of obligation away from Swift, and not to- 
ward him. The protege could hardly have refrained from looking 
hopefully forward to (not perhaps depending upon, or expecting 
as his due) some particular kindness from his employer, a per- 
haps supererogatory but direct assistance toward a career, fitted 
to Swift's visible talents, and of an eminence proportioned to his 
ambition. If we think of Hobbes and Cavendish, or Locke and 
Shaftesbury, we shall have an idea of what could happen. 

An incident which could hardly have weakened such hopes is 
also an example of what Swift meant when he wrote that 'grow- 
ing into some confidence' with Temple after 1691, he 'was often 
trusted with matters of great importance'. 1 The king's faith in 
Temple had not evaporated when the diplomat refused to come 
out of retirement. Even while the international crisis spread, and 
the English were more and more deeply engulfed in the war of 
the League of Augsburg, William found occasions to be his old 
friend's guest, and (according to Swift) 'took his advice in affairs 
of greatest consequence'. 2 

During the reign of William and Mary an issue which had 
every urgency except novelty was the tug-of-war between court 
and parliament. After securing his throne and pacifying his three 
kingdoms, William had gone forward with the campaigns against 
Louis XIV. But the staggering defeat at Beachy Head and the 
surrender of Mons to Louis did not enhance his authority. To 
keep parliament with him, he allowed his prerogative to be hem- 
med in, though he considered no government so futile as that of 
a monarch without independent power. 3 The extreme Whigs, 

1 Atitob., t 8 V - * Ibid., ff. 

3 G. N. Clark, The Later Stuarts (Oxford 1934.), p. 143. 


who were keenest against France, were least willing to enlarge 
the king's freedom of action. 

Dependent for supplies upon grants of money, William had 
no way to avoid summoning parliament annually; and he laid 
down a tradition which has lasted ever since. What he did resist 
was the revival of Charles I's Triennial Act, which required the 
summoning of a new House of Commons at least once every three 
years. As Burnet remarks, Whiggish men felt afraid that in a 
house of unlimited life the members 

might be so practised upon by the court, that they might give all the 
money and all the liberties of England up, when they were to have a 
large share of the money, and were to be made the instruments of 
tyranny; as it was in king Charles the second's time. 1 

While a bill with this object had been unsuccessful in 1689, a 
second attempt passed both houses in 1693, when the king's pres- 
tige was reeling under a load of military disasters ; for the victory 
of La Hogue seemed to be obliterated by the loss of Namur and 
the massacre at Steenkirk. 2 

First read in the House of Lords on 12 January, the 'triennial 5 
bill hung fire for two months. 3 'The king let the bill lie for some 
time on the table ; so that men's eyes and expectations were much 
fixed on the issue of it.' 4 He felt so sharply worried about its pos- 
sible effect that he had the Earl of Portland, his favourite, take 
the opinion of Temple. The latter, to make sure that his argu- 
ments in support of the bill were fairly set forth, had Swift present 
them at Kensington, early in 1 693, when he was only twenty-five. 5 
The young secretary carried a written account from Temple and 
added reasons of his own; he spoke briefly to the king, and at 
length to Portland. 'But in conclusion', says Burnet, the king 're- 
fused to pass it ; so the session ended in ill humour'. 6 On 1 4 March 

1 Burnet iv. 185. a Ogg n. 3852-4. 

3 JJJ.L., xv. 181. The concurrence of the Commons was desired 32 Jan. (J.H.C., 
x. 786;, and the bill was read for the first time in the lower house on 28 Jan. 

4 Burnet rv. 187. 

5 Not twenty-one or twenty-three, as Swift said (Autob., f, gv). rv. 187. 


the bill was vetoed and parliament was prorogued. 1 Two months 
later, William arrived in Belgium to begin another miserable 
campaign; for this was the year of the battle at Landen e even 
more bloody than Aughrim' where the young Duke of Or- 
monde distinguished himself. 2 

Swift, forty-five years later, does not conceal his still-rankling 
annoyance : 

This was the first time that Mr Swift had ever any converse with 
courts, and he told his friends it was the first incident that helped to 
cure him of vanity. 3 

Yet he would have been mean-spirited indeed if the reliance upon 
him which Temple showed in this business did not raise expecta- 
tions of a full step toward making him a career. The errand had 
been precisely the sort on which Temple had been accustomed 
to send his son. No offer was, however, forthcoming. Yet as the 
months passed, Swift saw more and more of his friends settled: 
Congreve the hope of English letters ; cousin Thomas about to 
be ordained and to take up (in 1695) t ^ ie rectorship of Putten- 
ham, Surrey; Peter Browne already a fellow of Trinity College; 
Henry Tenison in the Middle Temple since 1692, and shortly to 
enter upon his Irish parliamentary service, which would com- 
mence in 1695. 

Reaching this period in his autobiography, Swift says that he 
was 'inclined to take orders'. 4 To his uncle William he wrote at 
the time (November 1692), e l am not to take orders till the king 
gives me a prebendary; and Sir William Temple, though he 
promises me the certainty of it, yet is less forward than I could 
wish, because I suppose, he believes I shall leave him, and upon 
some accounts, he thinks me a little necessary to him.' 5 

The complexity of the situation is barely intimated by Swift. 
There are several layers of conflict: first, whether to remain in 
the comfortable but temporary status at Moor Park, or choose a 
permanent way of life; second, if he should leave, whether to be- 

1 J.H.L., xv. 389; the bill came up again in 1694. and was enacted. 
2 Oggii. 385. *Autob. 9 f, 9 V . 'Ibid., 9, 8 Ball r. 10, 



come a priest, or enter another profession; third, if he should take 
orders, whether to do so immediately, and risk seeming mercen- 
ary, or wait until he was in a position really to select the priest- 
hood as preferable to other available openings. The last dilemma, 
while not Swift's alone, was characteristic of him. Of his respect 
for the priestly calling there can be no doubt; that his own inten- 
tion was sincere is obvious from the straightforwardness with 
which he mentions it. He never deceived himself, however, as to 
the likelihood of his therewith improving his worldly substance. 
So much can be inferred from Swift's words to his cousin Thomas, 
then in a predicament like his own but further knotted by an en- 
gagement to be married. Thomas felt properly hesitant; he had 
no income, and he did not like marrying without one : yet marry 
he would. 'All I can say*, Swift answered tartly, 'is I wish to God 
you were well provided for, though it were with a good living 
in the church.' 1 When Thomas communicated further qualms, 
Swift saw only one path for a man who, though he might prefer 
other trades to that of cleric, could be sure of a place in the 
church : 

I cannot . . . give a judgement near enough upon your other hopes, 
but if they be not certain, I think there is no avoiding the choice of 
what is. . . [Yet] if that curacy were not disposed of which I once 
mentioned you, I think I should say it was, for it fits your present 
prospects almost as ill as it did your merit then. 2 

He was realistic about it all. Though piety was the first ingredient 
of a priest, it could not support sublunary life. And if this was 
Swift's counsel to another, he surely reckoned with the obstacles 

The conscience beneath his conflict, however, appears in one 
extraordinary phrase that he e had a scruple of entring into the 
church meerly for support'. 8 Could a gate be straiter? In these 
refined scruples. Swift shows again his leaning toward what St 
Jerome described when he wrote that *many, who screen from all 
men's sight their poverty, charity, and fasting, desire to excite 

x Ball L, 365. * Ibid., p. 367. B Autob., f. 9. 



admiration by their very disdain of it.' 1 Swift had to give up the 
Temple establishment because it led nowhere. Having no estate, 
he could not go without an employment before him. His strongest 
patron. Temple, would arrange nothing respectable and wished 
to keep him at Moor Park. The church, his natural terminus, 
would be profaned if he went to it seemingly for lack of an alterna- 

Not that Swift's compunctions were unique : they seem paral- 
lel, for example, to those felt by Jane Austen's Edmund, in 
Mansfield Park, more than a century later though the turn of 
mind there is less fastidious : 

I see no reason why a man should make a worse clergyman for 
knowing that he will have a competence early in life. [His lady 
friend demurs; so he goes on :] But the motives of a man who takes 
orders with the certainty of preferment, may be fairly suspected, 
you think? To be justified in your eyes, he must do it in the most 
complete uncertainty of any provision. [No, she bursts out, to do so 
would be madness. He then concludes:] Shall I ask you how the 
church is to be filled, if a man is neither to take orders with a living, 
nor without? 2 

Swift was a more eager casuist than Edmund, but he managed 
to kill his sphinx. He extracted from Temple the offer of a post 
easily in his gift, c an employ of about ioo u a year* in the office 
of the master of the rolls in Ireland (a sinecure which Temple 

Whereupon Mr Swift told him, that since he had now an oppor- 
tunity of living without being driven into the church for a mainten- 
ance, he was resolved to go to Ireland and take holy orders. 3 

Free to exercise his franchise, Swift refused the clerkship, left 
Moor Park in the early part of May 1694, and joined his mother 
in Leicester. The closing scene must have been dramatic; a 
month later, he described it to his cousin Deane : 

1 To Eusiochium, par. 27. 

a Condensed from R. W. Chapman's edition (Oxford 1934), p. 109. 
8 Autob., f, 9. 
L [147] 


He was extremely angry I left him; and yet would not oblige him- 
self any further than upon my good behaviour, nor would promise 
any thing firmly to me at all, so that everybody judged I did best to 
leave him. I design to be ordained September next, and make what 
endeavours I can for something in the church. 1 

He was clearly without a concrete opportunity, for he went on to 
say that he would gladly become chaplain to the English traders' 
settlement in Lisbon if either Deane or Willoughby Swift, then 
merchants in Portugal (Deane had gone there from the Kil- 
kenny School in August 1688), possessed 'interest to bring me 
in'. 2 Such a proposal, conceived at the height of a vast military 
campaign, a year after the loss of the Smyrna convoy, and the 
same month as the disaster of Camaret Bay, is a mark of how far 
the bitterest war could at this time seem from the destinies of pri- 
vate citizens. More pertinently, it also indicates that Swift never 
anticipated the delays which ordination itself might require. 

Temple's side in this business should seem the effect of neither 
whimsy nor spite. If he had indeed any fatherly feelings for Swift, 
they would have worked in the wrong direction. Not only would 
he have wished to hold on to the young man's company, but the 
memory of his son's fate would have made protective measures 
appear kinder than efforts to give Swift independence. So long as 
John Temple had lived under his father's loving authority, he 
had been safe. He might carry messages from Temple to either 
Charles II or James II, but he was not allowed to meet the Prince 
of Orange at Torbay. When John Temple took a post of his own, 
though given him for his father's sake, the result was the most 
painful tragedy of Temple's life. Swift too might carry messages 
to court and lighten his master's solitude; but one could hardly 
expect Temple to play Daedalus a second time, and to Swift's 

On the level of pure selfishness, Temple could not have hoped 
to secure another secretary so talented as Swift. Here were an, ex- 
cellent listener, a witty speaker, a good book-keeper, an admiring 
disciple, and a humble subordinate, all ready to serve him at a 
i Ball i. 12. *Ibid. 

[I 4 8] 


gesture. Swift could make fair copies of his essays, translate his 
French and Latin letters, explain his wishes to menials or visitors, 
keep his accounts, provide society for his wife and sister, manage 
the household when the family were away all at a negligible 
expense and even with some appearance of benevolence on 
Temple's part. 

In Temple 9 s own past there were few experiences which might 
give him an insight into Swift's condition. Any diplomat spends 
most of his career waiting for others to make up their minds. But 
after the nightmare impasses at Nimeguen, the longest suspense 
which Swift might endure would sink below notice. Apart from 
his courtship of Dorothy Osborne, Temple had been the most 
obedient son possible by his political philosophy, fatherhood 
and the very notion of authority were merely two different views 
of the same object. He could not have felt that his demands upon 
Swift's confidence were unfair. 

Furthermore, it was not in Temple's nature to be a strong 
pleader, whether for himself or for others. It is even possible that 
his proteg^ caught from him the disdain which Swift was often to 
show for the normal methods of rowing one's own boat. It might 
indeed be no imprudence that Temple, heir to a considerable for- 
tune, should brush past occasions for enriching himself and say, 
*I have resolved never to ask [the king] any thing, otherwise than 
by serving him well', 1 or C I never could go to service for nothing 
but wages' 2 ; but it would seem either naive or arrogant (and all 
the more admirable) that a man so ill provided for as Swift should 
suppose he owed it to his integrity to act a similar role. Mean- 
while, if Temple were meaning to impose his code of honour up- 
on Swift, he would certainly not meet the young man's demands. 

Nor is there much reason to suppose that Temple regarded the 
priesthood as the ideal career for Swift or anyone else. With his 
cool piety, his resentment of the church's material strength, and 
his leanings toward a Dutch toleration, he would probably have 
hoped that Swift would interest himself in a more dashing or 
more public occupation. 

1 Letters u. 289; cf. i. 301-2. * Memoirs m. 170. 


Chapter Six 

Although one cannot describe Swift's becoming a priest as 
f-\ inevitable, it should appear quite natural. From the cir- 
JL A. cumstances of his infancy, he grew into a lofty idealist. 
His intense moral anger, a frequent and noble heritage of men 
who have endured severe discipline in childhood, could have 
found no better vent than a pulpit. His appreciation of a father's 
employment must have made the paternal privileges of a clergy- 
man seem deeply attractive. The tradition of his family was heavy 
with clerical ties (both grandfathers and two uncles Thomas, 
namesakes of the Goodrich hero). Cousin Thomas, his nearest 
thing to an older brother, and perhaps the most steadying ele- 
ment of his school and college days, was preparing for the church. 
Until the end of the eighteenth century a young man with Swift's 
education and attachments would have seemed fated to take up 
the vocation. One can only salute the integrity with which he de- 
ferred the obvious until it should possess the purity of a free de- 
cision. In his power of recognizing the tough, nasty facts of a case 
at the same time as he chose the high-minded, rather than Realis- 
tic', resolution of it, Swift reveals his finest (and most poetic) trait. 
By the canons of the church, he might not be ordained either 
before he was twenty-three or until he could certify to having c a 
living in readiness*. 1 Although these conditions might sometimes 
be dispensed with, they would in the normal course have kept 
him a layman until the end of 1690. In February 1692 he said he 
was thinking e of entering into the church*. 2 The Oxford M.A. 
that summer was probably a step on the way to ordination. In 

1 Landa, pp. 3-4. Ball i. 5. 



November 1692 he told his uncle William that Temple was hav- 
ing the king provide him with a prebend. 1 When, in June 1694, 
he said he intended to be ordained 'September next', 2 he was 
thinking no doubt of the embertides of September, one of the 
four periods set by the canons for the admission of candidates to 
orders. 3 

From Leicester, then, he went to Dublin, where he worked 
through the summer to move favourable powers in his own be- 
half. Several bishops were familiar with his uncles ; and Narcissus 
Marsh, recently provost of Trinity College, was now Archbishop 
of Dublin. Nevertheless, Swift found that holy orders were closed 
to him, because the passage of so many years since his taking a 
university degree aroused suspicious curiosity about what he had 
been doing in the meantime. It was a canonical requirement that 
one should present testimonials of 'good life and behaviour' for a 
period of three years preceding ordination; and although the 
application of this rule had been notoriously lax, Swift had struck 
an era and a regime of tightening standards. Unless he could pro- 
duce a certificate of good conduct, he was blocked. 

To the delicate and humbling task of winning an endorsement 
from Temple he at last nerved himself, in a letter of October 
1694, the only direct communication preserved between the two 
men. The conclusion of his appeal must stir the sympathies of 
anyone who has had to beg a benefit from an already resentful 
superior; Swift was not to use such language again until Gulliver 
left the land of the Houyhnhnms. Nevertheless, if the explanation 
we have given of their relationship is reliable, Swift neither for- 
feits his dignity here, nor sinks to self-abasement: 

May it please your honour, 

That I might not continue by any means the many troubles I 
have given you, I have all this while avoided one, which I fear 
proves necessary at last. I have taken all due methods to be ordain- 
ed, and one time of ordination [i.e., the embertides of September] 
is already elapsed since my arrival without effecting it. Two or three 
bishops, acquaintances of our family, have signified to me and 

1 Ball i. 10. a Ibid., p. 12. 3 Landa, p. 6. 


them, that after so long a standing in the university, it is admired I 
have not entered upon something or other, above half the clergy in 
this town being my juniors, and that it being so many years since I 
left this kingdom, they could not admit me to the ministry without 
some certificate of my behaviour where I lived ; and my Lord Arch- 
bishop of Dublin [i.e., Narcissus Marsh] was pleased to say a good 
deal of this kind to me yesterday, concluding against all I had to 
answer, that he expected I should have a certificate from your hon- 
our of my conduct in your family. 

The sense I am in, how low I am fallen in your honour's thoughts, 
has denied me assurance enough to beg this favour, till I find it im- 
possible to avoid; and I entreat your honour to understand, that no 
person is admitted to a living here, without some knowledge of his 
abilities for it, which it being reckoned impossible to judge in those 
who are not ordained, the usual method is to admit men first to 
some small reader's place till by preaching upon occasions, they can 
value themselves for better preferment. This (without great friends) 
is so general, that if I were four-score years old I must go the same 
way, and should at that age be told, every one must have a be- 
ginning. I entreat that your honour will consider this, and will 
please to send me some certificate of my behaviour during almost 
three years in your family; wherein I shall stand in need of all your 
goodness to excuse my many weaknesses and follies and oversights, 
much more to say any thing to my advantage. The particulars ex- 
pected of me are what relate to morals and learning, and the reasons 
of quitting your honour's family, that is, whether the last was oc- 
casioned by any ill actions. They are all entirely left to your hon- 
our's mercy, though in the first I think I cannot reproach myself any 
farther than for infirmities. 

This is all I dare beg at present from your honour, under circum- 
stances of life not worth your regard. What is left me to wish, next 
to the health and felicity of your honour and family, is, that Heaven 
would one day allow me the opportunity to leave my acknow- 
ledgments at your foot for so many favours I have received, which, 
whatever effect they have had upon my fortune, shall never fail to 
have the greatest upon my mind, in approving myself, upon all 

Your honour's most obedient and most dutiful servant, 

J. Swift. 

I beg my most humble duty and service be presented to my ladies, 
your honour's lady and sister. The ordination is appointed by the 


Archbishop by the beginning of November, so that, if your honour 
will not grant this favour immediately, I fear it will come too late. 1 

Temple was above spite, and he sent the statement so quickly that 
Swift could be ordained deacon 528 October and priest 13 Janu- 
ary 1694/5. Both ceremonies were performed in Dublin, at 
Christ Church Cathedral, by William Moreton, Bishop of Kil- 
dare (Moreton held the deanery of Christ Church in commendarri) , 2 

Swift had now resigned himself to profit in his own country 
rather than honour in England. But he had also to interest some 
large-handed, powerful person in his needs, for merit was no 
guarantee of a benefice. The name of the philanthropist is still 
unknown, but somebody recommended Swift to Lord Capel, 
chief of the three lords justices then ruling Ireland for the court. 3 
This action was so effectual that only two weeks after his final 
ordination the new priest was presented (by the crown) to his 
first living, the prebend of Kilroot in the diocese of Down and 
Connor. 4 Of course, the speed was to be expected, with the pro- 
mise of a benefice being a prerequisite for admission to orders. 

This lord justice, Henry, Lord Capel of Tewkesbury, was the 
younger brother of that Earl of Essex who had been so treacher- 
ous a colleague of Temple's under Charles II. He had been a 
privy councillor with Temple, and his residence at Kew was al- 
most next to Temple's at Sheen. CapeFs father, the heroic Lord 
Capel of Hadham, had devoted a life of c courage, virtue, and 
fidelity' (as Clarendon said) to the royalist cause; the Puritans 
beheaded him in 1649, soon after the king. The son, shortly to be 
sole governor of Ireland with the title of Lord Deputy, 5 was an 
immoderate 'Whig* (as the word was used at the turn of the cen- 

1 Ball i. 13-15. * Landa, pp. 7-8. 

a Deane Swift (p. 106; identifies Swift's recommender as Temple; but he is 
practically excluded by the language, at this point, of Swift's autobiographical 
essay. Landa (pp. 9-10) suggests Ashe, who is indeed the most likely person. How- 
ever, it is safest to leave the issue unresolved : e.g, Richard Tenison, Bishop of Meath, 
was the father of Swift's friend, had been chaplain to Capel's brother, and was one 
of Swift's predecessors at Laracor ; he too might have recommended Swift to Capel. 

4 He was presented to the prebend of Kilroot by patent dated 28 Jan. 1695; he 
was instituted 5 March and installed 15 March (Cotton). 

6 27 May 1695; he died a year later. 



tury) , steadily placing Protestant English welfare before Catholic 
Irish,, 1 and pressing the Nonconformists 5 claims to legal tolera- 
tion. 2 Dartmouth called him c a very weak, formal, conceited 
man' ; but Shrewsbury said he was 'liked and beloved by all par- 
ties*. 3 Expressly proud of his father's glorious reputation, Capel 
did not apparently find Swift's affiliations distasteful; and 
CapePs connection with Ireland had some depth, his brother 
having been lord lieutenant about twenty years earlier. It is clear 
of course that Swift outwardly had not yet assumed a political 
colour, and simply accepted the powers that be. However, if he 
had been labelled a fellow-traveller of the nonjurors, or a critic 
of the king's measures in Ireland, Capel would not have been 
selected as the man to prefer him. 


The Church of Ireland which is of course the name not of 
the Roman Catholic Church but of the Anglican or Established 
Church in Ireland was to be the scene of Swift's entire eccle- 
siastical career. Twenty years of manoeuvring would bring him no 
bishopric, deanery, or prebend in England. Not that Swift wholly 
mistook his chance. The drive against him had begun before he 
was born. An English university man commonly had precedence 
over a rival from Ireland when benefices were assigned in the in- 
ferior kingdom; but a clergyman of Irish breeding and educa- 
tion though of purely English descent was in the natural 
course shut out not only from every good employment in the 
mother country but also from the major dignities of his native 
land. Yet a historical irony arises just here where none is want- 
ing; for out of the three Trinity College students who were trans- 
lated to the English episcopal bench in the eighteenth century, 
two were classmates of Swift Thomas Wilson and Edward 

The Church of Ireland at this time was an ailing giant, asth- 
matic within and throttled from without. After the Reformation 
1 Burnet iv. 277. a Landa, p. 20. 3 Burnet iv. 278-9, n. 



it was almost a foreign, thing to the natives, for the proportion of 
Protestants among them was of the order of a fifth to a twentieth; 
and by 1695 most of the Protestants were dissenters from the Es- 
tablished Church. Having so few worshippers to employ them, 
the buildings fell away with disuse ; the wealth drifted into the 
wrong hands ; and the pastors failed in their sense of mission. 
Only c a face of Christianity' was saved. 1 

A typical diocese was Ferns, practically identical with the 
county of Wexford, about fifty miles long and twenty miles wide. 
It contained a hundred and thirty-one parishes, of which William 
King said (when he became Archbishop of Dublin), 

7 1 were in the hands of impropriate lay proprietors, 28 appropriat- 
ed to the bishop and cathedral dignitaries, and only 32, usually the 
poorest, in the hands of the officiating clergy. Neither bishop, dean 
nor archdeacon was resident in the diocese. It was served by only 
thirteen beneficed clergymen, few of whom had an income of as 
much as 100 a year, and nine curates, each paid about 30 per 
annum. 2 

The livings, from the most meagre curacy to the primacy of all 
Ireland, were tending to become sinecures for ecclesiastical poli- 
ticians and pamphleteers ; and the clergy were pressed to use di- 
vine service, property rights, literary gifts, and seats in the House 
of Lords as devices for strengthening the government in power. 
By the time Swift took orders, a parson who owned no political 
identity might get a good name but hardly great riches. 

To resist the defiance of Catholics and Presbyterians, and an 
English government's greed, the Church of Ireland had the sup- 
port of Anglican landlords the 'ascendancy' who wished to 
keep the nation under their own control. But between them and 
their priests stood one bitter grudge : the payment of tithes. By 
paying less than the true value of the tithes, by converting land 
from tillage to pasture (for which the tithes were less), by paying 
an agreed-on but permanently set sum (rather than a proportion 
of the harvest, etc.), by taking long leases of episcopal lands at 

1 Davis ix. 58. * As summarized in Phillips in. 182. 


low rents, or by simply refusing to pay tithes at all (and engaging 
tlie clergy in expensive, often disastrous litigation) , the landlords 
tried to reduce their obligations to their ministers. Yet the prin- 
ciple by which the same men laid claim to their vested political 
ascendancy was that they were Anglicans and not Presbyterians 
or Roman Catholics. As the church weakened, under the ham- 
mer of its other destroyers, it had to compromise more and more 
ruinously in this relationship. 

At the head of the Church of Ireland stood the Archbishop of 
Armagh, who was primate. Three other archbishops followed 
him in the hierarchy: Dublin, Cashel, and Tuam. Under each 
of these four were three to seven additional bishops, whose dio- 
ceses the archbishops had to oversee in addition to tending their 
own. The primate managed the diocese of Armagh, supervised 
the seven inferior prelates of his archbishopric, and was also head 
of affairs of the body of the church. Although the rest of the clergy 
might receive their appointments from a variety of sources, the 
bishops were always and deans usually chosen by the English 
government. Their incomes were seductive, their powers were 
broad (every bishop sat in the House of Lords), and the faithful 
wanted feeding. No administration could let the deanery of St 
Patrick's, valued at a thousand pounds a year, or the archbishop- 
ric of Armagh, abundant in patronage, slip into unco-operative 

Normally, a cathedral had a body of clergymen to carry on its 
business and services; they were the 'canons', and made up a 
'chapter', the head of which was called 'dean'. Certain of the 
properties belonging to the cathedral were usually assigned to in- 
dividual canons, and the right to use the income so arising was a 
'prebend*; the holder of it, a 'prebendary'. To most prebends 
were attached parochial benefices in the diocese, which the pre- 
bendary served as vicar or rector and from which the bulk of his 
income was derived. 

When Swift went to Kilroot, the primate was Michael Boyle, 
now closing a long life of eminence in the church. He was a de- 
cayed octogenarian, senilely absent-minded, very hard of hear- 


ing, and nearly blind 1 an appropriate emblem of the Anglican 
establishment in Ireland. Edward Walkington, Bishop of Down 
and Connor, presided over the diocese to which Swift's first par- 
ish belonged. Like Boyle, he was a native of Ireland, educated at 
Trinity College. Capel had advised his promotion as a man who 
would get along with the dissenters concentrated in the area; 
and he was popular, learned, well connected. The other virtues 
which recommended him to the lord deputy were moderation, 
sobriety, a talent (regularly exercised) for preaching, and a habit 
of residing at the scene of his duties. 2 


The parish which Swift served had since 1609 been a union of 
two vicarages and one rectory: Kilroot, Templecorran, and Bal-_ 
lynure. Together they formed the 'corps* of the prebend of Kil- 
root in the cathedral of Connor. The distinction between vicar- 
age and rectory was not at all one of spiritual duties or authority. 
In theory the priest of any parish church had originally been a 
rector, supported by tithes which his parishioners set aside for 
him, and enjoying during his tenure the use of certain buildings 
and lands. In the Middle Ages and at the Reformation these 
'rectorial' tithes or properties had often been handed over to 
somebody else 'impropriated*. But some resources had still to 
be held back for a Vicar', who possessed spiritually the same rank 
as a rector but received different usually more modest 'tem- 
poralities'. Logically, there could be a vicar only where there was 
no rector. 

Since the income of many livings was too little for the needs of 
even a parson's family, several parishes, adjoining or not, were 
frequently thrown together as a single 'cure'. The merger might 
be occasional and for the sake of one man, who was simply grant- 
ed from time to time extra benefices. His holdings were then a 
'plurality' ; and if they were assigned in commendam, he enjoyed 
the receipts so long as no regular incumbent was named. Michael 

1 Ware i. 130. a SPD, 1694-95, pp. 480, 512. 


Boyle, when Bishop of Cork, Cloyne, and Ross, had for a while 
taken the profits of six parishes to himself, 'under colour that he 
could not get clergymen to serve them*. 1 But his pluralism 
brought down a scolding from the government, and he filled the 
openings. Some benefices were attached to particular dignities 
ex qfficio, and whoever got one got the other : so the Bishop of Kil- 
dare, from 1681 to 1846, held in commendam the deanery of Christ 
Church Cathedral, Dublin. When several livings were perman- 
ently combined, and all their duties and temporalities were en- 
trusted to the same minister, they became a 'union', such as that 
of Kilroot. 

Swift was not entering a healthy and prosperous diocese. The 
prebendal stall of Kilroot was one of four (in the cathedral of 
Connor), created and endowed by James I (1609) during a cam- 
paign to strengthen the Church of Ireland. But the diocese of 
Connor did not flourish. Out of sixty-seven of its churches which 
were visited in 1622 only seven were found to be in good repair. 
As for county Antrim, where Kilroot lay, an inquisition of 1657 
revealed that 'in the 65 parishes 30 churches were in ruins, 27 had 
no incumbents, 51 had no glebes, and the tithes of 32 were im- 
propriate to laymen'. In 1693 another report on Down and Con- 
nor emphasized its lamentable condition 

the clergy wrongfully dispossessed of their glebe lands, incumbents 
restrained from claiming their legal rights for fear of offending their 
patrons or parishioners, non-residence widely prevalent, churches 
falling into disrepair and ruin. 2 

So many complaints had been made that an ecclesiastical com- 
mission was appointed to look into the diocese. The then bishop 
(1693), Thomas Hackett, c was facetiously known as the Bishop of 
Hammersmith [near London] because of his prolonged residence 
there' ; he found himself finally deprived. In the wake of the com- 
mission's work there were additional 

suspensions, excommunications, and deprivations for such varied 
1 Ware i. 569. a Landa, pp. i o-i i . 


offences as drunkenness, fornication, adultery, neglect of cures, 
pluralism, diversion of funds, excessive procuration and visitation 
fees, non-residence, illegal use of the bishop's seal, and simony. 1 

The stables were still being cleansed when Swift arrived from 
Dublin. During March and April 1695 he was installed in his 
prebend at Lisburn, went through the required forms for induc- 
tion at Templecorran and Ballynure, and at the end of April read 
divine service and preached at Lisburn. 2 His parishes were in the 
north-east of Ireland, on the shore of Belfast Lough. Even today 
it is a bleak and exposed neighbourhood, though with fine views 
of the hills behind the farmlands and of the edge of county Down 
across the bay. Templecorran and Kilroot together formed an 
uneven rectangle, with one short side of sea-front. Inland and 
due west lay Ballynure, cut off from them by St Nicholas's parish, 
which included the important fishing port and market town of 
Garrickfergus. There was no church at Kilroot, only a non- 
functioning ruin. Swift did not appear there to read divine ser- 
vice at the time of his induction, and he probably never preached 
there. 3 At Templecorran and Ballynure there were churches, less 
than eight miles apart the one at Ballynure seems to have been 
in poor repair but usable and his Kilroot parishioners may have 
worshipped at nearby Templecorran. 4 

Every parish was supposed to have some land or 'glebe 9 
and a house (manse or glebe house) for the priest's own use. It 
was a common thing, especially in unstable Ulster, for the house 
to be uninhabitable and the glebe impropriated. There is no sign 
that Swift found a manse in any of his parishes, though there may 
have been a glebe in one. 5 He may have lived either in the village 
of Templecorran or in Carrickfergus. From the town he could 
have ridden to each of his churches in less than an hour, unless the 
weather and the roads were bad. 

Neither had Swift (on a hundred pounds a year) an abundance 
of income to enjoy, 6 nor was there a glut of parishioners to over- 
work him. An accurate description of his territory, written on the 

1 Landa, p. 13. 2 Ibid. 9 p. 15. *Ibid., p. 22. * Ibid., p. 23. 
5 Ibid., p. 22. ' Ibid., p. 16. 



spot (1683)5 by Richard Dobbs, one of his parishioners-to-be, is 
still preserved : 

The parish of Kilroot is but small, the whole tithes not worth forty 
pounds per annum, and the great tithes belong to the Earl of Done- 
gal, the small tithes to the prebendary [i.e.. Swift's predecessor], 
one Milne, a Scotchman; the inhabitants (except my family and 
some half a dozen that live under me) all Presbyterians and Scotch, 
not one natural Irish in the parish, nor papist, and may afford i oo 
men. Next adjoining to this parish, adjacent to the sea, is ... the 
parish of Templecorran; the small tithes belong to the prebendary 
aforesaid, the great to the bishop, and may be worth fifty pounds 
per annum . . . the inhabitants all Scotch, not one Irishman or 
papist, all Presbyterians except the parson and dark, who I think 
is his son. 1 

For an Anglican clergyman in Ireland, one of the severest trials 
was that the government, which could not afford to alienate the 
powerful community of Nonconformists, actually gave encour- 
agement to them. 'King William not only renewed the regium 
donum to the Ulster Presbyterians; he increased it.' Capel tried 
to strengthen their legal toleration. The Act of Uniformity was 
not enforced. 2 

Not only was Ulster in general heavily Presbyterian,, but 
Swift's diocese, especially Antrim, was still more densely so than 
the rest of the province. 'The Nonconformists are much the most 
numerous portion of the Protestants in Ulster' , one bishop com- 

Some parishes have not ten, some not six, that come to [the Estab- 
lished Church], while the Presbyterian meetings are crowded with 
thousands covering all the fields. This is ordinary in the county of 
Antrim especially, which is the most populous of Scots of any in 

Even within Antrim, no parishes were more seriously infected 
than Swift's; and he himself said his sermons were 'calculated for 
a church without a company or a roof'. 4 Shortly before his ap- 

1 McSkimin, p. 378. a Laada, p. 20. 8 Ibid., pp. ao-i. * Ball i. 29. 



pointment, the dean of Connor had said that many cures in this 
part of the diocese were neglected, 'particularly the cure of Bally- 
nure and Killrott belonging to Mr Miln [is] not served by him- 
self or any curate 9 ; as a consequence, 'several considerable per- 
sons in the parrish of Ballynure particularly Mr Dobbs & his lady 
& Mrs Stewart & several others . . . were forced to frequent the 
Presbyterian meetings for want of a fitt minister to attend that 
cure 9 . The first Presbyterian minister in Ireland had been pre- 
sented to Templecorran and Kilroot. Swift's immediate pre- 
decessor had been a Commonwealth minister until, at the Res- 
toration, he decided to conform. 1 Belfast, only a few miles from 
Kilroot, was the centre of the strength and wealth of Presby- 

As a final sign of the weakness of Swift's outpost, there is the 
startling fact that his own bishop, Edward Walkington, believed 
the prebend was not rightly Swift's, for he had been (erroneously) 
persuaded that it still belonged to the previous incumbent. Wil- 
liam Milne had emigrated to Ireland from Scotland in 1657. He 
was ordained by the presbytery of Antrim and given a parish 
next to Kilroot. In 1660 he abandoned Presbyterianism, and in 
1 662 Jeremy Taylor reordained him ; a year later, he was pre- 
sented to Kilroot. 2 During the general investigation of the dio- 
cese by the ecclesiastical commission of 1693, Milne was charged 
with non-residence, intemperance, and incontinence. He was 
first suspended but later deprived, although a maintenance of 
twenty pounds a year was set aside for him out of the income of 
the prebend. 3 

Early in. 1697, Walkington said that Milne had never been de- 
prived of his prebend but merely suspended. Writing to Anthony 
Dopping, then Bishop of Meath, Walkington mistakenly argued, 

[Milne] was censured by your lordship here at Lisburn for some 
misdemeanours, and his penalty was a suspension from, his office 
and benefice during the kings pleasure, this censure was by the poor 
man himself and by most of the hearers mistaken for deprivation, 
and the mistake was so current, and universall that one Mr. 

1 Landa, p. 19. 2 Ibid., p. 12, n. 8 Ibid., p. 14. 


Jonathan Swift putt in for the poor man's living, and obtain'd the 
kings patent for it, during which time the poor man has been main- 
tain 3 d by the charitable contribution of the clergy or otherwise he 
must have starv'd. 1 

Further research must have shown that Milne had no case; and 
he was certainly not reinstated. Swift probably never heard of 
the bishop's doubts, being in England at the time; nevertheless, 
the proceedings of the commission were notorious ; and one of the 
witnesses against Milne had been John Winder, rector of Carn- 
money, who became Swift's friend and eventually his successor 
at Kilroot. It was impossible for Swift to avoid knowing the de- 
tails of Milne's scandalous conduct. Even if he had not just come 
from the civilization of Moor Park, even if he were not infused 
with the highest ethical ideals, even if he did not possess a both 
hereditary and inculcated repugnance for Puritanism, the cir- 
cumstances of his entry into the church must have shocked him. 
The geographical isolation of his parishes, the meanness of their 
church buildings and temporalities, the paucity of worshippers, 
the threatening crowds of the hostile sect, the profound corrup- 
tion of his own prebendal antecedents to be undisturbed by 
such a congeries of evil symptoms would seem a mark not of wis- 
dom but of vice. 


Yet he found some people who were worth knowing. In Carrick- 
fergus and Belfast princely homes were maintained by the Earl of 
Donegal, heir of the Chichesters. The rectory of Kilroot was of 
course impropriate to the Earl. Swift put himself on speaking 
terms with the Countess, 2 and was sooner or later introduced 
to her niece, Lady Jenny Forbes. 3 Within Kilroot parish the great 
house was Castle Dobbs^ belonging to Richard Dobbs, who was 
one of Swift's churchwardens at Templecorran and who had 

1 Landa, p. 1 2. 

2 Ball i. 1 8. The lines praising her, in Apollo's Edict, I believe to be not Swift's 
but Mary Barber's : see Poems i. 355-6. 

3 Ball H. 248-9. 



often been mayor of Carrickfergus. 1 He was now about sixty or 
more, loved to improve his house and his lands, and felt obvious 
pride in his family's long connection with the place. Since 1664 
he had been high sheriff of county Antrim. Swift paid visits to 
Dobbs, borrowed GlanvilPs Scepsis Scientifica from him, and lent 
him a volume of Temple's Miscellanea. The Glanvill he found, as 
one might have guessed, 'a fustian piece of abominable curious 
virtuoso stuff'. 2 

Carrickfergus was still the focal centre of the Lough, only start- 
ing to be eclipsed by Belfast. Carrickfergus Castle, thrusting out 
to sea on a rocky promontory, was among the best-preserved an- 
cient fortresses in Ireland. During the late Middle Ages it had been 
e the only firm holding of the English in Ulster'. Around it grew 
a flourishing maritime trade. On the pier under its walls, after 
the castle surrendered to Schomberg, William III had landed in 
1690, five years before Swift's arrival. On the edge of the town 
stood its most imposing residence, the Chichesters' many-win- 
dowed, many-bayed Joymount, built by Sir Arthur at the be- 
ginning of the century. The handsome church of St Nicholas (as 
yet unsteepled) dated from Norman times ; although the original 
structure fell into ruin, it was repaired, at Chichester's expense, 
while Joymount was being finished *a mean beggarly restora- 
tion', says a recent rector. The north aisle still ends with the 
elaborate Jacobean monument, alabaster and marble, bearing 
effigies of Sir Arthur and his family. 3 

In 1696 the mayor of Carrickfergus, elected for a second time, 
was Henry Clements, one of Swift's churchwardens at Bally- 
nure. 4 His family, like those of Dobbs and Donegal, had done 
what they could to resist James II during the Troubles; both he 
and his brother Edward were attainted by the Dublin parliament 
of 1689. Clements witnessed Swift's preaching and reading di- 
vine service at Ballynure in April 1695. Swift liked him and ac- 
cepted an invitation to his mayor's feast. 6 

1 McSHmin, pp. 417-19; Landa, p. 15, n. a. * Ball i. 28. 
3 See E. M. Jope, A Guide to Carrickfergus Castle (Belfast 1957), P- 3i J- C. 
Rutherford, St Nicholas* Church, Carrickfergus, 1957, p. [9], 

* Landa, p. 15, n. a; McSkimin, pp. 418-30. 5 Ball i. 29; Landa, p. 15, n. a. 
M [163] 


Most of the parishes next to Swift's were the corps of George 
Warter Story, who had succeeded Ward as dean of Connor. 
When Schomberg's army came over in 1689, Story came with 
them as a regimental chaplain. He served throughout the Wil- 
liamite campaigns, and wrote a history which became c by far 
the most important authority for the war on the Williamite side'. 
At the same time as his deanship (December 1694) he had re- 
ceived the rectory of St Nicholas. 1 

Of all the acquaintances which Swift formed at this time, the 
longest-lived was with John Winder, whose sea-front vicarage of 
Carnmoney lay between Ballynure and St Nicholas's parish. 
According to one report, he too had been an army chaplain, and 
was the son of a colonel. 2 The husband of one of his great grand- 
daughters sketched Winder as having c a moderate estate* in 
county Antrim and as forming 'several respectable connexions'. 3 
Yet what Swift says to him suggests a man not well provisioned 
with money or great friends. Although he was probably married 
at this time, his first child was born after Swift left Kilroot. Win- 
der's two sons became parsons ; his elder daughter, a parson's wife; 
but his younger daughter married one of the most eligible gentle- 
men in the county, with whom she had an only child eventually 
to be famous as a brilliant diplomat and created Earl Macartney. 4 

Swift had enough respect for Winder to tell him, e [I] never en- 
tertained one single ill thought of you in my life,' 5 The vicar of 
Carnmoney was a considerate friend, abnormally sensitive, who 
admired Swift so much that he made copies of his sermons. How- 
ever, sincere intimacy with an earnest young clergyman, ad- 
mittance to what elegance the society of Belfast and Carrick- 
fergus could afford, and the freedom of an independent style of 
life, all came short of the heights which Swift had breathed in 
Moor Park ; he says himself that he grew Sveary in a few months* . 6 
In the spring of 1 696, a way of release opened : Temple asked him 
to rejoin the family. 7 Swift did not hesitate long. 

1 DNB+ a Cotton v. 246. 8 Mason, p. 235, n. 

* Ball iv. 282, n. 3; DJVjB (Macartney). * Ball i. 26. Autob,, f. g v . 

7 Ball i. 18. 

[l6 4 ] 


Though he resided no more than a year at Kilroot, 1 he made 
up his mind to go away weeks or perhaps months before he actu- 
ally left. One cause for the delay was the comfort of Winder's 
presence: 'Had I been assured of your neighbourhood, I should 
not have been so unsatisfied with the region I was planted in.' 2 
A much stronger deterrent was Jane Waring. 

The family home of the Warings was the inland town named after 
them, about thirty miles southwest of Carrickfergus. Jane's 
father, Roger, had died in 1692, when he was not yet fifty. 3 An 
accomplished pluralist, he had, from 1682 to 1690, been Arch- 
deacon of Dromore, residing in Waringstown until the Troubles. 4 
Then he must have left Ireland about the same time as Swift and 
for the same reasons, since he was another person attainted in 
Dublin, 1689. The act of attainder had fallen mainly upon those 
who went to England so as to escape from the government of the 
dethroned king. It took in about twenty-five hundred people, of 
whom almost a hundred were clergymen. William Waring, the 
head of the family, moved to the Isle of Man in March 1689. 
Though he returned alone, a few months later, his wife and child- 
ren remained behind until peace was established. 6 Roger may 
have stayed abroad too long, for a new archdeacon was collated 
in August iGgo. 6 

When Swift met Jane Waring, she had poor health and was 
probably living in Belfast with her widowed mother. 7 He gave 
her a poetical nickname, 'Varina', by latinizing the W of * War- 
ing' and altering the g to a feminine a; she became the first of the 
three frail, fatherless, first-born young women to whom he suc- 
cessively attached himself. So large and prominent a family as the 

1 Landa, p. 24. a Ball i. 30. * Burtschaell gives his age in 1 658 as fifteen. 
* Edward D. Atkinson, An Ulster Parish (Dublin 1898), pp. 31-3. 
*/&*., pp. 34, 37, 51. 

6 H. B. Swanzy, Succession Lists of Dromore (Belfast 1933), p. 43. 

7 His letters were apparently addressed to her in Belfast; see Nichols, 1808, 
xvra. 243, n. 



Warings 1 Swift could hardly fail to meet, especially as his uncle 
Adam (whose property lay on the south coast of county Down, in 
Carlingford Bay) knew them, 2 and Swift had been at Trinity Col- 
lege with two of Jane's cousins. 3 At the same time, in his travels 
and visits through the area, he would have seen something of 
Henry Jenney, Roger Waring's successor ; for the Jenneys yielded 
only to the Warings as personages in the local society. But Swift 
could ride a horse from Kilroot to the city of Belfast in a few hours, 
and he probably did so often. 

Jane herself was the eldest of eight children, but not yet twenty- 
one years old when Swift moved to his new situation. 4 Her half- 
orphaned state and her illnesses (reminding him of Hetty Johnson 
at Moor Park) must have made it easy for him to sympathize 
with her and to wish to supply the guidance which he had missed 
at her age. The comparatively wide gap between them, in years 
(he was a third again as old as she) and in education, would have 
encouraged him to assign Hetty's filial role to the girl. Like his 
mother's flirtatious cousin, Elizabeth Jones, she was a parson's 
child. Like his sister, whose name she bore, she was the eldest in 
her family. There were plenty of elements to make Swift feel at 
home. But this time, if he began (as in 1691) * without any other 
design than that of entertaining myself*, 5 the affair soon found a 
deeper level. No doubt Swift was impelled in that direction not 
only by his solitude but by the effect of his cousin Thomas's 

By the end of the year, Swift had enough assurance to be cor- 
responding with Varina during a visit which he made to Dublin. 6 
Although the earliest letter preserved between them is dated 
29 April 1696, the marriage engagement which he proposes in it 
has obviously been for some time an unfailing source of heady 
conversation to the pair. Varina apparently considered his pro- 
posal ill-timed. She could not see herself enjoying life either in the 
vicinity of Kilroot or on Swift's income. e ls it possible', he wrote, 

1 Atkinson, pp. 24-54. a Ball i. 31, n. 2. * Ibid. i. 16, n. i. 
* She was born 8 Oct. 1674 (Swanzy, p. 43). * Ball i. 5. 
e Nichols x. 14, n. 



you can 1 be yet insensible to the prospect of a rapture and delight so 
innocent and so exalted ? Trust me, Varina, Heaven has given us 
nothing else worth the loss of a thought. 

His conditions are mild. She may live where she pleases until the 
wedding day; and the ceremony will take place only when she is 
satisfied with his income. C I will push my advancement with all 
the eagerness and courage imaginable, and do not doubt to 
succeed. 9 

Although the temperature of the message is properly exalted, 
the argument is didactic and even avuncular. Apart from talk 
about his love, he repeatedly begs Varina to mind her health and 
to give up her whimsies and affectations: e lt is true you have 
known sickness longer than you have me, and therefore perhaps 
you are more loath to part with it as an older acquaintance.' 

His urgency was due to a shift in his plans. Ireland was a dis- 
appointment. He held himself ready to remain there if she would 
engage herself to him, and not otherwise. If she preferred to re- 
main sickly and single, and he left the kingdom before she was 
his, he would 'endure the utmost indignities of fortune rather 
than ever return again, though the King would send me back his 
Deputy'. For Temple had felt Swift's loss and was not only in- 
viting him to return but promising not to leave him where he 
found him: 

I am once more offered the advantage to have the same acquaint- 
ance with greatness that I formerly enjoyed, and with better pro- 
spect of interest. 

For all Swift's fountains of eloquence, Varina remained un- 
willing. That her prudence was well founded is almost certain. 
Swift accuses her of being mercenary and makes acid reflections 
on the character of all women ; yet his concrete terms do not quite 
reach the unqualified promise of marriage. If one pokes among 
the rhetoric, one finds his admission that his friends 'still continue 
reproaching me for the want of gallantry, and neglecting a close 

1 Ball misprints 'cannot* (i. 19) ; other texts read *can*. 



siege'. Evidently, the hectic tone of these appeals made a serenade 
which Varina did not often hear from him, and which arose in 
part from Swift's very sense of their hollowness. She was a good 
listener ('your pity opened the first way to my misfortune 9 ) ; but 
if she had behaved herself with less caution at this point. Swift 
would probably have expressed himself with more. Although he 
uses the style which one associates with utter recklessness, he 
manages to make a very hard demand that she give him her 
decision immediately, because he is going to England in two and 
a half weeks. Yet he knows she desires more time, for he complains 
of 'your unreasonable scruples to keep me here 9 . That he looks 
for a negative seems clear, since in the midst of his rhapsody he 
says, e ln one fortnight I must take eternal farewell of Varina.' 
Swift was being true to the character he had given himself four 
years earlier: e l confess I have known one or two men of sense 
enough, who, inclined to frolics, have married and ruined them- 
selves out of a maggot [i.e., an infatuation] ; but a thousand 
household thoughts, which always drive matrimony out of my 
mind whenever it chances to come there, will, I am sure, fright 
me from that.' 1 

While their correspondence was nevertheless to continue, 2 the 
tone was to alter. His stipulations would grow more exact; her 
demurrals would soften. Soon it would no longer be a question of 
manners or of physical health. Varina's family and her character 
would be found wanting. Unless she radically rearranged her 
way of life and cut down her material needs, he would find him- 
self unable to keep up his own side of the understanding. Varina 
would infer, correctly, that she had a rival. 

1 Ball i. 6, 18-19. 

a Nichols (x. 14, n.) mentions unpublished letters of 20 Dec. 1695, from Dublin; 
and 29 Jun, 1696 and 28 Aug. 1697, from. Moor Park (besides the two in Ball). 


Chapter Seven 

Swift did return to Moor Park in 1696. Without giving up 
his prebend, he accepted Temple's invitation. During his 
absence the household had suffered. Thomas Swift had 
stepped into the office of secretary, perhaps at his cousin's recom- 
mendation. 1 In January 1693/4, however, Thomas was presented 
to the rectory of Puttenham, about five miles west of Moor Park, 
toward Guildford. While Temple may have helped him to obtain 
this living, 2 the duties must have limited the attention which he 
could give to his patron. Here was one occasion for the recall of 

Another was the death of Lady Temple, buried in Westminster 
Abbey, February 1694/5, soon after the funeral of her friend, 
Queen Mary. Temple's brother John commented to a relative, 
'You have heard, I believe, of the greate losse, that wee have 
lately had here, by the death of my Sister Temple, which hath 
been a greate trouble to us all in this family, and with a greate 
deale of reason, for a better woman I never knew, and no body 
could be kinder, then she had alwayes been, both to mee and my 
children. 33 So Sir William's immediate circle shrank to the di- 
mensions of his sister, his French daughter-in-law and her 
mother, his two grand-daughters, Mrs Dingley, Mrs Johnson, 

1 In a sermon published 1710, Thomas Swift labelled himself 'formerly chaplain 
to Sir William Temple*; in its dedication, he called Temple *my own -wise patron' ; 
he -witnessed Temple's -will 1695; ^^ he wrote a letter for Temple to a bookseller, 
Feb. 1694/5 (Ball i. 387-8; Woodbridge, p. 231). 

a See William Wotton, Observations on A Tale of a Tub, 1705, p. 67. 

3 Dated from East Sheen, 9 Mar. 1694/5, to Thomas Flower. 



and young Hetty. 1 Yet the summer before this was probably the 
time when he impressed a visitor as leading an ideal life : 

far enough from the town to be protected from visits, the air whole- 
some, the soil good, the view limited but pretty, a little stream which 
runs near making the only sound to be heard ; the house small, con- 
venient, and appropriately furnished; the garden in proportion to 
the house and cultivated by the master himself. He is free from 
business, and to all appearances free from ambition; he has a few 
servants, and some sensible people for company. . . I saw Monsieur 
Temple healthy and gay; and though he is gouty and getting on in 
years, he tired me in walking, and except for the rain which inter- 
rupted us would, I believe, have forced me to ask for quarter. . . 
This good old man thought I should not be sufficiently repaid for 
my trouble if I saw only his little house; and though I assured him 
that I was more interested in men than in buildings, and that I was 
content with the honour of having seen him, he insisted that before 
returning to London I should go to [Petworth], the country house 
of the Duke of Somerset. He gave me horses and servants to take me 
there, and fearing that the Duke might be gone to London, he asked 
my Lady Temple to write to the Duchess. 2 

The picture of her asking the Duchess of Somerset to receive a 
young Swiss traveller is the last direct view we have of Dorothy 
Osborne. What her passing meant to the man who had courted 
her for six years, through war, smallpox, and the disapproval of 
both fathers, is foreshadowed in his sister's record : c He was often 
heard to say how happy his life had been if it had ended at fifty.' 8 
About the middle of May 1696 (just after James IFs 'invasion* 
collapsed) Swift departed from his parishes. 4 He probably moved 
from Belfast to Dublin, to Holyhead, to Leicester, and there 
paused to visit his mother. Then came London, while the govern- 
ment's worst financial crisis was being complicated by the arrest 
of Sir John Fenwick; and at last Moor Park. Here, at the end of 

1 Woodbridge, pp. 217-18, 233. 2 Ibid., p. 232. 

8 Temple, Early Essays, p. 21 (the remark was written before the death of Lady 

4 On Wednesday, 29 Apr. 1696, he said he planned to leave Kilroot for Dublin on 
Monday fortnight ("18 May), there to take leave of the lord deputy and go quickly 
to England; but if CapeTs illness was as serious as he had heard, he might sail 
directly from Belfast or Carrickfergus; Gapel died 30 May (Ball i. 18). 



June 3 one of Swift's first gestures was to write Varina a letter, in 
diction perhaps less molten than before. 1 

Although he waited another year and a half before resigning 
Kilroot, Swift does not seem to have felt disappointed by the re- 
ception Temple gave him. He had been granted a licence of ab- 
sence from his cure, extendible if he so desired. 2 However, ac- 
cording to Swift's sister. Temple *was so fond of him . . . that he 
made him give up his living in [Ireland] , to stay with him at 
Moor Park, and promised to get him one in England'. 3 Her im- 
pression exaggerates the warmth of Temple's feelings, for he 
urged Swift to write for a 'further licence'. Yet we know that 
Swift decided to resign the prebend while he was still in Ireland. 
He even promised Winder that he would do what he could to get 
him appointed as Swift's successor. Few people besides Winder 
seem to have encouraged such imprudence; and even Winder 
was reported to have advised against it. 4 Never thelesSj Swift car- 
ried out the scheme; and his friend was installed exactly three 
years after himself. 5 

One can hardly miss the smell of bridges burning. Not only is 
there the violence of Swift's threat to Varina C I will endure the 
utmost indignities of fortune rather than ever return again' but 
on the very eve of his resignation, Swift's hopes in England were 
dashed. Temple had recommended him to the Earl of Sunder- 
land, whom the king brought out of seclusion, at the time of the 
peace negotiations, to serve as lord chamberlain. But the earl 
gave up his office soon after the treaty was signed at Ryswick. 8 In- 
stead of hesitating at this turn, Swift plunged ahead : c Ten days 
before my resignation [of Kilroot], my Lord Sunderlaud fell and 
I with him/ 7 One seldom shoves a treasure aside with such eager 
reluctance if it does not hold a strong allurement. Swift was prob- 
ably showing that he lived, as he had hoped Varina would do, 
above the c paltry maxims' of that prudence which is 'calculated 
for the rabble of humanity*. 8 

1 1 am assuming that Swift followed his usual course and that Nichols is right 
about the correspondence with Varina (x. 14, n.). 

* Landa, p. 24. s Ball r. 30. * Ibid. i. 24. 5 Cotton nr. 266. 
e Ogg n. 4.37-41 . 7 Ball i. 24. 8 Ibid. I. 20. 




Meanwhile, in the new order at Moor Park, Swift functioned less 
as a clerk than as a companion and literary aide. After 1695, 
Temple wrote practically nothing for publication. However, his 
poems, 1 a number of essays, a volume of memoirs, and what ap- 
peared as three volumes of letters, had all to be prepared for the 
press. The first two volumes of the letters had been mainly col- 
lected by one of Temple's secretaries; and Swift did little with 
these but transcribe them in chronological order and translate 
those in French or Latin. A few in Spanish were probably trans- 
lated by Lady Giffard, if we may judge from the flourish of praise 
which Swift bestows upon that labour. 2 The additional letters in 
these volumes (besides those collected by the earlier secretary) 
and all the letters in the third volume were chosen as well as re- 
copied by Swift, under Temple's direction: *I had begun to fit 
them for the press during the author's life 5 , Swift said later; c but 
never could prevail for leave to publish them: Tho' he was 
pleased to be at the pains of reviewing, and to give me his direc- 
tions for digesting them into order. 53 Similarly, Temple's 'third' 
volume of Memoirs had to be e copied from the originals by Sir 
William Temple's direction, and corrected all along by his 
orders'. 4 Exactly this sort of work had been Swift's province from 
his first days at Moor Park. 5 

The measure of his advance over those days is given by the 
tone of a letter which he wrote early in 1698. Sir William and his 
sister had gone to London, taking with them Mrs Dingley, Mrs 
Johnson, and her daughter Esther a kind of companion and 
waiting woman, it seems, to Lady Giffard. Ralph Mose was 
Temple's steward; Swift and he had charge of the establishment 
during the family's absence. The letter was almost certainly ad- 

1 The poems, though not published, were privately printed; and a copy, perhaps 
unique, is in the British Museum. 
a Davis i. 258. 8 Ibid. i. 259. 

* Ball i. 172; the 'first 7 volume of Temple's Memoirs was not published or pre- 
served. Cf. Swift's remark on the third volume of Temple's Letters: 'They were 
corrected by himself; and fairly transcribed in his life time* (Davis I. 266) . 

* Ball i. 171-2. 



dressed to the seventeen-year-old girl; for in it the recipient is 
coupled with Lady Giffard's pet bird of paradise, as 'fellow-ser- 
vant' to her ladyship, and Swift jokes about an association with 
e my lady's chamber floor'. In one sentence he suggests the ease 
with which he now lives in Moor Park, the warm-but-teasing 
relationship with Hetty, and his famous habit of saying the kind- 
est things by their opposites : 

I desire your absence heartily, for now I live in great state, and 
the cook comes in to know what I please to have for dinner : I ask 
very gravely what is in the house, and accordingly give orders for a 
dish of pigeons, or, &C. 1 

When everybody was home, there must have been pleasant, 
if hierarchical, social evenings together. Frequently, Temple 
would give the young people a little money for stakes, and they 
would all play cards. 2 Swift also used to visit London, though we 
do not know how often; he later described himself as being at this 
time a 'gentleman much in the world 9 . 3 Temple sometimes sent 
him on errands to court, possibly in order to remind the king of 
Swift's name. 4 

The uncertainty of his position went on, of course, and he did 
not like being a dependent; yet though he tried 'other courses 9 
(unspecified) for preferment after Sunderland's withdrawal, they 
were fruitless. 5 Temple did recommend him to the king, but with 
how much force we do not know. 6 The baronet took himself 
seriously; there was not much in his past that he could think 
about without the danger of painful contrasts; and Swift had 
sometimes to put up with more reserve or gloom than he was 
ready to make allowances for. Years later, when a powerful 
friend acted diffident. Swift said, 

One thing I warned him of, Never to appear cold to me, for I would 
not be treated like a school-boy; that I had felt too much of that in 
my life already (meaning from sir William Temple) ... I think what 
I said to [him] was right. Don't you remember how I used to be in 

1 Ball i. 21 -st. * Journal n. 561. 8 Tale, p. 4. 4 Longe, p. 316. 
6 Ball i. 24. 6 Ibid. m. 301. 


pain when Sir William Temple would look cold and out of humour 
for three or four days, and I used to suspect a hundred reasons P 1 

However familiar he might feel with Moor Park, it was not 
Swift's home. His sister, who had been there with him, and was 
eventually to return and work for Lady Giffard, now lived in 
Dublin. 2 His cousin Thomas, with a parish and a wife to look 
after, perhaps found himself too busy to haunt the young genius. 
Anyhow, Swift, for all the confidences which he had reposed in 
him, had already begun to sound contemptuous of Thomas in 
1 693.3 Like many apparently close friends who have been con- 
nected by less sympathy than proximity, the two parsons lost 
rapport with advancing age. They possessed so many experiences 
in common that any clash of temperament must have made a 
bold impression. Swift obviously acted the dominant part; and if 
he had wished to strengthen the tie, he could have done so : it was 
Thomas who saved his letters, not Swift who saved his cousin's. 

If Swift felt ambitious, the strict morality but thin emotional 
indulgences of his childhood had trained him to put a low rating 
upon his natural abilities or charm. He did not look for disin- 
terested affection, though he could evoke it. To signs of contempt 
he was unreasonably open. Frequently, he gave offence to begin 
with, for fear of receiving it to end with. Instead of finding prac- 
ticable standards of conduct in himself, he therefore made heroes 
of men like Temple and Sancroft. The comparative lack of 
friction between Sir William and his protege is a credit to the 
older man's urbanity. 

Nothing about Temple impressed Swift so much as the high 
road which he had left. Refusing twice to become secretary of 
state was an act of as much virtue as genius. There is no sign that 
Temple underplayed the role of Gato Uticensis. After Swift came 
to know intimately the men who ruled England, he might smile 
at the 'splutter Sir William Temple [made] about being secre- 
tary of state'. 4 While he still lived at Moor Park, however, Swift 

1 Journal, 3-4 Apr. 1711. a Ball i. 8. * Ibid., pp. 367-8. 
* Journal^ 3 Nov. 1711. 


had no reservations about his patron's magnificence. Through- 
out Swift's life, he never referred to Temple except by his full 
name ; the most private allusions are always respectful ; the public 
allusions are always admiring: 'this great and good person' 
(1700)3 fi a certain great man . . . universally reverenced for every 
good quality that could possibly enter into the composition of the 
most accomplished person' (I7O9). 1 


Temple's glory and integrity seemed not superior to his literary 
gifts : C I never read his writings'. Swift said in 1692, 'but I prefer 
him to all others at present in England.' This author', he said in 
1 700, e has advanced our English tongue, to as great a perfection 
as it can well bear.' 2 Both Temple and his sister helped to direct 
the aspiring writer's first exercises. They had him undertake a 
translation of Virgil. 3 He wrote his ode to the Athenian Society 
only after c Sir William Temple speaking to me so much in their 
praise, made me zealous for their cause.' 4 

The strenuous course of reading which Swift undertook at this 
time may have owed something to Temple's advice and was 
probably meant as a propaedeutic to literary composition. When 
he prepared sketches of English history, he used Temple's Intro- 
duction and started from Temple's sources. After Temple gave up 
the idea of writing a complete history of England, he spoke of 
turning the job over to younger men, in words which could easily 
have referred to Swift. 5 Of a list of books read by Swift in 1697 
and i6g8/ history political and ecclesiastical accounts for a 
third of the titles. The classics, from Homer to Petronius, come to 
less than a quarter. Literature of travel, some books concerning 
Christian doctrine, French belles-lettres (e.g., Voiture, Fonte- 
nelle), and works on esoteric aspects of religion are (ranged from 
the more to the less important) other distinct groupings. Lucre- 

1 Davis i. 259, 6. * Ball i. 365; Davis r. 258. 8 Ball i. 365. 
4 Ibid., p. 363. 

8 See my 'Swift's History of England', JEGP., LI (Apr. 1952), 177-85* 
6 See Talg, pp. Ivi-lvii. 


tius he read three times; Virgil, twice; Lucius Florus a com- 
pilation based upon Livy three times. 

Copying out, during more than a decade, the works and letters 
of his patron set Swift a model of the highest value, Temple's 
prose style; for Lamb underestimates that in calling it 'plain 
natural chit-chat'. It is marked at its best by the use of short sen- 
tence elements with little affectation of periodic structure. The 
phrases are idiomatic, the vocabulary colloquial, the imagery 
striking and homely. In place of the slow cadences of Jacobean 
and Caroline rhetoric, Temple has a varied, unstressed rhythm. 
There are good examples in even his early essays : 

Amongst all those passions which ride men's souls none so jade 
and tire them out as envy and jealousy; their journey is longer 
than any of the rest, they bate seldomer, and commonly ride 
double. 1 

The unobtrusive parallelism gives coherence; the quiet meta- 
phor gives life; the tone, learned from Montaigne, of independent 
observation and personal reflection, gives character. Temple had 
a gift for anecdote, character-drawing, narrative history, and 
lucid (if superficial) exposition. Accuracy of information hardly 
mattered, since his data were rather illustrative than demon- 
strative; he used facts to embody moral generalizations, not to 
record what happened. He possessed little genius for large 
organization in the manner of Hooker; like most essayists of the 
Restoration, he conceived his works in paragraphs raying out 
from a central theme and unified by attitude or style. Structure 
in the sense of an articulated skeleton, the kind of outline-logic 
taught at the universities and employed to support a sermon or a 
formal oration, he hardly attempted. Recreation, not proof, was 
his aim. 

Taking over the essential qualities of Temple's style, Swift 
transfigured it through his violence, his wit, and his irony. How- 
ever, of the brilliant style which makes A Tale of a Tub the great- 
est prose satire in English, there is small promise in Swift's early 

1 Early Essays 9 p. 166, 



poems. The proper source to examine for such foreshadowings is 
not his verse but the first specimens of the genre in which Swift 
shows his most shining literary virtues letters. Although now 
seldom handled as aesthetic objects, there is no doubt that the six 
volumes of his general correspondence and the two volumes of 
the letters to Esther Johnson contain a higher achievement than 
all but the best-known of his essays and poems. In this art, Swift 
may have an equal; he has no superior. 

In his good letters, Swift plays with the correspondent as a 
witty author plays with a reader. 1 The famous irony appears here 
when Swift's poems are still grim with sobriety and sticky with 
one-sided idealism. He tries a great range of comic devices, es- 
pecially with intimates on whose affection he can rely : practical 
jokes, flattery disguised as insult, word play, farcical exaggera- 
tion. But when he wishes to sound serious, he can communicate 
bitter self-containment, gratitude, warm friendship, sarcastic 
contempt, deep admiration, all through turns of phrase rather 
than superlative epithets. The heart of Swift's literary effective- 
ness is not his vocabulary but his syntax. 

The earliest letter preserved, written when he was twenty-four, 
to his cousin's husband, is a remarkable accomplishment as a re- 
spectful but sternly dignified epistle. His gift for phrase-making is 
already ripe e my mind was like a conjured spirit 5 , e the people is 
a lying sort of beast'. His habit of humble, Bunyanesque meta- 
phor appears throughout : *But whenever I begin to take sober 
resolutions, or, as now, to think of entering into the church, I 
never found it would be hard to put off this kind of folly at the 
porch.* The purity of language and expression, the distinctly 
articulated sentence structure, the tough but elegant manage- 
ment of periodic units, the clarity of meaning, the intensity of 
feeling, the boldness of self-analysis, which permeate the letter, 
make it easy for one to understand why the recipient, to whom it 
was in part a rebuke, decided to save it: 

1 There is a fine analysis of this aspect of the Journal to Stella in Dr Mackie L. B. 
JarrelPs unpublished Ph.D. dissertation, 'Swift's "Peculiar Vein of Humor" ', Uni- 
versity of Texas, 1954. 



And as to that of my great prospects of making my fortune, on 
which as your kindness only looks on the best side, so my own cold 
temper, and unconfined humour, is much greater hindrance than 
any fear of that which is the subject of your letter, I shall speak 
plainly to you, that the very ordinary observations I made with go- 
ing half a mile beyond the university, have taught me experience 
enough not to think of marriage till I settle my fortune in the world, 
which I am sure will not be in some years; and even then I am my- 
self so hard to please that I suppose I shall put it off to the other 
world. 1 

A letter to cousin Thomas, three months later, demonstrates 
Swift's infinite capacity for turning his ordinary tasks into a 
drama : 

It makes me mad to hear you talk of making a copy of verses next 
morning, which though indeed they are not so correct as your 
others are, [is] what I could not do under two or three days, nor 
does it enter into my head to make anything of a sudden but what 
I find to be exceedingly silly stuff by great chance. I esteem the time 
of studying poetry to be two hours in a morning, and that only when 
the humour sits, which I esteem for the flower of the whole day, and 
truly I make bold to employ them that way, and yet I seldom write 
above two stanzas in a week I mean such as are to any Pindaric 
ode and yet I have known myself in so good a humour as to make 
two in a day, but it may be no more in a week after, and when all is 
done I alter them a hundred times, and yet I do not believe my- 
self to be a laborious dry writer, because if the fit comes not im- 
mediately I never heed it, but think of something else. 

The same letter shows his talent for exposing his own frailties 
without sacrificing our respect, because the style of the exposure 
is so masterly that we must envy the writer: 

I have a sort of vanity or foiblesse, I do not know what to call it, and 
which I would fain know if you partake of: it is not to be circum- 
stantial that I am overfond of my own writings ; I would not have 
the world think so, for a million, but it is so, and I find when I write 
what pleases me I am Cowley to myself and can read it a hundred 
times over. I know it is a desperate weakness, and has nothing to 

1 Ball i. 4-5 ; Ball's text reads 'myself I am'. 


defend it but its secrecy, and I know farther, that I am wholly in the 
wrong, but have the same pretence, the baboon had to praise her 
children, and indeed I think the love in both is much alike, and 
their being our own offspring is what makes me such a blockhead* 1 

Even in addressing Vaiina, when he sinks nearest the bathos of 
the pindaric odes, he cannot suppress his nervous genius : *I am 
once more offered the advantage to have the same acquaintance 
with greatness that I formerly enjoyed/ e lt is true you have 
known sickness longer than you have me, and therefore perhaps 
you are more loath to part with it as an older acquaintance.' 2 But 
it is not until we reach the message to Esther Johnson that true 
humour appears: sympathetic identification of oneself with the 
subject ridiculed; affectionate raillery; harmless, careless wit, 
more to be valued for the good nature than the bright mind be- 
hind it 'Here are three letters for you, and Molly will not send 
one of them 3 ,, he lies; 'she says you ordered her to the contrary* : 
but then he relents, confessing, 'Here is a great bundle and a let- 
ter for you; both came together from London. 33 Perhaps this air 
of direct speech, unscreened by ink and paper, but simple, open, 
and fresh, is the most solid foundation of Swift's epistolary art. 


In comparing the styles of Temple and Swift, the question of in- 
fluence is, as always, perilous, since their common traits often 
seem more innate than acquired; and even when a trait does ap- 
pear to have been taught or learned, it is often the effect either of 
general fashion or of common masters. During the Restoration, 
for example, nobody required an introduction to the names of 
Montaigne and Sir Thomas Browne. Yet if Swift was not brought 
to them by Temple, he was certainly encouraged to read them. 
They are palpable influences upon Temple's early essays of the 
1650*8 and on Swift's writing a generation afterward. 
That Swift derived his literary style immediately from Temple 

1 Ball i. 363-4; Ball's text reads 'partake of it: it'. * Ibid., pp. 18-19. 
* IbitL, pp. 21-2. 

N [179] 


one can take for granted, since besides transcribing and editing 
hundreds of pages of his patron's writings, he also translated 
scores of the French and Latin letters into English, all under 
Temple's correction. (William Wotton thought he could identify 
Temple as the author of A Tale of a Tub by the vocabulary. 1 ) In 
both men's prose one finds the same colloquial elegance, the same 
command of idiom, the avoidance of technical terms, of preten- 
tious jargon, or the appearance of pedantry ; they share the taste 
for homely similes, 2 though Swift's seldom run so long as Temple's ; 
and they like to use lists. But Dryden also has these traits; and 
while he had welcomed technical terms into the Annus Mirabilis, 
he explicitly disavows them in the dedication of his Aeneid, where 
he claims to be writing like Virgil, c to men and ladies of the first 
quality, who have been better bred than to be too nicely knowing 
in the terms' a remark which might have come from Temple. 
(Hobbes and Rymer made similar attacks upon * terms and jar- 
gon' 3 ; and when Swift came to write on language, he too would 
show less interest in enrichment than in purgation. 4 ) 

In his letters, Temple employs a graceful style, direct but 
charming, much like Addison's ; but Swift has far more variety of 
tone and change of level. While Temple often brightens his man- 
ner with the risque* gallantries which Swift avoids, he never uses 
the coarse expressions, the word play, or the complicated ironic 
flatteries with which his protege* (who learned some of them from 
Voiture) is identified; and of course he lacks the crisp, nervous, 
aphoristic talent that gives distinction to many of Swift's rou- 
tine messages. Temple avoids giving details even where he ob- 
viously should (and could) supply them 6 ; Swift glories in his 

Temple's periods are much longer (though not weightier) than 
Swift's ; but the degree of difference is less than at first appears : 
in a persuasive analysis, Macaulay, distinguishing the truly in- 

1 Tale, p. 314. 

2 Cf., in Temple's Observations upon the U.P., p. 257 : 'But as any rough hand can 
break a bone. . .* 

8 SpjLnganx n. 64, 68, 170. 4 Davis rv. 10-16. 
5 E.g., Observations upon the U.P., p. 250. 



volved syntax of Hooker or Clarendon from that of Temple, says 
that a careful judge 

will find that [Temple's apparently long sentences] are not swollen 
by parenthetical matter, that their structure is scarcely ever intri- 
cate; that they are formed merely by accumulation, and that . . . 
they might, without any alteration in the order of the words, be 
broken up into very short periods. 1 

Temple has only the seed of Swift's genius for phrase-making, 
aphorism, and startling analogies; Swift's conciseness, energy, 
variety, and violence are foreign to Temple's style. Yet when 
he does happen to strike out a 'thought', the result can be re- 
markably like Swift: e We bring into the world with us a poor 
needy uncertain life, short at the longest, and unquiet at the 
best.' 2 

Rarely does Temple indulge in either bitter or teasing senti- 
ments ; but when he does, he comes nearest the vigour we associ- 
ate with Swift: 

I shall not engage in answering the complements of your letter, 
tho' I should have much more justice on my side; but I am very ill 
furnished with that sort of ware, and the truth is, there is required 
so much skill in the right tempering, as well as the distribution of 
them, that I have always thought a man runs much hazard of losing 
more than he gains by them; which has made me ever averse, as 
well as incapable of the trade. 3 

This lacks Swift's nervous syntax, his degree of insinuation, his 
conciseness, and his irony; but it has his command of idiom, his 
simplicity of phrase and language, his talent for making structure 
follow feeling, and most important his directness, which lets 
one feel an emotion as if it were carried by the tone of a voice. 

On the contrary, it is in Swift's rare elegiac cadences that his 
style most closely agrees with Temple's, the prolonged, receding 
rhythms in which both of them take a melancholy view of his- 
torical ironies : e.g., 

1 Works, cd. Lady Trevelyan, 1873, vr. 280. * Miscellanea i. 175-6. 

* Letters I, 73-4- 



[Temple:] And thus an adventure has ended in smoak, which had 
for almost three years made so much noise in the world, restored 
and preserved so long the general peace, and left his majesty the 
arbitrage of all affairs among our neighbours, by the Emperor and 
Spam's resolutions, as well as Sueden and Holland's, to follow his 
measures for the common safety and peace of Christendom. 1 
[Swift :] And now it is done, it looks like a dream, to those who will 
consider the nobleness of his birth, the great merits of his ancestors 
and his own, his long unspotted loyalty; his affability, generosity, 
and sweetness of nature. 2 

Such Virgilian pathos is of course far more characteristic of 
Temple than of Swift. It is, indeed, the essential strains or im- 
pulses that most deeply divide the two. Where Temple chastens 
his natural formality to give it the ease of speech. Swift must guide 
the vigour of his colloquial instincts and impose the clarity of 
form upon them. Temple works to keep his language from sound- 
ing too fastidious; Swift's danger is leaning towards coarseness. 
Temple's periods tend to rhetorical fullness, even where an 
abrupt movement would be more effective; Swift's rhythms, 
more broken and more spontaneous, have the air of nature but 
seldom grow awkward. It is in A Tale of a Tub, therefore, where he 
is under least restraint, that Swift stands farthest from his master. 

1 Letters n. 307. Cf. the Ciceronian echo in Temple's curtain on the Battle of 
Hastings : "This was the man, these the forces, and such the circumstances that con- 
tributed to so famous an enterprise, by which the fate of England was deter- 
mined. . .' (Introduction to the History of England, pp. 1 19-20). 

2 Davis vm. 132-3. 


Part III 

Chapter One 

The composition of A Tale of a Tub produces little stir in 
Swift's biography because he has left no reference to the 
work made during its progress and almost none after its 
completion, apart from the Apology and a letter to the publisher. 1 
As the book now stands, much of its apparatus belongs to a later 
time than the first finishing of it. The notes, the Apology, the dedi- 
cation to Lord Somers, the c Bookseller [i.e., publisher] to the 
Reader' were not part of the original plan. We are left with the 
allegory of the three brothers, spread over five 'sections 5 (H, rv, 
vi, vnr, xi) ; five digressions interrupting the e tale' ; a dedicatory 
epistle, preface, introduction, and conclusion. Two of the five 
digressions have an evident order. That on madness is proper- 
ly in the climactic position, and, since it expounds a theory 
of vapours, follows the 'Aeolism* episode in the tale. The last, 
un titled digression ('Section x') is certainly intended as a con- 
clusion to all five. The other three have no more excuse for 
their locations than what Swift provides for the third ('Section 

I have chosen for it as proper a place as I could readily find. If the 
judicious reader can assign a fitter, I do here empower him to re- 
move it into any other corner he pleases. 

Although the scheme is odd enough deliberately so it does 
have precedents, such as L'Estrange's version of Quevedo's 
Visions and Donne's Ignatius His Conclave, both of which are ex- 

1 Ball 1. 183-5. 


amples (known to Swift) of discursive, segmented satire with a 
centrifugal design. 

The earliest part of the Tale to be conceived was probably the 
religious allegory. Of this the root ideas a magical legacy and 
three brothers who fall out are common in folk lore and litera- 
ture. 1 Swift's version is similar to the old story of three rings left 
by a rich merchant to his three sons : though all the rings look 
alike, only one is genuine, and the owner of it is heir to the whole 
paternal estate. Conventionally, the father is God ; the children 
are the Jewish, Christian, and Mohammedan faiths ; and.thevim- 
plied meaning, is that we should, regardless of our church,,, isdve 
through^ ..i&anty. to deserve-God s 5, .blessing. Boccaccio took this 
plot for his third novella in. the Decameron, and Lessing has em- 
bodied it in Nathan the Wise. After the Reformation it is some- 
times applied to the threefold division of Christianity, with the 
sons representing the Lutheran, Calvinist, and Roman Catholic 
communions; the father's will (i.e., the New Testament) grows 
important; the rings fade out; the meaning becomes not eirenic 
but anti-Catholic. There are close parallels to Swift's treatment 
in an early seventeenth-century German play (based upon a 
Renaissance Latin source) 2 and in a notorious sermon preached 
(1686) by John Sharp, later Archbishop of York. 3 Swift replaces 
the estate by three seamless coats (the Christian faith) ; he gives 
the will a fundamental role; names the sons Martin, Jack, and 
Peter; and turns the ridicule against both Geneva and Rome : the 
plea for charity is now an aggressive, satirical defence of the 
Church of England. Since all the elements of his plot were in the 
air, we need not exercise our minds as to his 'sources'. 

There may be some truth in the tradition that Swift produced 
the allegory while he was at Trinity College, and that he showed 
sketches of it to friends there or in Kilroot. 4 Temple, of course, 

1 Cf. J. A. Macculloch, The Childhood qf 'Fiction, 1905, pp. 350-80. 

8 Martin Rinckhardt, Der Reichslebischer Christlicher Ritter, 1613. See G. M. 
Webster in MLN., XLVTO (1933), 551-3; also C. J. Home and H. Powell, A Ger- 
man Analogue for A Tale of a Tub', MLR,, LV (1960), 488-96. 

3 Tale, pp. xxxi-xxxvi. 

4 Ibid., pp. xxxiv-xxxvi, Lyon says it 'appears from some sketches ... in his own 
hand' that Swift (c. 1685-8) wrote an account of the kingdom of absurdities 



frowned on satire. If Swift had such works under way when he 
first came to Moor Park, he probably tabled them. Returning to 
Ireland in 1 694, he might have felt emancipated enough to let his 
instincts have their way. Nevertheless, no allusion which can be 
dated points to a period earlier than 1696; and the Tale shows 
traces of many books which he read in 1697 and 1698. In 1709 
Swift said, 'The greatest part of that book was finished above 
thirteen years since, 1696, which is eight years before it was pub- 
lished.' 1 Since he had a habit of remembering dates as earlier than 
the truth, this statement probably puts the farthest limit on the 
writing of the text which we know. Delivered from the cures of 
three parishes, untroubled by Varina's proximity, restored to the 
leisure and the bluestocking ambience of Moor Park, and moved 
by Temple's authorial example, he probably began to compose 
the Tale during the latter half of 1 696. 

The five digressions (sections in, v, vn, ix, x) hardly refer to 
the course of the intervening plot, and seem to have been struck 
off later, during 1697-8. They contain attacks upon pedantic 
scholarship, 'modernism' in literary style and taste, political and 
military crime, mechanistic natural philosophy, occultism, 
Roman Catholicism, and, above all, Puritanism, The 'Introduc- 
tion' and 'Conclusion 3 , very much to the same purposes, carry 
signs of the 1697-8 period of composition. The 'Preface' antici- 
pates most of the themes in the succeeding pages ; it offers a mock- 
ing but not misleading declaration of aims, and ends in a pseudo- 
dissertation on the nature of satire. Like the 'Epistle Dedicatory*, 
just preceding, it sounds as though it had been done right after, 
or at the same time as, the body of the book. The 'Epistle Dedi- 
catory', which contains much free-flying bombardment of 
contemporary authors and scholars, is subscribed December 

(p. 15). Lyon also believed that several people saw a draft of the Tale while Swift 
was at T.G.D. (p. 25). 
1 Tale, p. 4. 



The character of Martin probably belongs to the earliest part of 
A Tale of a Tub. Certainly it is the weakest ingredient. If one seeks 
a biographical explanation, one may say that in a book dominat- 
ed by the principles which Swift shared with Temple, the figure 
representing Swift's positive religious values could not be strong, 
because Christian devotion, never meant so much to Temple as 
it did to Swift. True, they both conceived of Christianity as a 
simple, easy, rational faith, single, coherent, and harmonious, 
independent of studied erudition 1 ; but the reverence which drew 
Swift into the clergy was alien to j Temple. One may also say, 
thinking historically, that after, experiencing the desolation of 
Kilroot, Swift could hardly recapture the mood of his ode to 
Bancroft. Or, thinking of Swift's character, we may say that es- 
pecially in a comic setting he had trouble paying straightforward 
tribute to what he cherished most. 

But, in its literary aspect ? the treatment of Martin is weak be- 
cause it clashes with the programtnc^of the book as a whole. This 
programme is for the author to imply a virtue which he desires to 
recommend, but to state it either not at all or else ironically by 
pretending to depreciate it. In seventeenth-century England it 
would scarcely have been feasible for a writer to communicate 
his love for the Established Church by the medium of mock- 
insults. Swift would have accomplished his aim best by leaving 
Martin out. However, the central allegory which he chose almost 
precluded that possibility. 

For one who admires the Tale of a Tub, therefore, the safest 
way to demonstrate its merits is to admit the flaw and then to 
consider what Swift accomplishes in spite of it. However, in 
order to succeed in this task, one must start from the objects 
which he implicitly proposes for our respect and emulation. 
That few of his favourable critics have done so, may be due either 
to their misunderstanding of the nature of his satire or to their un- 
willingness to face the conventionality of his recommendations. 

* Temple, Miscellanea ra. 260-70. 


6 A TALE OF A TUB 9 (l) 

Unfortunately, those who occupy themselves wholly with 
techniques and targets can find little genuine structure in A Tale 
of a Tub. The patterns which they offer are usually so abstract 
and so remote from the concrete detail of the book as to seem, for 
practical purposes, irrelevant. If we root ourselves firmly in the 
principles which Swift advocates, however, we shall have to con- 
fess that these are neither original nor dazzling. Literary critics 
who dislike the conclusions of the supreme moralists, from St 
Luke to Freud, will find A Tale of a Tub diminished by this per- 
spective. But there is no other way to bring order into the chaos 
of commentary upon what is, after all, an affirmation of quite 
accessible wisdom. Those who believe that rationality carries one 
away from virtue, or that agony and mindless mysticism are the 
invariable tokens of true religious faith, may call A Tale of a Tub 
a great satire; but unless they agree that it satirizes them, they 
cannot understand it. 

A good clue to Swift's scheme is his remark in an Examiner 
paper written many years after the Tale: *In describing the vir- 
tues and vices of mankind, it is convenient, upon every article, 
to have some eminent person in our eye, from whence we copy 
our description.' 1 For the virtues implicitly recommended by A 
Tale of a Tub, the eminent person is of course most often Sir 
William Temple; for the vices, Richard Bentley and William 
Wotton lead the train, but other antagonists of Sir William's vir- 
tues march close behind. This is in no way to suggest that one 
cannot comprehend and enjoy the book without a knowledge of 
Temple; certainly Swift never made such a demand upon his 
reader. Swift's view of Temple was itself a construction, and he 
respected him as embodying ideals which were largely implanted 
before the two met. With this in mind, we may state the problem 
as not to discover how Temple appears in A Tale of a Tub but, 
using the figure of Temple as a reference, to see what sort of per- 
son Swift is urging us to imitate. My only postulate is that behind 
the book stands not a list of philosophical propositions but the 
idea of a good man. 

1 Davis DJ. xo-ix. 



This idea belongs to the tradition usually called Christian hu- 
manism and descending to Swift from the great line of Spenser, 
Sidney, and Milton. However, the hero has been tamed into the 
shape of a gentleman; for nobody so appalled as Swift was by the 
effects of Puritan self-confidence could advise men to use epic 
standards as the test of merit. When Temple examined 'heroic 
virtue' in his finest essay, it was the Roman talents that he exalted 
-justice, government, prudence, and advances in the practical 
arts not self-sacrificing magnanimity or sublime patriotism 
demonstrated on the battlefield; it was his grandfather who had 
fought beside Sir Philip Sidney, but it was Henry Sidney who be- 
came Sir William's adored friend. 

Swift sets before us the Christian gentleman, a landed pro- 
prietor educated in humanist culture, conscious of his duties as a 
subjec.t and a master. This model conforms cheerfully to the 
Established Church; he willingly supports the government of 
England as redesigned in 1 688-9, an( i ne * s a responsible head of 
a family. As a father and as a landlord, he lives by the rules of 
Christian morality and labours to improve the condition of those 
who depend upon him. He uses polished manners, rereads the 
Greek and Latin classics, keeps his hereditary property in order, 
and maintains his elevated place in society. While everybody 
cannot hope to enjoy such a position, Swift argues, only fools 
would prefer another. 

The satire is therefore directed against the fools and their ante- 
cedents : those who oppose 'the direct rules of the gospel, the cur- 
rent of antiquity, the religion of the magistrate, and the laws of 
the land 9 . 1 Swift takes up the alternatives to his model in religion, 
in culture, in politics, and in manners. When he described the 
genesis of the book, he said, 

The author was then young, his invention at the height, and his 
reading fresh in his head. By the assistance of some thinking, and 
much conversation, he had endeavour'd to strip himself of as many 
real prejudices as he could ; I say real ones, because under the notion 
of prejudices, he knew to what dangerous heights some men have 
1 Davis rv. 27. 

'A. TALE OF A TUB' (l) 

proceeded. Thus prepared, he thought the numerous and gross 
corruptions in religion and learning might furnish matter for a 
satyr, that would be useful and diverting: He resolved to proceed 
in a manner, that should be altogether new, the world having been 
already too long nauseated with endless repetitions upon every sub- 
ject. The abuses in religion he proposed to set forth in the allegory 
of the coats, and the three brothers, which was to make up the body 
of the discourse. Those in learning he chose to introduce by way of 
digressions. 1 

The danger in this account is that it encourages one to split the 
book into parts and to search for coherence in what qualities the 
objects of his satire possess together. But since Swift is ridiculing 
the alternatives to the good life, his targets are wildly various; 
and the search would only take us into a jungle. 

The convenient frame is the limitation of Swift's era. If we con- 
sider the possibilities available to religious men, for instance, in 
16965 we may say that those who rejected the Church of England 
could be either Roman Catholics or Dissenters ; and if Dissenters, 
they were most likely to be Presbyterian. They might also belong 
to other Protestant sects: Independents (or Congregationalists), 
Baptists, Anabaptists, Quakers, and so forth ; or they might doubt 
the divinity of Christ and be Socinians or deists. Nobody admit- 
ted to atheism (Tree- thinking'), and hardly anybody would pub- 
licly accept the name of deist ; in ordinary usage the type of great 
mystic, Boehme, was classified as a maniac. Though most of 
these groups felt small respect for one another, the Calvinist sects 
and the Roman Catholic Church were normally conceived as the 
most extreme opposites along the same axis. While the Roman 
Catholics made the bulk of the people in Ireland, they were as 
powerless there as in England, where they constituted a docile 
minority. Among the sects which acted as dangerous, expanding 
rivals of the Established Church, the Presbyterians (who were 
usually the ones intended by 'nonconformists' or 'dissenters') 
were easily the strongest. 

In politics the situation seemed rather less knotty. Absolutism, 

1 Tale, p. 4. 



as illustrated by Louis XIV, could have no open supporters in 
England (no matter how many statesman accepted French 
bribes)., least of all when the two nations were at war; the Jaco- 
bites, for self-evident reasons, had to veil their longing for the 
'abdicated' king. But the Commonwealth was a fresh memory, 
especially to Swift, whose Antrim had been smothered in Crom- 
wellians. In Swift's experience, therefore, a man who disliked the 
constitution of 1688-9 would look much more menacing if he 
criticized it by republican principles than by regrets over the 
death of divine right. Since open advocacy of a republican con- 
stitution was not possible, such tendencies took the form of 
*new-Whiggish' political manoeuvres, descended from the con- 
troversy over the Exclusion Bill. An older body opposed to the 
court was the 'country party', or the grouping of landed gentry 
fearful of any demand that might increase the taxes upon their 

In literature and learning, one division was philosophical. 
Here, Swift assumed that the central position belonged to the ac- 
cepted moralists, from Plutarch to Montaigne, who warned men 
against the frailty of their nature and praised the stern but 
humble pursuit of duty. Swift was able to celebrate the same vir- 
tues in Socrates as in Sir Thomas More, in Cato as in 'King 
Charles the Martyr'. To this plain sort of moral 'philosophy' was 
opposed either the old scholasticism or the new systems of Des- 
cartes and Hobbes ; and in a lunatic fringe were to be found a 
procession of quack sciences (alchemy, astrology), hopeless re- 
searches (the longitude, squaring the circle, the philosopher's 
stone), occult studies (numerology, the cabbala, rosicrucianism, 
the work of Paracelsus). Both scholastic metaphysics and the 
modern (but outmoded) systems of Hobbes and Descartes ap- 
peared, to Swift, useless speculations beside the irrefutable 
validity of moral wisdom. The ethics of Plato (i.e., the early 
dialogues), Epicurus, and Zeixo deserved respect in so far as they 
anticipated Christian doctrine; but neo-platonism, extreme 
stoicism, epicurean physics, all belonged to another department 
and had more significance as illustrating folly than as teaching 

C A TALE OF A TUB 5 (l) 

virtue. Scholasticism of course possessed few defenders by this 
date. Long before Bacon, it had been the fashionable butt of 
humanist sneers. 1 But the geometrical 'method' of Descartes and 
the parallel, mechanistic scheme of Hobbes (whose critique of 
scholasticism Swift accepted) had admirers. Nevertheless, their 
reduction of the world and of mankind to soulless machines, their 
loud boasts of great accomplishments by easy devices, their dis- 
missal of earlier philosophy as futile, their promising panaceas 
while supplying little to help man's condition such appear- 
ances made it necessary for Swift to reject their work. As for the 
virtuosi and amateur experimenters the Royal Society and the 
Dublin Philosophical Society not only did they fall into the 
same class; but if Swift had any knowledge of his university 
teachers' extracurricular studies, the very pedants who expound- 
ed Aristotle's physics and logic must have seemed the first con- 
verts to the new (though now fading) obsession. 

In polite letters the things of worth were again plain to the dul- 
lest eye: Homer, Virgil, Horace, and the usual diet of classics; 
Moliere, Boileau, Jonson, Cowley, Milton, Temple, etc., among 
the moderns. Swift felt no inclination to limit his literary taste to 
ancient works; he enjoyed and recommended a healthy assort- 
ment of English, French, Spanish, and Italian authors. He must 
have agreed thoroughly with Temple's protest against *that 
which is called the authority of the ancients' ; for Temple wrote, 

I suppose authority may be reasonably allowed to the opinions of 
ancient men in the present age; but I know not why it should be so 
to those of men in general that lived in ages long since past ; nor why 
one age of the world should be wiser than another; or if it be, why 
it should not be rather the latter than the former ; as having the same 
advantage of the general experience of the world, that an old man 
has of the more particular experiments of life. 2 

Similarly, Temple, probably under St-Evremond's influence, 
wrote a panegyric on English dramatic humour which was 

1 Neither A Tale of a Tub nor The Battle of the Books was an attack on Bacon or on 
Baconian tradition. 
* Miscellanea I. 68. 



echoed by Congreve : 'There is no vein of that sort either antient 
or modern', said Temple, "which excels or equals the humor of 
our plays.' 1 It is true that this judgment, though accepted by 
Rymer and Dennis, was later qualified by Swift in favour of 'the 
Spaniards and Italians' 2 ; but the recognition of modern superi- 
ority remains unchanged. For Swift, the true enemies of the pro- 
ponents of good taste were under whatever comprehensive 
label pedantic critics, sensational journalists, and self-adver- 
tising poetasters : e.g., Bentley, L'Estrange, Blackmore, Wesley. 
When he grows sarcastic against 'moderns', it is such hacks and 
mercenaries that he means, and not modern authors generally. 
That these contemptibles had sometimes the insolence to find 
fault with the classics, or to argue that merely being modern gave 
them advantages over the ancients, in no way touched the merits 
of a Cervantes or a Congreve. 

We should remark Swift's great unconventionally in one re- 
spect, the weighting of his ridicule. Of three standard subjects for 
satire women, the law, and medicine he gave hints, but no 
more. With physicians and courts of justice, his experience, 
whether personal or second-hand, was modest; he had as yet no 
aspirations tied up in them, no disappointments sad enough to 
curse them for. Out of his inbred modesty, he denied himself the 
usual style of anti-feminine campaign; for he had not yet created 
his own expression of that theme. Anyhow, as he said, he was 

to proceed in a manner, that should be altogether new, the world 
having been already too long nauseated with endless repetitions 
upon every subject 3 

and so avoided the topics most commonly affected. There is one 
further motif which in a more particular way singles itself out by 
its absence; this is the parvenu. Except in an episode of 'Section 
n' 3 where the three brothers, freshly come up from their late 
father's country home, try to copy city fashions, there is hardly a 

1 Of Poetry, p. 337. See Spingarn i. Iviii-lxiii, m. 31 1. 2 Davis xn, 33. 
a Tab, p. 4- 


C A TALE OF A TUB 5 (l) 

stab at the newly rich who pretend to be aristocrats. Plenty of 
middle-class traits are tumbled about : the narrow education ; the 
hypocrisy; above all, the dissenting creeds. Yet the social climber 
of Moliere and Congreve gets barely a nod. 1 It may be another 
case of recognizing and therefore rejecting a coin too current to 
have character; or it may have been a target so close that he 
might cut himself with the knife he threw at others. He was, 
many years later, to say (misleadingly), 

All my endeavours, from a boy, to distinguish myself, were only for 
want of a great title and fortune, that I might be used like a lord by 
those who have an opinion of my parts whether right or wrong, it 
is no great matter, and so the reputation of wit or great learning 
does the office of a blue ribbon, or of a coach and six horses. 2 

To appreciate how false this calculated misrepresentation is, one 
must add up the number of times Swift let his talents and in- 
tegrity spoil his career. But his saying it first, indicates how alert 
he was to the chance of others' saying it. 

Having briefly surveyed the positives and negatives of A Tale of 
a Tub, we may turn to its occasion. Though there is no direct evi- 
dence on this problem, most scholars assume that the heaviest 
impulse behind the composition of the satire was Swift's year at 
Kilroot. The most bitter and extensive ridicule in A Tale of a Tub 
is that devoted to the Dissenters; and neither Moor Park nor 
London could have provided Swift with the frightening hordes of 
them which he met in Ulster. The long removal from Temple's 
presence may have enabled Swift to recognize his congenital 
leanings toward comic irony and to exercise them in essays 
which he did not preserve. Returning to Moor Park and the sight 
of literary eminence, as well as visiting London and receiving 
the stimulus of conversations there, Swift may have been im- 
pelled to make a fresh assault upon Parnassus. 

Besides such currents, the only occasion discoverable which 
has any concreteness is Temple's fight with Wotton and Bentley. 
William Wotton had courteously taken issue, in print, with 

1 The nod occurs in the clothes satire of section n. * Ball rv. 78. 


Temple's essay on Ancient and Modern Learning ; Richard Bentley 
had added, to a second edition of Wotton's book, an ungloved 
attack on Temple for praising the Epistles of Pkalaris; and the 
Hon. Charles Boyle, later Earl of Orrery, had lent his name to an 
elaborate riposte, ridiculing Bentley more than Wotton. Ac- 
cording to Swift's prejudiced account, Wotton had 

in a way not to be pardon'd, drawn his pen against a certain great 
man then alive [i.e., Temple], and universally reverenced for every 
good quality that could possibly enter into the composition of the 
most accomplish* d person; it was observed, how he was pleased 
and affected to have that noble writer calPd his adversary, and it 
was a point of satyr well directed, for I have been told, Sir W.T. 
was sufficiently mortify'd at the term. All the men of wit and polite- 
ness were immediately up in arms, through indignation, which pre- 
vailed over their contempt, by the consequences they apprehended 
from such an example . . . till my Lord Orrery had a little laid the 
spirit, and settled the ferment. But his lordship being principally 
engaged with another antagonist [i.e., Bentley], it was thought 
necessary in order to quiet the minds of men, that this opposer 
should receive a reprimand . . . and the author [i.e., Swift] was 
farther at the pains to insert one or two remarks on him in [A Tale 
of a 

If the satire on Dissenters was immediately inspired by Kilroot, 
the satire on Corruptions in learning' might be similarly due to 
the Pkalaris affair; for Bentley and Wotton take more punish- 
ment than any other persons referred to in A Tale of a Tub. 

But the literary appeal of the book resides in its armoury of 
satirical techniques. The power, vigour, and joy which Swift 
communicates flow from the abundance of his resources. He 
moves so fast from one device to another, employs each with such 
skill, combines them so suddenly, and exhibits such manifest 
pleasure in his own motions, that the reader can only share his de- 
light. The fundamental method of the satire and the irony is for 
Swift to make believe that he possesses the faults which he means 
to attack, and that he detests the qualities which he means to 
praise. By exaggerating and exemplifying the 'corruptions' as he 

1 Tale, pp. ii 12* 



recommends them, he of course makes them absurd. Most of the 
people whom he satirizes are writers of one sort or another; so 
Swift can, as it were, act out their vices for us through parodies of 
their work. When he cannot directly bestow a fault upon "him- 
self, he praises it. The result is to bewilder anyone who tries to 
find a consistency in the character of the 'author*. As he switches 
masks from Presbyterian to Roman Catholic, from schoolman to 
Cartesian, from pedant to poetaster, from republican to abso- 
lutist, he in effect repudiates such coherence. If one should suc- 
ceed in isolating some threads of uniformity, they would have 
to be either genuine traits of Swift himself or else qualities so 
vague as to be detached from the workings of this book. Either 
way, they become irrelevant to Swift's literary art. At the source 
of his incandescence there is not a consistent persona but an ironi- 
cal pose, which wins its literary effect only to the degree that it is 
seen through. 

Swift's more subtle and original method of ridicule is to treat 
each object as if it were its opposite. Swift can do this because 
both are false alternatives to his honest recommendations. 
Robert Boyle, measuring the elasticity of air, may feel contempt 
for a mad astrologer; but to a man convinced that moral philo- 
sophy is the only kind worth studying, the virtuoso seems as fool- 
ish as the quack. Similarly, the Puritan may loathe the papist as 
an idolater of anti-Christ; yet the Anglican not only spurns them 
both as distractions from true religion, but he also charges them 
with being in league. 1 Swift will therefore 'praise 3 the scientist for 
being a brilliant quack, or 'recommend 3 Calvin as the best guide 
to Rome. The modern reader has difficulty telling the opposites 
from the parallels, and without an annotated edition he is sure 
to go astray* However, Swift's contemporaries, familiar with 
other satires against Puritan enthusiasm, met little trouble of this 
sort. 2 Even today, anyone acquainted only with the plays ofjbn- 

1 W. P. Holden, Anti-Puritan Satire (New Haven 1954), p. 61. 

8 See Clarence M. Webster, 'Swift's Tale of a Tub compared with Earlier Satires 
of the Puritans', PMLA., XLVH (1932), 171-8; and 'Swift and Some Earlier 
Satirists of Puritan Enthusiasm*, PMLA., XLvra (1933), 1141-53; also Holden, 



son or with Butler's Hudibras should find the clusters of associated 
traits which are damned in A Tale of a Tub at least slightly fami- 
liar. Temple seems to make such links automatically : 

I have always look'd upon alchymy in natural philosophy, to be 
like enthusiasm in divinity, and to have troubled the world much 
to the same purpose. And I should as soon fall into the study of the 
Rosycrucian philosophy, and expect to meet a nymph or a sylph, for a 
wife or a mistress, as with the elixir for my health, or philosophers 
stone for my fortune. 1 

Malvolio already has the combination of arrogance, hypocrisy, 
lechery, credulity, superstition, avarice, and commercial inter- 
ests which Swift includes among the attributes of the Puritans. 2 

The most elusive but powerful of Swift's methods is his revers- 
ing the normal connotations of the 'polarized' kind of imagery 
which he had employed in his early poems. Emile Pons has 
shown how near the metaphors of A Tale of a Tub are to those of 
the odes. 3 The difference is that while in the odes they were in- 
tended to be sincere and to support the poet's argument, in the 
satire they are intended to be ironical and to undermine the 
'author'. Illusion and truth, appearance and reality, vapour and 
solid, light and darkness, hero and crowd, upward flight and 
sudden fall, these are applied just where they do not belong. The 
vapour which means spirit becomes the vapour which means 
flatulence; the light which illuminates becomes the phosphores- 
cence of rotting wood in the dark. The intangible is treated as 
tangible. This is Hobbes's sceptical method of reducing ideals 
to delusions; and Swift, whose flavour is remarkably close to 
Hobbes's, may have learned it from him. However, as Empson 
says, 'the language plays into [Swift's] hands' because 'the 
spiritual words are all derived from physical metaphors'. 4 

But this level of his technique also reflects the deepest moral 

1 Some Thoughts upon Reviewing the Essay of Antient and Modern Learning, in Miscel- 
lanea m. 255. 

a Gf. the cluster of people in Eachard's Free and Impartial Inquiry, 1673. 

8 Swift, Les Annies dejeunesse et le "Conte du tonneau" (Strasbourg 1925), pp. 246-7, 
325-6, and passim; cf. Davis t. xvi. 

4 Some Varieties of Pastoral, 1935, p. 60. 

C A TALE OF A TUB 3 (l) 

implication of Swift's comic irony, a principle which he shares 
with St Augustine : that inter wrinas et faeces nascimur,' 1 that the seat 
of love is the foulest place in our body, that our most exalted, 
most spiritual aspirations are bound to our soiled flesh. As Mon- 
taigne says, e Man is the onely creature, whose wants offends his 
owne fellowes, and he alone that in naturall actions must with- 
draw and sequester himselfe from those of his owne kinde. . . 
Whereas in other creatures, there is nothing but we love, and 
pleaseth our senses : so that even from their excrements and or- 
dure, we draw not only dainties to eat, but our richest ornaments 
and perfumes.' 3 Swift's repeatedly forcing this truth on our atten- 
tion suggests that it frightened him more than it does most (but 
not all) men; yet Swift was perfectly capable of keeping himself 
clean and falling in love. Temple, who was also fascinated by the 
imagery of air and vapours, 3 worried less about cleanliness and 
enjoyed a more conventional family life; but he never faced this 
paradox. Swift's warning is not to confuse the intangible with the 
good, or the arms with evil. For purposes of invective it may be 
proper to use such associations (and Swift rarely failed to do so) ; 
but in religion and morality it is disastrous to believe either that 
refining a sin makes it a blessing, or that the humble origin of a 
noble impulse makes it vicious. If one falls into the mud through 
staring at the heavens, the way to keep on one's path is not to 
darken the sky. 

As the framework for these patterns, Swift often adopts the 
rhetorical schemes taught him at the university: the oration, the 
sermon, the formal praise or dispraise, the demonstration or dis- 
proof. In academic rhetoric, praise and dispraise were treated as 
equivalent: one ran through identical heads, consistently ap- 
proving or condemning the facts marshalled under each : as St 

1 For all his gross errors of fact, attribution, and interpretation, Professor Norman 
O. Brown has, on this limited subject, something important to say, in his Life 
against Death, 1959, pp. 186-8. 

2 Florio's translation, ed. 1908, n. 222-3; see also K. Williams, p. 150. 

s Temple's preoccupation with air and vapours, his referring of human character 
to the state of the 'finer spirits of the brain* or to the 'fumes of indigestion', and 
similar chains of imagery are illustrated in his Miscellanea I. 45-6 and Observations 
upon the U.P., pp. 125-7, 186-7. 



George Ashe said, 'All things are capable of abuse from the same 
topics by which they may be commended.' 1 This is one reason for 
Swift's power: he was accustomed to follow the same plan for 
opposite ends; to be ironical, he had only to change expected 
positives into negatives. By a reverse form of the same device, one 
may treat a noble character as a subject for either a direct attack 
or meiosis ; in the dedication to Somers, Swift gives this treatment 
to his lordship. In rhetorical tradition, the device is an ancient 
one, going back to Isocrates' eulogy of Busiris, the wicked king of 

Syllogistic demonstration had an analogous effect: it was 
taught purely as method, detached from the truth or falsity of the 
result. One recalls that in the final exercises for the B. A. degree at 
Trinity College, each candidate had to write twenty-four syl- 
logisms on the wrong side of each question but only twelve on the 
right. To be ironical, Swift therefore goes through the gestures of 
logic, but gives such absurd reasons and reaches such absurd con- 
clusions that the proposition he offers to prove becomes incred- 
ible. 2 Contrariwise, he will pretend to explode a proposition 
which in reality is one of his rules of life; but again, though em- 
ploying all the machinery of logic, he operates it on glaring fal- 
lacies, and strengthens the case which was ostensibly to be des- 
troyed. Thus he can make great play with the school distinction 
between accidental and essential properties, especially in ex- 
amining the central problem of the satire the nature of man. In 
the logic handbooks, for instance, it was regularly explained that 
man was vestitus by accident but reasonable by essential nature. 
Swift merely treats dress as essential and reveals the degree to 
which we are (though we ought not to be) our clothes 3 ; or he 
treats unreason (i.e., madness), rather than reason, as essential to 
humanity, and unmasks the vulgar conception of what a great 
man is. 4 

Another framework is the type of fable or anecdote which be- 

i See Appendix D. 

8 Cf. John M. Bullitt, Jonathan Swtft and the Anatomy of Satire, 1 953, pp. 1 16-22. 

8 Section n. 4 Section ix. 


gins to appear in the poem, to Gongreve. Swift has little gift for 
narrative plot; and he changes this defect into a satirical device 
by substituting bathos for climax at the turning-point of an epi- 
sode. Instead of a dramatic struggle followed by triumph or 
tragedy, we get an impasse or an anticlimax; any action is hur- 
riedly summarized, but the enveloping commentary, and the 
speeches, are extended. Swift's power lies in the extraction of 
many meanings from the juxtaposition of contrasting figures^ 
usually in a static but tense tug-of-war ; so his fables succeed best 
either when they are short or when they comprise a series of such 
evocative but quick juxtapositions filled out by speeches (usually 
one to a side) , mock- moralizing, and commentary. In A Tale of a 
Tub the fable of the three brothers, though spread over four sec- 
tions (ir, iv, vi, xi), will be found to contain little sequential nar- 
rative but many discrete sub-fables, static speeches (i.e., not con- 
troversy progressing dialectically), and authorial exegesis. In 
these occupations, Swift's ear for talk is a formidable talent, and 
supplies an excitement which makes up for the lack of dramatic 
suspense. The little anecdote in the preface to A Tale of a Tub is a 
good specimen; it is of course a parable concerning the rivalry of 
writers for public attention : 

A mountebank in Leicester-Fields, had drawn a huge assembly 
about him. Among the rest, a fat unweildy fellow, half stifled in the 
press, would be every fit crying out, Lord ! what a filthy crowd is 
here; Pray, good people, give way a little, Bless me! what a devil 

has rak*d this rabble together: Z ds, what squeezing is this! 

Honest friend, remove your elbow. At last, a weaver that stood 
next him could hold no longer: A plague confound you (said he) 
for an over-grown sloven; and who (in the devil's name) I wonder, 
helps to make up the crowd half so much as your self? Don't you 
consider (with a pox) that you take up more room with that carkass 
than any five here ? Is not the place as free for us as for you ? Bring 
your own guts to a reasonable compass (and be d n'd) and then 
1*11 engage we shall have room enough for us all. 1 

1 Tale, p. 46. 




The attitude behind the Tale, then, is intensely moral. The book 
is a criticism of men, not of theologies or tastes. Swift means to 
contrast two types of behaviour; the symptoms by which he 
identifies the wrong type are chosen from aspects of life familiar 
to him but not already exhausted as material for comedy. He is 
therefore persuaded that any one of these streaks will serve to 
diagnose the same underlying disease, and each therefore stands 
for the others. 'Modernism', properly defined, shows links with 
atheism; an epicurean philosopher will make a bad literary 
critic ; for all originate in one failing : the dismissal of orthodox 
goods (Anglicanism, humane letters, reason) and their replace- 
ment by eccentric satisfactions (pedantry, superstition, enthusi- 
asm). This error is too gross to arise from self-interest; Swift 
therefore attributes it to ignorance or madness. 

What makes the Tale bewildering is of course that Swift does 
not put his real merchandise before the reader. It must be re- 
fined from the ironies which he marshals instead. Since he is at- 
tacking tendencies and not products, he can use a single example 
in several contexts, protecting it when it has benign implica- 
tions, destroying it when it has others. Through his centrifugal 
symbolism, each motif can be transformed into any of the rest. 
They are riot even limited by chronology, for human nature is 
stable beneath the changes of civilization : our essential character, 
as Temple said, 'seems to be the same in all times and places'. 1 
So Jack blends with Peter, Tigellius with Bentley, critic with 
fanatic. But Homer as an epic poet is differentiated from Homer 
as a reputed seer, the statesman as patriot from the statesman as 
egotist, erudition, as wisdom from erudition as pedantry. 

It is correct to read every part of the Tale as an adaptation of 
one attitude : that wilful rejection of the Established Church, 
limited monarchy, classical literary standards, and rational 
judgment is an act of pride, and leads to corruptions in govern- 
ment, religion, and learning. The book would then be a collec- 

1 An Essay upon the Original and Nature of Government, opening sentence ( Mis- 
cellanea i. 45). 


6 A TALE OF A TUB* (l) 

tion of essays treating, from this point of view, several of the end- 
less varieties of corruption. Ironically, the author pretends to ad- 
mire his exhibits. He is so sure of his own creed that he assumes 
intelligent readers share it and will not fail to penetrate his plain 

Under the stunning complications of its surface, the Talehas an 
intellectual symmetry. Everywhere in it, Swift is preaching the 
same lesson. His countrymen's lives should possess the virtue and 
cultivation of Graeco-Roman antiquity, improved by Christian 
ethics and a cheering hope of salvation. To put themselves in this 
condition, they had only to accept the doctrines of the Church of 
England and the political constitution of 1688, to let their under- 
standing instruct them in decisions outside the province of reve- 
lation, and to form their taste from the monuments of unageing 
intellect. Instead of doing so, they swarmed after crazy sectarians, 
threw the reins to their passions, and either ignored the classics or 
anatomized them to dust. Where order belonged, they estab- 
lished chaos. 


Chapter Two 

To see how the full apparatus of the satire actually works, we 
may look at three of its sections, beginning with a rather 
straightforward one, A Digression concerning Criticks. Before 
taking even this up in detail, however, we should remark that the 
whole intricate pattern of preliminaries and "sections' in A Tale 
of a Tub amounts to a general parody of the elaborate arrange- 
ments often bestowed in the seventeenth century upon books 
whose content did not deserve them; it embodies the spirit of La 
Bruyere's epigram, 'Si Ton ote de beaucoup d'ouvrages de mo- 
rale Tavertissement au lecteur, Fdpitre d^dicatoire, la preface, la 
table, les approbations, il reste a peine assez de pages pour meri- 
ter le nom de livre.' 1 Through the superficial formality of his pat- 
tern, Swift encourages the reader to look for a cleverly knit argu- 
ment, advancing with the rhythm of reason to a set of proposi- 
tions evolved from it if not proved by it. Nothing of the sort is 
forthcoming. The Digression in the Modern Kind might easily be 
moved to the position of c Section x' ; the Conclusion delivers no 
conclusions; 'Section xi 3 contains an apparently impertinent 
study of ears ; the text is irregularly broken by files of asterisks 
where we expect crucial transitions. An object of this game is to 
show up the excessive elaboration of framework in much polite 
literature of the age. But such structural parody glances more 
specially at Dryden's Virgil, with its texts of the great poems set 
adrift among commendatory verses, separate dedications, notes, 
observations, a life of Virgil, a preface to the Pastorals, an essay on 
the Georgics, and a postscript. Similarly, the business of presenting 

1 De VEsprit, no. 6. 


C A TALE OF A TUB* (ll) 

the expository parts of A Tale of a Tub as e digressions' probably 
points at Fontenelle's 'Digression* sur les Anciens et les Modernes> in 
his Poesies Pastorales (1688). 

However, the general ideas which are themselves defended in 
the digression on critics can be traced to Temple's Some Thoughts 
upon Reviewing the Essay of Antient and Modern Learning and to Dr. 
Bentlefs Dissertation Examined (published over the name of Charles 
Boyle and therefore called 'Boyle's' Examination, though largely 
prepared by Francis Atterbury). 1 The objects attacked are 
identified mainly with Richard Bentley, William Wotton, and 
their writings ; yet Wotton here is a thoroughly subsidiary figure, 
receiving most of his punishment elsewhere within the Tale. As 
the fundamental method of his satire. Swift uses a parody of the 
polemical and erudite manner adopted by Bentley in the Dis- 
sertation upon the Epistles ofPhalaris. 2 Finally, the internal structure 
of the digression is that of a formal speech, part demonstration, 
part eulogy. 

Several of Swift's minor devices appear here. One is to men- 
tion his true values briefly near either the outset or the end of a 
satire, and to dismiss them with a sneer. Thus he defines two ad- 
mirable senses of critic an impartial judge of literature and a 
textual editor but says they are both "utterly extinct'. 8 What he 
does in the remainder of the 'digression' is to pretend he admires 
the kind of pedantry which he attributes to Bentley and Wotton. 
For this project, the main device is the animal metaphor which 
had found only casual expression in the poems. Now Swift makes 
capital of a phrase used in the Boyle-Berxtley exchange, and pre- 
tends that the ass, in antiquity, was employed as a symbol of the 
'true' critic. 4 Since he explores the rather unsubtle implications 
of this theme through a parody of Bentley's own philological 
proofs, his success is irresistible. 

The exhilarating texture of the section depends more on 
Swift's wild, punning conceits than on any other ingredient; as a 

1 See Temple, Miscellanea ra. 256-60 ; Boyle's Examination, pp. 224-7. 

2 See Miriam K. Starkman, Swift's Satire on Learning in A Tale of a Tub (Princeton 
I950). PP- 101-4. 

8 Tale> p. 93. 4 Boyle's Examination, p. 220; Starkman, pp. 103-4. 



device of comic irony, these have an almost generous flavour : the 
reader feels that the author is grateful to his objects for giving 
him such happy occasions to exercise his intellectual muscles. 
Here, for example, is a figure in which Swift compares critics with 
deadly substances which never change their natures; the pun- 
ning close depends on the fact that hashish and a gallows rope are 
both derived from the hemp plant : 

For it hath been observed both among antients and moderns, that a 
true critick hath one quality in common with a whore and an alder- 
man, never to change his title or his nature; that a grey critick has 
been certainly a green one, the perfections and acquirements of his 
age being only the improved talents of his youth; like hemp, which 
some naturalists inform us, is bad for suffocations, tho' taken but in 
the seed. 1 

Another example, almost adolescent in its rawness, is the ex- 
planation of why critics are like ancient mirrors; here the joke 
hinges on the moral connotations of 'brass' and on the use of 
'mercury' (which was of course the backing of glass mirrors), in 
Swift's day, to signify wit : 

Now, whoever considers, that the mirrors of the antients were made 
of brass, and sine mercurio, may presently apply the two principal 
qualifications of a true modern critick, and consequently, must needs 
conclude, that these have always been, and must be for ever the 
same. For, brass is an emblem of duration, and when it is skilfully 
burnished, will cast reflections [i.e., aspersions] from its own super- 
fates, without any assistance of mercury from behind. 2 

Swift's tone suggests the cocksure arrogance of one who has 
taken over wholesale the opinions of others. The quarrel is ex- 
pressly with 'pedantry'. Temple said he had e no mind to enter 
the list, with such a mean, dull, unmannerly pedant'. 3 What may 
trouble the modern reader is the confusion, all around, between 
manners and morals. In his attack upon pedantry as exemplified 
by some university dons, 4 Swift was following the consensus of 

1 Tale, p. 1 01. a Ibid., p. 103, 8 Ibid., p. xlviii. 

4 It would be wrong to describe the Temple-Boyle-Swift campaign as either an 


e A TALE OF A TUB* (ll) 

thoughtful men, from Montaigne to Locke. When vulgarized 
among polite gentlemen in the late seventeenth century, the atti- 
tude assumed the form of a 'fashionable flight from learning'. It 
was, said John Norris in 1678, c a piece of errant pedantry, and defect 
of good breeding to start any question of learning in company' ; 
while, said Norris, the fellows of the Royal Society might im- 
prove in learning, c in point of civility they decline 5 . As Boyle put 
it, 'The first and surest mark of a pedant is, to write without ob- 
serving the receiv'd rules of civility and common decency : and 
without distinguishing the characters of those he writes to, or 
against : For pedantry in the pen is what clownishness is in con- 
versation; it is written illbreeding.* Pedantry thus becomes a mat- 
ter of manners; and for this invidious distinction anyone not a 
gentleman by birth is implicitly at a disadvantage; Bentley's 
fault is to be an upstart. Swift was later to fling contempt upon 
such notions, but in 1696 he agreed with Temple, who had as- 
signed much of the blame for the decline of true learning to 

the scorn of pedantry, which the shallow, the superficial, and the 
sufficient among scholars, first drew upon themselves, and very 
justly, by pretending to more than they had, or to more esteem, 
than what they [could] deserve; by broaching it in all places, at all 
times, upon all occasions ; and by living so much among themselves, 
or in their closets and cells, as to make them unfit for all other busi- 
ness, and ridiculous in all other conversations. 1 

Bentley's hubris was admittedly colossal. He was the greatest 
classical scholar in England before the nineteenth century; but 
his erudition hardly excuses his offensive condescension. Yet 
though he gave ample provocation, and though the answers of 
both Swift and Atterbury ('Boyle') are so clever that their lack of 

attack upon or defence of the universities as such : Bentley, like Temple, was an 
ornament of Cambridge, and 'Boyle' regarded himself as the loyal son of Christ 
Church, Oxford (Examination, sigg. A4 V ~5). 

1 Temple, Miscellanea n. 71 ; W. E. Houghton, *The English Virtuoso in the 
Seventeenth Century', JHL 9 in (1942), 216-17; T. Scott xi. 49-57; Boyle's 
Examination, p. 93, and cf. p. 222. Wotton himself made remarks sympathetic with 
Temple's position (Reflections, pp. 415-16), and Boyle praised Wotton's book for *a 
vein of learning running through it, where there is no ostentation of it' (p. 25) . 



scholarship is irrelevant, the attacks have a hollowness which 
finally weakens their achievement, because the clergymen and 
the scholars really belong in the same party. For all his arrogance, 
moreover, Bentley stood solidly behind his Dissertation ; he did not 
pretend to be writing anything except a piece of scholarship, but 
classical philology was half his life ; not only are his facts sound : 
his deepest feelings are bound up in the work; it is his own ex- 
position of his original ideas. A Tale of a Tub was anonymous ; Dr 9 
Bentley 9 s Dissertation Examined was pseudonymous ; but everybody 
knew who was the author of both the Dissertation upon Phalaris and 
the Reflections upon Ancient and Modern Learning. 

What Swift resented was not in fact the kind of learning cor- 
rupt or pure which his victims possessed. Wotton had, he said, 
'in a way not to be pardon'd', drawn his pen against Temple ; c all 
the men of wit and politeness 9 were up in arms. 1 Yet the writers of 
Boyle's Examination say Wotton c is modest and decent, speaks 
generally with respect of those he differs from*. 2 Wotton actually 
fought, as a modern scholar says, 'in reasonable and accepted 
terms, and distinguished himself by the thoroughness and just- 
ness of his analysis. Although Swift engaged in a good deal of per- 
sonal diatribe against him, Wotton's ideas were perfectly ortho- 
dox.' 3 Temple, furthermore, never seems to have asked Swift 
to write against either Bentley or Wotton. 4 After all, a riposte 
had been published which Temple had approved; and Swift's 
parody of Bentley comes perilously near to being as well a parody 
of Boyle. One suspects that Swift's hopes were treading on his 
sense of reality, and that he felt eager to render a service which no 
one had called for, because while the talents of the participants 
were inferior to his own, their prestige was much higher. Who, 
after all, was an obscure Irish parson to put himself in the place 
of the distinguished baronet and the heir of a great earl ? Was not 
Swift showing that he was a better ally for Temple than the 
Christ Church wits ? that a Wotton in Swift's clothing could out- 
lord the Hon. Charles Boyle? 

1 Tale, pp. ii-ia. * Ibid., p. n, n. 2. 8 Starkman, pp. 15-16. 
4 Tale > p. xlviii* 


C A TALE OF A TUB' (ll) 

At the heart of the digression on critics one finds little that 
Swift really stood for. The arrogance which he denounced ap- 
pears more in him than Wotton (who is a peripheral figure in this 
digression but a focus of the ridicule elsewhere). Swift funda- 
mentally respected learning of the sort which Bentley exemplifies. 
Neither victim could be described as a hack pamphleteer or a 
poetaster. Swift's attack on Wotton (a parson his own age) was 
not for pedantry but for daring to criticize Temple, however 
courteously; and Swift may be said to have borrowed his emo- 
tion. The wit, for all its fanciful exuberance, is narrow in con- 
notation and lacks the rich cross-references to religion and to the 
human body that will be found in other sections; it touches no 
moral depths. The comic thrusts are skinny word-play, lacking 
'humour 5 . The c true critic' is indeed apotheosized so that the link 
with Bentley becomes a shadow; the whole essay is a dazzling 
performance ; yet these admirable transmogrifications and pyro- 
technics are not what we mean by Swift. 


The remarkable violence of Swift's irony is not only due to his 
'polarized' imagery and to his playing foul against fair with the 
human body, but also to what might be called his nettle-grasping 
instinct. He often writes as though he were disproving the in- 
sinuation that he feels uneasy about his subject. It is as if he wish- 
ed to appear so sure of his own piety, his continence, his modesty, 
his rectitude, that common prudential maxims should not apply 
to him. In the holiest places he could affect to sound coarse not 
out of irreverence but as a demonstration that his faith was too 
serene for any such expressions to impugn it. What he fails to an- 
ticipate is that other worshippers may hear the coarseness with- 
out appreciating the piety. 

This leaning belongs to the hypocrite-in-reverse character 
which is often assigned to Swift. Through it, he separates himself 
from those who substitute the show for the reality in religion. By 
caricaturing the postures of Tartufe, he does not ridicule the 



church but the hypocrisy. To an uninformed outsider, unfortu- 
nately, the two sorts of ridicule look alike. Similarly, in literary 
forms, the mock-heroic Swift does not attack the epic but the 
pretensions of men who hope to achieve greatness by putting on. 
the trappings of Achilles. To some outsiders, however, the parody 
seems an affront not to pretensions but to poetry. Yet the genius 
who assumes that his religion or his literary taste is unquestioned 
does not stoop to concern himself with their blunders. Hence the 
troubles of Moliere and Swift with their contemporaries. 

"Section xi 3 of A Tale of a Tub has some good illustrations of 
Swift's nettle-grasping. It is part of the satire upon Jack, the one 
of the three brothers in Swift's religious allegory who stands for 
Calvinism. In spite of the explosive effect of the satire, Swift, in 
his charges against the Puritans, is Tar less fierce and unrelenting' 
than many of his predecessors. His ultimate conception of Jack is, 
after all, 'more of a zealous fool, with no sense or reason in him 5 
than of a cunning, dangerous Tartufe. Furthermore, much of 
Swift's material was not very fresh : he said little that was new 
about the Puritan alone. Many of his most amusing thrusts were 
a century old when he wrote them.' 1 Even motifs which seem 
Characteristic 9 of Swift had been exploited by a train of writers 
including not only obvious names like Burton, Jonson, and But- 
ler but also Sir Thomas Browne, Meric Casaubon, Henry More, 
and John Eachard. Accusations that Puritans mistook flatulence 
for inspiration, that their zeal was a cover for sexual excitement, 
that they suffered from a martyr complex, that women were 
peculiarly susceptible to enthusiasm were staples of such litera- 
ture. 2 

Swift's audacity in manipulating these motifs is what makes 
them his own. In a satire on predestination, for instance, he has 
Jack first bang his nose against a post and then claim that Trovi- 
dence thought fit' to arrange this collision, and again that 'Provi- 
dence either forgot, or did not think it convenient to twitch me 

1 Webster, in PMLA., XLVH. 177-8. 

2 Webster, ibid., XLVHI. 1141-53; see also William P. Holden, Anti-Puritan Satire 
(New Haven 1954), pp. 40-3 and passim. One burlesque of Puritan sermons was 
entitled A Tale in a Tub (John Taylor, 1643; see Holden, pp. 73-5). 


S A TALE OF A TUB 5 (ll) 

by the elbow, and give me notice to avoid it.' One wonders how 
Swift could have been puzzled to find Wotton describing such 
phrases, delivered by an ordained clergyman, as c a direct 
prophanation of the majesty of God 3 . Later, Swift himself con- 
fessed the rashness of his involving the deity so directly in a piece 
of broad ridicule, when he altered 'Providence' to 'Nature' and 
'Fortune 5 . 1 

Not only does Swift use words which in an absolute sense ap- 
pear shocking, but his similes have cross-references to other bold 
implications of his argument. Although, therefore, "Section xi' 
possesses less fluency and a less regular organization than the di- 
gression on critics, its imagery has more depth and the whole 
essay more power. In Jack, Swift treats the doctrines, practices, 
and evolution of the Calvinists, employing far-fetched tropes 
which would nevertheless be easily penetrated by an audience 
familiar with the theme. The most bizarre of these tropes is a his- 
tory of ears which breaks into the already rough-hewn tale of 
Jack. 2 This operates analogously to the c ass s in the digression on 

As Swift develops the c ear 5 figure, he suggests two funda- 
mental connotations of Puritan zeal: political rebellion and 
sexual excitement. Ear-clipping was a penalty for certain crimes ; 
and the Puritans, by cutting their hair short, gave themselves 
round heads and prominent ears. Swift can therefore suggest, 
through the rise and fall of ears, the Civil War and the Restora- 
tion. But he also draws a parallel between ears above and genitals 
below; by such displacement he suggests that in their avowed de- 
sire for spiritual light, the Puritans were really looking for erotic 
pleasure. A weaker but distinct innuendo is that the dissenting 
minister's overwrought, sensational preaching addressed to the 
'ears' of the congregation was a substitute for true piety ration- 
ally expounded. Finally, there is a reference to philosophy; for 
Swift emphasizes the eagerness of the ignorant Puritan worship- 

1 Tale, p. 193 and n, i ; p. 324. 

* Cf. the faces pictured on the title page of Heads qf All Fashions, reproduced in 
Holden, p, 62. 

P ^Il] 


pers to hear the minister's wild speculation; their idle a dangerous 
curiosity, undermining established wisdom, is the kind of hunger 
which maintains the systems of the scholastics and the Cartesians 

9 Tis true, indeed, that while this island of ours, was under the 
dominion of grace, many endeavours were made to improve the 
growth of ears once more among us. The proportion of largeness, 
was not only lookt upon as an ornament of the outward man, but as 
a type of grace in the inward. Besides, it is held by naturalists, that 
if there be a protruberancy of parts in the superiour region of the 
body, as in the ears and nose> there must be a parity also in the in- 
Jerior : And therefore in that truly pious age, the males in every as- 
sembly, according as they were gifted, appeared very forward in 
exposing their ears to view, and the regions about them. 1 

One of the happiest extensions of 'cross-referential* overtones is 
the treatment of avowed opposites as equivalents. Much as the 
Calvinists might fear the Established Church, it was Rome that 
loomed for them as the anti-Christ. Swift therefore has a fine pas- 
sage describing Jack as if he were Peter : 

Their lodgings were at the two most distant parts of the town, from 
each other; and whenever their occasions, or humors called them 
abroad, they would make choice of the oddest unlikely times, and 
most uncouth rounds they could invent; that they might be sure 
to avoid one another : Yet after all this, it was their perpetual for- 
tune to meet. The reason of which, is easy enough to apprehend: 
For, the phrenzy and the spleen of both, having the same founda- 
tion, we may look upon them as two pair of compasses, equally ex- 
tended, and the fixed foot of each, remaining in the same center; 
which, tho' moving contrary ways at first, will be sure to encounter 
somewhere or other in the circumference. 2 

In the abruptness with which he introduces such overtones,, 
the speed with which he circulates among them, and the naked- 
ness with which he exhibits them, Swift seems to be protesting 
too much that these things hold no terrors for him that he is not 
worried or tempted by political unrest, by lust, by emotional 

1 Tale, p. 201. * Ibid., pp. 198-9. 


6 A TALE OF A TUB* (ll) 

oratory, by Hobbes and Descartes. Both as a moralist and as a 
literary artist, Swift miscalculates; for however worthy his ideals 
may be, and however sinful his opponents, he never reckons with 
the fright which he may give to the reader. By their very success, 
his methods of communicating a violence of feeling distract one 
from Swift's values and his targets both : one looks instead at the 
author, and marvels at his audacity. 

Here (in part) is Swift's satire on Jack's obsession with Scrip- 
ture as the sole recourse in every function of life : 

He had a way of working it into any shape he pleased ; so that it 
served him for a night-cap when he went to bed, and for an um- 
brello in rainy weather. He would lap a piece of it about a sore toe, 
or when he had fits, burn two inches under his nose,* or if any thing 
lay heavy on his stomach, scrape off, and swallow as much of the 
powder as would lie on a silver penny, they were all infallible reme- 
dies. With analogy to these refinements, his common talk and con- 
versation ran wholly in the phrase of his Will [i.e., Scripture], and 
he circumscribed the utmost of his eloquence within that compass, 
not daring to let slip a syllable without authority from thence. Once 
at a strange house, he was suddenly taken short, upon an urgent 
juncture, whereon it may not be allowed too particularly to di- 
late; and being not able to call to mind, with that suddenness, the 
occasion required, an authentick phrase for demanding the way to 
the backside ; he chose rather as the more prudent course, to incur 
the penalty in such cases usually annexed. Neither was it possible 
for the united rhetorick of mankind to prevail with him to make 
himself clean again: Because having consulted the Will upon this 
emergency, he met with a passage near the bottom (whether foisted 
in by the transcriber, is not known) which seemed to forbid it. 1 

To digest such a satire without suspecting possible sneers at one's 
own beliefs, one would have had to be not only Protestant but 
Anglican, and not only Anglican but a high-churchman, and not 
only a high-churchman but a very special distruster of other Pro- 
testant sects. In 1688, comprehension or the inclusion of the 
Presbyterians in the Anglican communion had been *a high- 
church policy'; Bancroft himself (unlike Francis Turner) had 

pp. 190-1. 



been c a warm friend of members of the foreign Reformed church- 
es in England and a cordial advocate of comprehension at home*. 
On Swift's treatment of the Roman Catholics a parallel, if 
weaker, comment can be made; for although the period (largely 
war time) from 1685 to 1714 was one of open hostility between 
the Gallican and Anglican churches, a rapprochement of the 
two had been canvassed, before the revocation of the Edict of 
Nantes, by men like Bossuet and Wake ; and when Wake became 
Archbishop of Canterbury (1716), he was to revive the project; 
even during the war years, nonjuring Jacobites like Charles 
Leslie and Henry Dodwell sought to heal this schism. Similarly, 
while Swift seems to have set great store by the authority of the 
primitive church, the reputation of the pre-Nicene fathers was 
sinking quickly between 1680 and I73O. 1 

In theology as in. philology, Swift's note of confidence should 
not imply that all right-thinking folk stood behind him, but that 
he was wholeheartedly borrowing the standards of a small, 
powerful group. As a contrast, we may quote a statement made 
by Sir William Temple's old tutor, Cudworth, in 1674: Cer- 
tainly in. our English Church, just as in Noah's Ark were all sorts 
of animals (if I may so express it), are all kinds of Protestants : 
Calvinists, Remonstrants [i.e., Arminians], and I believe even 
Socinians, all dwelling here, united with no apparent discord in 
one and the same communion. * a In the face of such varied evi- 
dence, Swift's apology for the Tale can only be described as 
question-begging viz., 

Why should any clergyman of our church be angry to see the follies 
of fanaticism and superstition exposed, tho' in the most ridiculous 
manner ? since that is perhaps the most probable way to cure them, 
or at least to hinder them from farther spreading. Besides, tho' it 
was not intended for their perusal; it raillies nothing but what they 
preach against. It contains nothing to provoke them by the least 
scurillity upon their persons or their functions. It celebrates the 
Church of England as the most perfect of all others in discipline and 

1 Norman Sykes, From Sheldon to Seeker (Cambridge 1959), pp. 83-7, 1 14-32, 142 ; 
George Every, The High Church Party 1688-1718, 1956, pp. 70-3. 

* Rosalie L. Colie, Light and Enlightenment (Cambridge 1957), p. 40. 

e A TALE OF A TUB 5 (ll) 

doctrine, it advances no opinion they reject, nor condemns any they 
receive. 1 

Unless, in fact, one either agreed with the details of Swift's 
principles, or else felt disrespect for all churches, one could hardly 
read such a passage as that on Jack's addiction to Scripture, and 
not flinch. Furthermore, by deliberately narrowing the limits of 
a sympathetic audience within the community of Christians, 
Swift effectively enlarged it outside ; and people of moderate 
piety might easily question the religion of a priest who seemed to 
encourage scepticism. Thus he laid himself open to charges 
which he could not refute without weakening the very self- 
confidence which lay behind his provocations. 

1 Tale, p. 5. 

Chapter Three 

Bit the full force of Swift's abilities is only felt when he 
widens the angle of his satire to include everybody. The 
logic is that if Puritans, pedants, mechanical philoso- 
phers, and the rest are so numerous, they must represent corrup- 
tions which are incipient in all, and which Swift associates with 
irrationality. Normally, these are considered not essential proper- 
ties but diseases, and in that sense accidents. Our most essential 
property is, by contrast, said to be reason, or the opposite of mad- 
ness. Swift stands the traditional contrast on its head, making 
madness the essential property and reason the corruption. 1 To 
prove the paradox, he shows how much better we can account 
for human history and behaviour as the outcomes of madness 
than as the effect of reason ; and his demonstration is the material 
of A Digression concerning the Original) the Use and Improvement of 
Madness in a Commonwealth. 

This digression embodies an irony familiar to Swift's genera- 
tion and recognized throughout the eighteenth century: that the 
corruption of reason can be cured by reason alone. Our restless 
itch for knowledge, when it trains the reason upon proper objects, 
brings wisdom; but when it thrusts the reason upon objects out- 
side reason's province, brings delusion. As Temple said, 

The same faculty of reason, which gives mankind the great advan- 
tage and prerogative over the rest of the creation, seems to make the 
greatest default of humane nature ; and subjects it to more troubles, 
miseries, or at least disquiets of life, than any of its fellow creatures. 2 

1 Martin Price, Swtffs Rhetorical An (New Haven 1953), p. 93. 
a Miscellanea xx. 75. 


C A TALE OF A TUB 9 (ill) 

Yet we can be reclaimed from delusion by nothing but reason, 
'called in, to allay those disorders which it self had raised'. 1 In 
attributing both political and philosophical upheavals to insa- 
tiable, idle curiosity, operating by way of the corruption of the 
understanding, Swift is not original, either. Although the bond 
between irrationality and evil knaves and fools is one of his 
leitmotivs, it is an article which he shares with the longest line of 
Western moralists. 'This restless humor, so general and natural 
to mankind*, writes Temple, 

is a weed that grows in all soils, and under all climates, but seems 
to thrive most, and grow fastest, in the best; 'tis raised easier by the 
more sprightly wits and livelier imaginations, than by grosser and 
duller conceptions. . . From this original fountain issue those 
streams of faction, that with some course of time and accidents, 
overflow the wisest constitutions of governments and laws, and 
many times treat the best princes and truest patriots, like the worst 
tyrants and most seditious disturbers of their country. 2 

Thus the revolutions of governments and of intellectual systems 
are traced to the same source. Eighty years after Swift's satire, 
Gibbon, when excoriating the neo-Platonists, still nursed this 
tradition: in his version, the link is made between religious, 
philosophical, and civil disorders: 

By mistaking the true object of philosophy, their labours contri- 
buted much less to improve than to corrupt the human understand- 
ing. The knowledge that is suited to our situation and powers, the 
whole compass of moral, natural, and mathematical science, was 
neglected by the new Platonists, whilst they exhausted their 
strength in the verbal disputes of metaphysics, attempted to ex- 
plore the secrets of the invisible worlds. . . As they agreed with the 
Christians in a few mysterious points of faith, they attacked the re- 
mainder of their theological system with all the fury of civil war. 3 

Still later, we find Hazlitt, whose moral constitution shared 
several ingredients with Swift's, independently asserting the tra- 

1 Miscellanea n. 87. 

* Miscellanea m. 10-11 (Of Popular Discontents) \ cf. Tale, p. 169. 

* Decline and Fall, ed. Bury, 1900, 1. 392-3. 


ditioniiiits widest, most 'Swiftian 9 form, and tying the irrational 
to ultimate evil : 

There is a love of power in the mind independent of the love of good, 
and this love of power, when it comes to be opposed to the spirit of 
good, and is leagued with the spirit of evil to commit it with greedi- 
ness, is wickedness. . . This character implies the fiend at the bottom 
of it; and is mixed up pretty plentifully (according to my philo- 
sophy) in the untoward composition of human nature. It is this 
craving after what is prohibited, and the force of contrast adding 
its zest to the violations of reason and propriety, that accounts for 
the excesses of pride, of cruelty, and lust. 1 

In Hazlitt's formulation, the principle becomes so general that 
it seems neither a special insight nor part of a current, but simply 
an attitude which occurs to many moralists when they reflect 
upon the vagaries of human behaviour. But in Swift the attitude 
has unique depth and strength because its roots are religious and 
emotional, starting from that centre where his innermost ideals 
met his innermost feelings. Although Swift's illustrations come 
from a great range of social types, the prime specimen always be- 
fore his eyes is the fanatic : the digression on madness takes the 
psychology of religious zeal to be the paradigm of all folly. 

Since, however, the object defended in the digression is reason, 
one must consider that before examining the targets attacked. 
Those faculties which nowadays are separated into intelligence, 
moral insight, and intuition were, in Swift's time, still included 
under a single head : the same power that enables us to tell true 
from false, he assumed, also distinguishes right from wrong; and 
in the shape of 'discursive reason', or ratiocination, it produces 
the kind of logical analysis of which Euclid is the model. 

The material on which the reason or understanding works is 
given to it by the senses, and the record of their reports is the 
memory. However, there is also a power which can take the re- 
membered sensations and divide and recombine them in new 
patterns independent of experience : this is the imagination or 
fancy; and it falls peculiarly under the influence of our feelings or 

1 Selected Essays, ed. Geoffrey Keynes, 1930, p. 274. 


*A TALE OF A TUB 5 (ill) 

passions ; for those people whose emotions are least controllable 
have the most powerful fancies and the weakest understanding. 
The lure of the imagination is the pleasure which it can give, 
illusory perhaps, but more attractive than the honest reports of 
the senses or the impartial analysis of reason. As Temple said 
(with echoes of Hobbes), 

All the pleasures of sense, that any man can enjoy . . . grow fainter 
with age, and duller with use; must be revived with intermissions, 
and wait upon the returns of appetite, which are no more at the call 
of the rich, than the poor. . . But the pleasures of the imagination, 
as they heighten and refine the very pleasures of sense, so they are of 
larger extent, and longer duration. 1 

Although, therefore, the senses are indispensable, they should 
only serve, not dominate, the reason; and although the imagina- 
tion is not essentially bad, it must never displace either the senses 
as the windows or the reason as the governor of the soul. But the 
understanding itself has boundaries too ; and these seemed to 
Swift most apparent in the realm of discursive (rather than moral 
or intuitive) reason; for when that is given free play, it can be- 
come dangerous speculation. Modern readers have trouble fol- 
lowing the twists of the digression on madness because Swift was 
more anxious about the menace of speculation than the menace 
of conformity, whether in religion, philosophy, politics, or learn- 
ing. He shared Locke's wish, 

to prevail with the busy mind of man to be more cautious in med- 
dling with things exceeding its comprehension, to stop when it is 
at the utmost extent of its tether, and to sit down in a quiet ignor- 
ance of those things which, upon examination, are found to be be- 
yond the reach of our capacities. 2 

Or as Swift would say in 1 7 1 2, 'True learning, like all true merit, 
is easily satisfied; while the false and counterfeit is perpetually 
craving, and never thinks it has enough/ 3 

1 Observations upon the U.P., sigg. A6 V ~7. 

8 Essay concerning Human Understanding, z. i. 4. * Davis rv. 20. 


It is in problems of faith above all that the issue grew urgent; 
and here it was expressed as a conflict between credulity and in- 
credulity. 1 The arguments by which Anglican divines attacked 
the papist doctrine of transubstantiation, for instance, had to be 
such as would not boomerang against their own doctrine of the 
trinity. To protect their own creed, they used a distinction be- 
tween what was 'above' and what was "contrary to' reason. While 
transubstantiation was said to be contrary to reason, the trinity 
was said to be above reason. Some quotations from South and 
Tillotson will make the case clear: 

The case between transubstantiation and the Trinity is very differ- 
ent; the former being contradicted by the judgment of that faculty 
[i.e., the senses], of which it is properly the object; the latter being 
not at all contradicted but only not comprehended by the faculty 
[i.e., the reason], to which the judgment and cognizance of it does 

He that can once be brought to contradict or deny his senses, is at 
an end of certainty; for what can a man be certain of if he be not 
certain of what he sees ? In some circumstances our senses may de- 
ceive us, but no faculty deceives us so little and so seldom: And 
when our senses do deceive us, even that errour is not to be cor- 
rected without the help of our senses. 

Credulity is certainly a fault as well as infidelity: and he who said, 
blessed are they that have not seen and yet have believed, hath no where said, 
blessed are they that have seen and yet have not believed, much less, blessed 
are they that believe directly contrary to what they see. 2 

With these commonplace distinctions before us, we may state 
the Anglican position generally as that the better one reasoned, 
the more Christian one became, that revelation and the under- 
standing, being in perfect harmony, must strengthen one an- 
other. Unreason and sin become thus almost equivalent. 
Through such arguments, moreover, one disposes not only of 
Puritan 'enthusiasm* and papist superstition, but also of me- 
chanical systems like those of Epicurus, Descartes, and Hobbes. 

1 Gf. Andrew Marvell, Mr. Smirke and Defence of John Howe, passim. 
3 Robert South, Sermons, 1824, n, 494-5 and passim; John Tillotson, A Discourse 
against Transubstantiation, 1684, pp. 38, 42. 


'A TALE OF A TUB 5 (ill) 

By leaving no place for 'spirit 9 in the Christian sense, by dividing 
faith from reason, and by making the laws of nature (human or 
physical) independent of immediate Providence, the 'systems' 
denied exactly what the Anglican position affirmed. Hence it is 
that with the digression on madness Swift 'satirizes, in the great- 
est detail, the contemporary neo-Epicurean and Hobbesian 
philosophies'. 1 


Though the implications of the digression on madness are com- 
plex indeed, its design is straightforward; for Swift's stratagem of 
camouflaging his real movements under an appearance of sys- 
tematic argument appears in each expository 'section' of the 
Tale. The digression on madness has two divisions : the theory of 
madness and the application of the theory. In the first, he begins 
with two propositions : that the greatest deeds are due to mad- 
ness and that madness is caused by vapour on the brain. As evi- 
dence, he uses three categories : acts of military conquest (Henry 
IV and Louis XIV of France), acts of philosophic innovation 
(Epicurus and Descartes), and acts of religious innovation (the 
nonconformist sects) . By analysis of these classes, Swift proves 
that new kingdoms, new philosophies, and new religions are all 
established by madmen. He provides only asterisks where there 
should be an explanation of how individual differences make the 
same madness lead to military activity in some, philosophical in 
others, and religious in others. The entire first section ends with 
two famous paragraphs on the function and value of madness. 

The second division begins with two propositions : that mad- 
ness is brought on by an excess of vapours; and that this excess, 
properly disposed, would be of service to the community. Swift 
takes his examples from Bedlam Hospital and shows how each 
kind of lunatic, correctly adapted, could fulfil some ecclesiastic, 
civil, or military responsibility. A violent maniac becomes a 
colonel; a babbler, a lawyer; an anxious compulsive, a city mer- 
chant; a megalomaniac and pathological liar, a courtier; a 

1 Starkman, p. xix. 



coprophiliac, a physician; and 'a taylor run mad with pride', a 
bishop. In a final paragraph he extends the plan to include other 
persons puffed up with vapours, such as fops, musicians, authors, 
and statesmen. 

This scheme has an air of logical discourse and the shape of a 
formal speech. It is a defence of reason in the Erasmian design of 
a praise of folly, and its development is a parody of logic. Swift 
makes one expect normal procedures : a principle followed by 
illustrations ; a series of postulates and corollaries building up to a 
proved conclusion. He then frustrates this expectation by using 
cases which contradict his ostensible principle, or arguments 
which defy his premise. If the reader knows the real, unexpressed 
proposition which is being advanced, he will see that the proofs 
and examples do, ironically, support it. But they do so in no arti- 
culated form. The composition which results amounts to a re- 
casting of the Moriae Encomium two centuries after the original. 1 

To Erasmus as well may be traced, ultimately. Swift's inverted 
decorum personae., his acting the dunce (here and elsewhere) be- 
cause he is exposing fools. But only in the work of Marvell had 
Swift's talent for applying that principle been adumbrated. Pre- 
tending that his Rehearsal Transpros'd must take a clownish form 
because its victim, the Bishop of Oxford, is a clown, Marvell has 
to excuse himself when he slips into serious argument, for that is a 
violation, as it were, of the crazy decorum; otherwise, however, 
Marvell consistently makes his own pose a caricature of the one 
he attributes to the bishop. 2 Swift has no serious argument to 
apologize for; but he picks up MarvelPs comical decorum and em- 
bodies it in the most brilliant shape it has yet received. 

The second of the two central paragraphs of the digression can 
be anatomized to show his full, elaborate method in detail. Here, 3 
Swift says that knowledge of surfaces (credulity) is better than 
knowledge of interiors (curiosity). To prove this, he says what 

1 Gf. Ports, pp. 385-8; Ronald Paulson, Theme and Structure in Swift's Tale of a Tub 
(New Haven 1960), pp. 79-80. 

* John S. Coolidge, 'Martin Marprelate, Marvell, and Decorum Personae as a 
Satirical Theme', PMLA. 9 LXXTV (1959), 526-32. 

8 Tale, pp. I73-4- 


*A TALE OF A TUB' (ill) 

would seem to argue against it : that man's reason pries into the 
depths while his senses are satisfied with appearances. To prove 
this in turn, he says what again seems a contradiction : that reason 
is correct (though therefore to be not followed but rejected) in 
deciding that most bodies, when investigated, prove worthless 
within though attractive without. The conclusion is that since 
outsides are thus 'superior' to insides, we ought to make the most 
of our appearance and ignore our essence. 

The real proposition here is that man's moral essence or reason 
his inside is infinitely more important than his physical acci- 
dents the outside. Reason and judgment are therefore analyti- 
cal or introspective., while the senses, passions, and imagination 
(when divorced from the understanding) may be tricked by ap- 
pearances. It is better to be sadly wise and know oneself, Swift 
really says, than to be complacently self-deceived. 

Such axioms are neither argued nor ironically implied in a co- 
herent order; they are repeatedly illustrated by innuendo, 
through absurd assertions and fantastic instances. In this para- 
graph, for example. Swift says with a straight face that surface 
counts for more than substance, but means ironically that it is 
worth nothing by comparison. He says with contempt that reason 
is mistakenly occupied with internals, but he means seriously that 
reason is correct in its business. He says that natural law requires 
us to put our s best furniture forward* but means that unnatural 
vanity is behind man's obsession with show. Supposedly to illus- 
trate how much appearances are preferable to reality, he com- 
ments, 'Last week I saw a woman flafd, and you will hardly 
believe, how much it altered her person for the worse.' Between 
the bland understatement of the reaction, and the loathesome- 
ness of the event, lies a gap the parallel to which separates the 
foolishness of the narrator in being surprised, from the inevit- 
ability of the operation's causing disfigurement. The least con- 
sequence of flaying is its effect on the victim's beauty. The agony 
is what matters. Only grotesque immorality would see things so 
far out of proportion: pride is so blind that hellish pain holds less 
terror for it than bad looks. Ostensibly, the anecdote shows the 



value of a good complexion; really, Swift is saying that irrational 
vanity will consider ugliness the most deplorable result of being 
skinned alive. 

The succession of these assertions and instances has a pseudo- 
logic in accordance with the outward argument. But actually it is 
irrational, except to this extent : that only certain motifs are re- 
peated, that each of them connotes others, and that they are 
juxtaposed so as to heighten their effectiveness through contrast 
and paradox. The complete digression admits of interpretation 
by this method; and its force is that only madness can account for 
the zeal of those men who, rather than live at peace with ortho- 
doxy, reason, and good taste, insist on upsetting themselves and 
others with dangerous, or at best unprofitable, newfangleness. 

By madness Swift implies two features : an extravagant addic- 
tion to one's own opinions and an excess of vapour on the brain. 
Stubborn insistence upon personal convictions however un- 
likely or iconoclastic had another name besides madness: 
pride. To be puffed up with clouds of self-importance meant to 
be filled with the sin of vanity. Swift is once more ironical when 
he says madness causes great and beneficent innovations; for he 
means that zeal against orthodoxy is the growth of egoism. What 
to a lunatic looks like progress, to a sane man looks like pride ; and 
this was Temple's judgment of natural philosophy. 

If such 'madness' be opposed to reason, we face the medieval 
definition of a person as 'the individual substance of a rational 
nature' : the common element, what we share with others, is by 
this definition reason; the separate element, whereby we differ 
from others, is will. Hence, the more wilful, proud, sceptical we 
are, the more irrational and individual we are; on the contrary, 
the more reasonable, the more like others. Madness, sin, and per- 
verse individuality are thus essentially opposed to sanity and 
conventionality. As Hooker said, 'The most certain token of evi- 
dent goodness is, if the general persuasion of all men do so 
account it.' 1 

1 Laws of Ecclesiastical Polity i. viii. 3; cf. Michael Oakeshott's introduction to his 
edition of Hobbes's Leviathan (Oxford 1946), p. Iv and n., also p. 28 of the text. 


*A TALE OF A TUB 5 (ill) 

So it is that Swift contrasts curiosity and credulity. Curiosity 
may be bad when it idly and insatiably pokes into the foundations 
of settled, unsearchable beliefs ; for it then becomes the pride or 
incredulity which caused Adam's fall. But curiosity is good and 
reasonable when it encourages self-criticism within the orthodox 
framework, when it enhances modesty or destroys falsehood. 
Credulity, however, is bad in so far as it idly follows a wicked or 
irrational authority, as it accepts immoral assurances as to the 
grace of an individual, as it strengthens complacency and cor- 

In this digression. Swift reaches the height of his literary 
achievement. He has given elaborately formal, evocative ex- 
pression to a rich, powerful moral insight. It was to be almost an- 
other thirty years before he would repeat, and even surpass, the 
achievement; and that was to be only once, in Gulliver's voyage 
to the Houyhnhnms. 


Chapter Four 

The Battle of the Booksis one of Swift's most detached and en- 
joyable productions. It is a shame^that the work must be 
closely identified with its occasion/ because Swift employs 
the presumed arguments as an excuse for independent comedy. 
To make the Battle of the Books into a chapter in the history of ideas 
can be as misleading as to make Le Malade Imaginaire an illustra- 
tion of seventeenth-century medical practice. Furthermore, the 
ancients-moderns querelle was not as has been asserted a con- 
flict between emerging science and reactionary humanism, but 
an essential part of the humanistic tradition itself, 1 Even moral 
issues are almost irrelevant here ; Swift's humour is ultimately as 
remote from 'ideas' as a brilliant piece of comic prose can be. Yet 
the achievement arises from a cross-play of allusions ; and until 
these are recognized, the charm of the 'history' cannot be .ap?.. 
preciated.JCJnless one knows that Dryden translated Virgil, .one 
misses the wit of his giving Virgil Crusty if on* armour iii exchange 
for gold. If one does know it, however, the episode becomes less 
an argument that the modern is infei^or to the ancient than a 
little farce out of commedia deWarte. (The biographical, moral, 
literary, and intellectual aspects of the Battle of the Books constitute 
its accidents, not its essence.} 

If we nevertheless examine the work as a moral satire, its posi- 
tive basis will be seen to be the teachings of Temple in his essay 
Of Poetry. The Battle is a defence of taste as an expression of char- 
acter ; and Swift implies ultimately that a good book is one which 
a *good' man approves. By this principle one may use literature as 

1 Hans Baron, 'The Querelle of the Ancients and the Moderns', J. H.L, xx (Jan. 



a measure of social health. In a good society, bad writers cannot 
flourish because they will find no readers. 

The enemies of good taste are those who offend on both sides : 
the poetasters, hack journalists, pedants, who manufacture trash ; 
and the corrupt judges who read and recommend it. That Swift 
gives this class the name of 'moderns' is less an attempt at defini- 
tion than a tribute to Sir William Temple ; for Swift traces their 
descent from antiquity. Similarly, the leaders of the moderns 
happen to be identified as William Wotton and Richard Bentley 
not because these are the worst writers or worst critics of the age 
but because they had disagreed with Temple. - 

The situation which led Swift to write his book goes back to the 
time , of his second departure from Moor Park. Wotton, in his 
methodical Reflections upon Ancient and Modern Learning (1694), 
had taken up Temple's essay Of. Ancient .and, Modern Learning 
(1690) and defended modern arts and sciences against Temple's 
intimation that no significant advance had been made on the 
accomplishments of the ancients. Temple had bestowed unusual 
praise on the letters of Phalaris and the fables of Aesop, which he 
accepted (though admitting the doubts of scholars) as the earliest 
works of their kind, ajid authentic compositions of the historical 
Phalaris and Aesop^ Unfortunately, the dean of Christ Church, 
Oxford, decided, partly as a gesture in Temple's direction, to 
make Phalaris the text of an editorial exercise which he used to 
assign to bright undergraduates. The student chosen was the 
Hon. Charles Boyle, but the main work of preparing the edition 
(1695) was carried out by older men in the college. Through a 
wicked coincidence, one of the manuscripts required belonged to 
the King's Library ( c St. James's Library'), of which Richard 
Bentley had just been made Keeper: and instead of facilitating 
the necessary collations, Bentley saw fit to hinder them. Here, 
therefore, were the tinder and wood for an academic blaze. 
/The match was soon applied : for Wotton, in the second edition 
of his Reflections (1697), made room for a Dissertation by Bentley, 
bluntly impugning both Temple's judgment and Boyle's edition 
(in which that judgment was paraphrased), and demonstrating 


that the fables and epistles could not be authentic. By now Swift 
had returned from Kilroot, and was at Moor Park, writing A TaU 
of a Tub. Soon a rebuttal of Bentley appeared, in the form of an 
Examination attributed to Boyle but again prepared by his seniors 
(March 1698); and this stimulated Swift either to create or to 
revise some of his own 'digressions' and other expository chap- 
ters, for material from it is incorporated not only into the Digres- 
sion on Critics but as well into the many quick thrusts at Wotton 
and Bentley which are scattered throughout the Tale.' 1 

It seems to have been, however, the memory of the recent war 
with Louis XIV that determined the plan of the new satire, for 
the Treaty of Ryswick was only settled in the autumn of 1697. 
The closing campaigns of the War of the League of Augsburg had 
produced a stalemate. The attempted invasion of England had 
been a fiasco (April 1 696) . Neither side could claim a triumph in 
the treaty: conquests made by either were to be returned to the 
other within six months. 2 Swift gave his story the appearance of a 
sensational pamphlet reporting a fresh battle; but he gave the 
battle an ending as inconclusive as the terms of the peace. Im- 
plicitly, he thus made the narrator a journalist, or one of the 
moderns, although the tone of the account obviously favours the 
ancients. If Swift seems to imply that England is the defender of 
the ancients and France of the moderns, he has convention on his 
side ; for Temple had derived the immediate controversy over the 
relative merits of the two literary parties from Perrault and Fon- 
tenelle; and one of the chief documents in the controversy was a 
poem by Perrault flattering Louis XIV by asserting that U si&cle 
de Louis le Grand was not inferior to classical antiquity. 

Swift may also have drawn analogies between the recent war 
and the war of the Trojans and Latins in the Aeneid. In the plates 
of Dryden's translation Aeneas was given the face of William III ; 
in Virgil's epic, Aeneas, an idealized Augustus, is the bringer of 

1 A. G. Guthkelch's edition of the Battle of the Books> in the King's Classics (1908), 
follows the controversy in detail, giving lengthy extracts from relevant works. For 
some additions to the list of books which Swift may have read while he was writing 
the Battle of the Books t see Appendix G. 

* Ogg n. 439. 


civilization; and Swift's London was commonly conceived of as 
a new Rome (Nova Augusta). Since the Battle of the Books takes its 
design mainly from the Ameid (Dryden's version appeared in 
July 1697), and since Temple is the hero. Swift could be suggest- 
ing a parallel, through King William, between Augustus- Aeneas 
and Sir William. In Swift's own career. Temple appears indeed 
to have been the great civilizer, covering provincial brick with 
cosmopolitan marble. 

Certainly the story is founded upon Virgil, though with many 
allusions to Homer. It consists of a historical introduction which 
is interrupted by the fable of the spider and the bee; then the 
mobilization of the opposing sides, interrupted by the episode of 
Momus and Criticism; and at last a fragmentary account of 
several battles, closed by the long episode of Bentley and Wotton 
meeting Temple and Boyle. The historical introduction, with the 
newly arrived Moderns challenging the property rights of the 
old inhabitants, the Ancients, seems parallel to the difficulties 
between Aeneas and Latinus (Aeneid vn) ; and the catalogue of 
the forces goes back to both Homer and Virgil (Iliad n. 484-877, 
Aeneid vu. 64 1-8 1 7) . The meetings of Momus with Criticism, and 
of Criticism with Wotton, are derived from the Allecto-Juno- 
Amata-Turnus sequence in the Aeneid (vn, 5286-465) . Among the 
battles, the encounter of Virgil with Dryden is a parody of 
Glaucus and Diomedes (Iliad vi. 119-236); Blackmore and 
Lucan recall Hector and Ajax (Iliad vn. 244-302) ; Creech and 
Horace recall Turnus and Aeneas (Aeneid x. 63688) ; Aphra 
Behn is Camilla (Aeneid xi); and Pindar attacking Cowley 
suggests Diomedes attacking Aeneas (Iliad v. 3024). In the 
final episode, the portrait of Bentley and his chastisement by 
Scaliger follow closely the Thersites-Odysseus exchange in 
Homer (Iliad n. 211-69); Bentley and Wotton are modelled 
on Nisus and Euryalus (Aeneid ix. 314-449); and the attack 
on Temple is a parody of Arruns* attack on Camilla (Aeneid xi. 

Additional allusions seem stitched into these. For example, 
Jupiter enthroned, with a chain of 'second causes' attached to his 



toe, suggests not only the famous gold chain of the Iliad (vm. 19) 
but Lucian's Icaromenippus 1 and Bacon's idea of second causes as 
'nature's chain', of which the highest link is e tied to the foot of 
Jupiter's chair'. 2 Of such allusions, the most interesting is an 
elaborate parody of King Arthur, Blackmore's doggerel epic on 
William III; for the whole of Swift's Momus-Griticism episode 
seems to caricature the parallel episode in that poem. 3 Black- 
more's manner invited parody, and others were already mimick- 
ing him 4 ; but Swift's effort may reflect certain personal align- 
ments. Blackmore was sympathetic with Dissent; he followed the 
new Whigs ; he associated himself with the commercial and finan- 
cial interests of the City of London. Swift was a devout Anglican 
who did not trust the Dissenters; an old Whig, with views which 
enlightened 'Tories' would soon accept; and an admirer of the 
landed gentry. To Swift, perhaps, Blackmore's principles seemed 
as detestable as his prosody. 

Certainly the mock-epic situation, in which these parodies and 
echoes are submerged, belongs to the seventeenth century. In 
one of Boccalini's Advertisements from Parnassus, Apollo has to 
marshal his belletristic followers for the defence of Parnassus 
against the powers of Ignorance. 5 In de Calibre's Histoire 
Poetique the ancients and moderns fight about the twin peaks of 
Parnassus. 6 In the fifth canto of Boileau's Le Lutrin (which is itself 
a document in the ancients-moderns controversy) , Chicane plays 
the role of Allecto (Swift's Criticism) ; and the two parties under- 

1 Paragraphs 25^6. 

a Advancement of Learning^ Bk. i, admit. (World's Classics ed., p. 1 1). This applica- 
tion of the image is commonplace: cf. Hobbes, Leviathan, pt. n, chap, ax, ad init. 
(Cambridge English Classics, ed. A. R. Waller, p. 148). See also A. O. Lovejoy, 
The Great Chain of Being, 1936, passim. 

3 See Blackmore's King Arthur, 1697, pp. 61-77 (Bk. m, 11. 1-451). For a detailed 
comparison, see MLN. t LXX (Feb. 1955), 95~7- (Blackmore combined Scylla in the 
Odyssey with Juno and Allecto in the Aeneid.) 

* See Albert Rosenberg, Sir Richard Blackmore (Lincoln, Nebraska, 1953), pp. 36, 
39-70; Richard C. Boys, 'Sir Richard Blackmore and the Wits', University of 
Michigan Contributions in Modern Philology, no. 13, Ann Arbor 1949. 

5 Advertisement LXXXVI. See Traiano Boccalini, Advertisements from Parnassus 
[trans, by N.N.], n (1704), 83-9. (The first English translation of Boccalini's Adver- 
tisements appeared in 1656.) 

6 For an acute comparison, see Craik i. 912. 



go a mock-heroic battle in a bookstore, with the fighters throwing 
famous authors at one another : 

La, pr&s d'un Guarini Terence tombe & terre, 
La y Xenophon dans Fair heurte contre un la Serre. 

As the Renaissance retrieval of ancient science had become the 
antiquarian philology of Augustan scholarship, so humanism had 
stiffened into neo-classical hierarchies, themselves no less sterile 
than the scholastic dogmas against which the tradition had been 
once directed; and it was natural that some geniuses first in 
Italy, then in France, and later in England should test the 
dimness of the decline by the vitality of the dawn. Of course the 
new learning of Vossius and Bentley was no continuation of Eras- 
mus and Montaigne/It grew out of a divorce between the ma- 
terials of scholarship and the ends of wisdom. And this divorce is 
what Swift was attacking. The mock-epic serves his need for pro- 
test because it reminds the reader of the original grounds of our 
esteem for the ancients, at the same time as it suggests the mean 
purposes into which the classical heritage has been betrayed/f n 
the single structure we recognize the Aeneid of Virgil, the Aeneiaof 
Petrarch, and the Aeneid of Bentley. Thus the hiatuses in Swift's 
battle scenes are meant to burlesque the scholarly arguments 
which Bentley founded upon Void spaces' in manuscripts. 1 The 
mock heroic poet is not ridiculing Homer or Virgil. He is at- 
tributing a heroic pretension to those moderns whom he wishes 
to satirize; then, standing them beside the monuments of Greece 
and Rome, he shows us how absurd is the gap between pre- 
tension and achievement. 

Swift is, in this sense, appealing to the old scholarship against 
the new, and his appeal reaches its climax in the fable of the 
spider and the bee. Here he was perhaps modelling his work 
upon that of Sir Roger L'Estrange; for in L'Estrange's popular 
versions of Aesop one finds the same coarse language, the same 
lively dialogue, colourful letterpress, and elaborate moralizing, 

1 See Boyle's Examination, p. 221. 



as in Swift's creation. 1 Swift may even have expected the reader 
to notice and feel amused by the parallel. In general, however, 
Swift's employment of Aesop as a character alludes to the second 
part of 'Boyle's' polemic : Dr. Bentlefs Dissertation upon the Fables 
of^Esop Examined; in this part, furthermore, occurs a discussion of 
the animated 'picture' of Aesop's Fables described by Philostratus 
in his Imagines. 2 In 1698 there was also a Christ Church edition of 
Aesop, in which the fable of the dog in the manger was so inter- 
preted as to reflect upon Bentley and his manuscripts of Phalaris. 
The fable of the spider and the bee makes the link between A 
Tale of a Tub and the Battle of the Books* The relation to the 
Digression on Critics is obvious, but deeper bonds exist. At his first 
appearance the spider was already 'swollen up to the first magni- 
tude' ; when he saw the bee's effect upon his fortress, he 'swelled 
till he was ready to burst' ; before undertaking the verbal duel, he 
'swelled himself into the size and posture of a disputant' ; and he 
began arguing 'with a resolution to be heartily scurrilous and 
angry, to urge on his own reasons, without the least regard to the 
answers or objections of his opposite'. As the bee points out, the 
contents of the spider's swelling 'over-weening pride' are 'dirt 
and poison'. 4 Bacon is probably the source of these hints, and he 
indicates how they refer to the Digression on Madness; for Bacon 
says it is not the 'quantity of knowledge' alone that can 'make the 
mind of man to swell' : 

It is manifest that there is no danger at all in the proportion or 
quantity of knowledge, how large soever, lest it should make it swell 
or outcompass itself; no, but it is merely the quality of knowledge, 
which be it in quantity more or less, if it be taken with the true cor- 
rective thereof, hath in it some nature of venom or malignity, and 
some effects of the venom, which is ventosity or swelling. This cor- 
rective spice, the mixture whereof maketh knowledge so sovereign, 

1 E.g., no. 258, the swallow and the spider (Fables, of Msop, 1692, pp. 224-5). 
Around 1698 there was a burst of anthologies of satirical poems in imitation of the 
fables, with titles like Msop at Bathe, Msop at Epsom, etc. 

a Boyle's Examination, p. 274. 

3 A similar comment is made by John Bullitt, in Jonathan Swift and the Anatomy of 
Satire (Cambridge, Mass. 1953), p. 117. 

4 Tale, pp. 228, 230-1. 



is charity, which the apostle immediately addeth to the former 
clause ; for so he saith, 'knowledge bloweth up, but charity buildeth 
up. 51 

So here, as in. the Digression on Madness, flatulence becomes a sym- 
bol of pride, and thus pedantry is a kind of flatulence. 

It is also Bacon who compares an introspective, reclusive, nar- 
row-read pedant to a spider: the scholastics, Bacon says, 

shut up in the cells of monasteries and colleges; and knowing little 
history, either of nature or time ; did out of no great quantity of 
matter, and infinite agitation of wit, spin out unto us those labori- 
ous webs of learning which are extant in their books. For the wit 
and mind of man, if it work upon matter, which is the contempla- 
tion of the creatures of God, worketh according to the stuff and is 
limited thereby; but if it work upon itself, as the spider worketh his 
web, then it is endless, and brings forth indeed cobwebs of learning, 
admirable for the fineness of thread and work, but of no substance 
or profit. 2 

For Swift, of course, the model of a contentious pedant, obsessed 
with the novelty of his futile minutiae, was Bentley: part of the 
division of Boyle's Examination devoted to Aesop was a burlesque 
dialogue between Bentley and another scholar, in which the 
former spiderishly refutes a charge of plagiarism: C I tread in no 
man's footsteps.' 3 

Although some scholars have argued that Swift's spider has 
more definite connotations than arid pedantry divorced from ex- 
perience of the world, and though they have tried to tie it specifi- 
cally to the new natural philosophers, their argument seems im- 
probable. 4 The image was a commonplace and might be applied 
with any overtones which suited an occasion. Locke, for ex- 
ample, arguing with Stillingfleet, denied that his own doctrines 

1 Advancement of Learning, Bk. I, ad init. Cf. Martin Price, Swift's Rhetorical Art 
(New Haven 1953), pp. 4-5. On Swift's relation to Bacon, see R. F. Jones, 'The 
Background of the Battle of the Books 9 , Washington University Studies vn (St Louis, 
1920), 159 and passim; Ancients and Moderns (St Louis, 1936), p. 55 and passim, 

* Advancement of Learning, Bk. I, in med. (World's Classics ed., p, 30). 
3 P. 250. 

* E.g., Ernest Tuveson, 'Swift and the World Makers', J.H.L, xi (1950), 54~74- 



were borrowed from Descartes; they were, he says, *spun bare- 
ly out of my own thoughts, reflecting, as well as I could, on my 
own mind, and the ideas I had there, and were not, that I know, 
derived from any other original'. 1 Stillingfleet made the image 
invidious : c Although those who write out of their own thoughts do it 
with as much ease and pleasure as a spider spins his web ; yet the 
world soon grows weary of controversy.' 2 Hereupon, Locke flung 
it back at him in the same spirit : "Should I ... think I had some 
right to return the general complaint of length and intricacy 
without force; yet you have secured yourself from the suspicion 
of any such trash on your side, by making cobwebs the easy pro- 
duct of those who write out of their own thoughts, which it might 
be a crime in me to impute to your lordship.' 3 Yet both men, 
in this combat, were opposed to Descartes; Locke stood for 
the Baconian tradition; and Stillingfleet kept the methods of 
scholastic logic which Bacon's invention had been designed to 

The final element of Swift's fable remains : the bee. This was 
supplied by Bacon in the Novum Organum; here is the famous pas- 
sage contrasting empiricists, in natural philosophy, with dog- 
matists : 

The empirics, like the ant, amass only and use : the [dogmatists] , 
like spiders, spin webs out of themselves : but the course of the bee 
lies midway; she gathers materials from the flowers of the garden 
and the field ; and then by her own power turns and digests them. 
Nor is the true labour of philosophy unlike hers : it does not depend 
entirely or even chiefly on the strength of the mind, nor does it store 
up in the memory the materials provided by natural history and 
mechanical experiments unaltered, but changes and digests them 
by the intellect. 4 

This distinction, which was to impress Coleridge deeply, also 
struck Temple, who admired Bacon as among the greatest 

1 Works, 5th ed., 1751, i. 369. 

8 The Bishop of Worcester's Answer to Mr. Locke 9 s Second Letter 9 1698, p. 4. 

8 Works 1.438. 

4 Transl. G. W. Kitchin (Oxford, 1855), aphorism no. 95, p. 78. 



modern thinkers ; and in his essay Of Poetry , Temple adopted the 
bee simile with the literary connotations already belonging to it; 
only narrowing its meaning to signify the far-ranging, elaborate 
art of ancient genius. 1 Swift, knowing his master's essay, may 
have read Bacon with the added stimulus of Temple's recom- 
mendation. He manifestly employed both the broad and the nar- 
row connotations of the images, ramified them by cross-reference 
to the Digression on Madness, and dramatized the outcome in his 

One source of the appeal of Swift's allegory is its underlying 
'naturalness' ; for the bee in fact goes with honey and wax, the 
spider popularly connotes dust and menace. Swift is also drawing 
upon such diverse cliches as aranearum telas texere (for hollow pole- 
mics) and 'Where the bee sucks honey, the spider sucks poison/ 2 
Others had already come near the application made by him and 
Temple. Florio, for example, had complained that literary critics 
c doo not seeke honie with the bee, but suck poyson with the 
spider' 3 ; and we find a similar figure in Beaumont and Fletcher: 

Sweet poetry? s 

A flower, where men, like bees and spiders, may 
Bear poison, or else sweets and wax away. 
Be venom-drawing spiders they that will; 
Pll be tlu bee, and suck the honey stilL* 

The fable therefore provides an early example of one of Swift's 
most characteristic and effective devices : to pick up some com- 
monplace distinctions, embody them in 'naturally' appropriate 
creatures, and dramatize the resulting juxtaposition in a comic 
scene with dialogue. He tells no story but brings home the ironies 

1 Miscellanea n (1696), p. 323. 

2 Cf. Herbert Davis, The Satire of Jonathan Swift (New York 1947), pp. 21-2; 
Morris Palmer Tilley, A Dictionary of the Proverbs in England in the Sixteenth and 
Seventeenth Centuries^ proverb no. B 208. 

3 Florios Second Frutes, 1591, 'To the Reader', 

* Pour Plays . . . in One, in the Works of Beaumont and Fletcher, ed. A, Glover and 
A. R. Waller (Cambridge 1905-12), x. 312. Professor Ronald Paulson traces the 
bee-poet analogy to Plato's Ion (Theme and Structure in Swift's Tale of a Tub [New 
Haven 1960], p. gx, n.). 



implicit in the cliches. To this genre belong the fat man and the 
weaver in the 'Preface 9 to A Tale of a Tub. 

Though Swift employs no suspenseful plot, he provides a suc- 
cession of unpredictable flowerings not unlike his sentence struc- 
ture in effect. Without periodicity., Swift's clauses turn on articu- 
lations which are the more evocative because one could not fore- 
see them; e.g., concerning the spider in his web : c ln this mansion 
he had for some time dwelt in peace and plenty, without danger 
to Ids person by swallows from above, or to his palace by brooms from 
below 51 one is prepared to end this sentence at each comma, but 
one is pleased by what appears when one goes on, and the very 
touch of rhetorical parallelism is itself ironic. So, in the fable, the 
reader feels ready to stop gratefully after the preliminary dia- 
logue, assuming there will be a brief, summary close. But two 
apparently climactic speeches follow, expatiating with unpre- 
dictable inventiveness on adumbrated themes. Again we pre- 
sume a conclusion. Yet the truly climactic 'moral' by Aesop still 
follows, and only then does the scene end. The prose style, mean- 
while, is Swift's best. Concrete, homely speech; fresh figures, 
with no varnish of 'rhetoric' ; quick, colloquial rhythms ; syntax 
so idiomatic, expressive, so plainly hinged and turned, that it 
seems the language of proverbs. 

Among Swift's satires this episode is further remarkable in its 
lack of negative emphasis. The tone seems that of confident 
affirmation. Swift states his principles openly, as he rarely does in 
comic writing : 

As for us, the Antients, we are content with the bee, to pretend to 
nothing of our own, beyond our wings and our voice: that is to say, 
ourJKghts and our language; for the rest, whatever we have got, has 
been by infinite labor, and search, and ranging thro' every corner 
of nature: The difference is, that instead of dirt and poison, we have 
rather chose to fill our hives with honey and wax, thus furnishing man- 
kind with the two noblest of things, which are sweetness and light. 2 

Flights, language; sweetness, light; that is, elegance of expression. 

1 Tale, p. 229. a Ibid.> pp. 234-5. 



and clarity of sense Lucian's x&P lT $ Kod oxxpriveioc these are 
the real definition of 'ancient* virtues in Poetry; they exist, for 
Swift, in Rabelais as in Lucian, in Cervantes as in Horace ; and 
they are embodied in his fable. 


Chapter Five 


The structure of The Mechanical Operation of the Spirit is 
simple. It has four parts: some preliminaries; a 'Section 
f, on Dissenting congregations; a 'Section if, of which 
the first part is on Dissenting preachers; and a closing part of 
e Section n* (not separately labelled), on the history of 'enthusi- 
asm'. The preliminaries include a 'Bookseller's Advertisement', 
probably by Swift, stating that the work is incomplete and of 
doubtful authorship; and an epistolary opening, in which the 
discourse is addressed to a virtuoso in Australia. 1 

In 'Section i' Swift says he proposes to treat of Dissenting zeal 
in a new way, viz., as enthusiasm artificially excited at first but 
become natural through habit: this is a doctrine concerning 
hypocrisy which Swift was to make peculiarly his own, that what 
may have begun as an affectation can through long use become 
an elementary part (bad or good) of one's nature. He then pro- 
ceeds to give a coarse caricature of the alleged manner and vices 
of Dissenters attending a conventicle, a caricature embodying 
most of the ancient, hack charges accumulated by their enemies. 
In not very comic irony. Swift simply describes these supposed 
hallmarks of Nonconformity as if they were premeditated tech- 
niques for inducing, in the minister and his hearers, fits which 
release their most uninhibited fantasies. Swift may be referring 
tangentially to the Roman Catholics' use of visible symbols and 
meditational exercises; for the title of the 'Discourse* probably 

1 The 'Bookseller's Advertisement* has hints that Swift himself felt the satire was 


comes from a couplet in Hudibras sneering at such papist prac- 
tices as 

The tools of working out salvation 
By meer mechanick operation* 

In 'Section n', after a long and dazzling introductory para- 
graph, Swift expounds, by mock-physiology, the relation be- 
tween the 'real' internal constitution of the brain (an Epicurean 
'crowd of little animals') and the workings of the mind. This is to 
explain how Nonconformist preachers think. Next, Swift ana- 
lyses the speech of the ministers, as he had earlier analysed the 
behaviour of their congregations. Employing traditional lines of 
attack, he describes, in caricature again, their enunciation and 
their language; and he pretends that all these traits are deliber- 
ate means of arousing 'enthusiasm' (i.e., delusions of being di- 
vinely inspired) in the worshippers. Finally, he traces the alle- 
gedly nasal whine of their voices to a decayed nose ruined by 
the effects of venereal disease. 

What follows amounts to a further section, though not so en- 
titled. 2 This is a brief, punning mock-history of enthusiasm, from 
ancient times to the present. Here, by a rough pseudo-allegory, 
Swift not only pretends that pagans, Roman Catholics, and Dis- 
senters are indistinguishable in this respect, but he also pretends 
to discover that the common element, and thus the essential 
cause, of all the phenomena is concealed (not stifled) lust. His 
irony, here, is in the style I do not suggest influence of 
Hobbes's amazing parallels, at the end of Leviathan, between 

too violent; and Swift later describes the portion deleted from *Section n* as 'neither 
safe nor convenient to print* (p. 276) ; cf. similar hints added in the fifth edition: 
pp. 1718 and the italic note, pp. 2612. There are bints of a different sort in the 
address to *T. H. Esquire', at the 'Academy of the Beaux Esprits in New Holland' : 
the initials may be an anagram of *the squire* ; the Academy is probably a thrust at 
the Royal Society; the epistolary style suggests the elaborate correspondence main- 
tained by the philosophical societies; New Holland apparently is to connote the 
self-important provinciality of the philosophical societies cf. the Iroquois and the 
Topinambo literati two pages later (p. 263) ; beaux esprits may possibly allude to 
hedonistic free-thinkers (see George Boas, The Happy Beast [Baltimore 1933], 
pp. 64-70). 

1 ra. i. 1497-8, noted in Tale, p. 261, n. i. * Tale, pp. 282-9. 



the papacy and the kingdom of fairies: 'The fairies marry not', 
Hobbes says, for instance; 'but there be amongst them incubi, 
that have copulation with flesh and bloud. The priests also 
marry not.' 1 Swift draws a closing parallel, appropriately but 
surprisingly, between a modern lover's manners in courtship 
and a Dissenting minister's manners in preaching; and this 
leads to the famous aphorism, c Too intense a contemplation 
is not the business of flesh and blood ; it must by the necessary 
course of things, in a little time, let go its hold, and fall into 

Swift's subject is thus Protestant Nonconformity in England 
or, we might say, a further consideration of Jack, from A Tale of 
a Tub. Yet the author ironically gives himself the character not 
only of a Dissenter but also of a 'modern' and a virtuoso. Swift's 
purpose in adding such references was probably to account for 
the mechanistic terms of his analysis; for he employs something 
like the language and methods of Hobbes, the interpretation of 
human nature as wholly governed by the laws of matter and mo- 
tion. Hobbes's mechanistic psychology was traditionally under- 
stood to encourage atheism, and all faiths abhorred him, the 
Puritans as much as the Anglicans. Swift is therefore implying 
that while Hobbes is wrong, his theories do fit the 'unnatural' be- 
haviour of the sects. Conversely, since Hobbes detested the Puri- 
tans, Swift is ridiculing him by implying that they are his most 
eager imitators. 8 If the Mechanical Operation is primarily an attack 
upon Dissent, it is, in a more limited way, also a satire upon 
pseudo-scientists (Descartes and Hobbes, not Bacon or Locke) 
and pedantic scholars. 

Among the Tale, the Battle^ and the Mechanical Operation, the 
connection is not obscure. For instance, the pretended 'spirit' 
(of the hypocrites whom Swift is attacking) operates like the 

1 Leviathan TV. xlvii (ed. A. R. Waller, p. 518). * Tale, pp. 288-9. 
* -* Much of Swift's language, in the Mechanical Operation, seems drawn from 
Hobbes: e.g., pp. 274-5, 'things invisible*, 'invisible power 5 , 'fear* and 'desire* as 
causes of religion; cf. Leviathan x. xi-xii (ed. A. R. Waller, Cambridge, 1904), 
pp. 68-71. See also David P. French, 'Swift and Hobbes A Neglected Parallel', 
Boston University Studies in English m (Winter 1957), 2 43~*55- 



Spider's genius, 'entirely from within'. 1 Butler, in Hudibras, had 
described the Presbyterians as 

Those spider-saints, that hang by threads 
Spun out o*tk 9 entrails of their heads? 

The ass which stood for a pedantic critic in the Digression on 
Critics and for a modern in the Battle of the Books f becomes a Puri- 
tan preacher in the Mechanical Operation. The displacement of 
senses and reason by imagination and fancy, as in the Digression 
on Madness., reappears in the Mechanical Operation with the as- 
sociated vapours of the Aeolists. 4 Jack's soliloquy upon predesti- 
nation and the unreliability of the senses reappears in a direct 
attack by the author upon Dissenting preachers. 5 But the deepest 
self-echo is the assertion that enthusiasm, like the madness of the 
Digression on Madness, not only can be found in men of every time 
and place, but also c has been able to produce revolutions of the 
greatest figure in history' 6 ; for of course Swift accounted for 
Puritan fanaticism as merely the most poisonous form of that 
irrationality which springs eternally in animal rationale and which 
it must be our constant moral purpose to control. The flatulence 
which seems spirit to self-tricked hypocrites is mediately the 
vapours of all foolishness and ultimately the swelling of pride. 


Although the Mechanical Operation of the Spirit stands beside the 
Tale and the Battle through always being published with them, 
it stands far beneath them as a literary achievement. 7 One might 
treat the Battle as a generalization of the Digression on Critics; and 
one might treat the Mechanical Operation as a generalization of 

1 Tale, p. 271. a in. L 1461-2. * Tale, p. 233. 4 P. 273. 

5 Cf. pp. 192-4 and 275-6. Professor Pons has traced such connections in his 
Swift, pp. 383-6 and passim. 

P. 266. 

7 Professor James Clifford's arguments^ to the significance of this grouping do 
not seem to me persuasive ('Swift's Mechanical Operation of the Spirit*, in Pope and His 
Contemporaries: Essays Presented to George Sherburn [Oxford 1949], pp. 135-46). 


Jack's character in 'Section xi' of the Tale. But where the Battle 
dramatizes and enlarges a hint into a fresh, full, imaginative de- 
sign, the Mechanical Operation takes a characterization which is 
heavy enough, and, without lightening its crudities, extends and 
labours its applications. Here is another case of Swift's obsession- 
al violence drawing so much attention to itself that it distracts 
the reader from the author's satirical aim. 

However, the parts compensate, to a degree, for the whole, 
since there are paragraphs in the Mechanical Operation as fine as 
any but the most brilliant passages of the companion works. One 
of these is the opening of 'Section i', based on the notable conceit 
of a reversed allegory. Swift pretends that he wishes to discuss the 
power attributed to an ass, by some authorities, of carrying its 
rider to heaven. He then says the problem is so tendentious that 
he will consider it e by way of allegory', using Nonconformist 
preachers as a symbol of the ass and their congregations as a sym- 
bol of the rider. By this artifice he becomes free to speak as di- 
rectly as he wishes about the Presbyterians, Independents, Bap- 
tists, and Quakers, and yet, within the frame of his irony, to claim 
that they are not his subject at all. Merely as exemplifying the 
skilful elaboration of a difficult rhetorical figure, the passage is of 
interest. As ridicule, however, it seems too broad for its purpose : 
the author sounds not as if he were disengaged but as if he were 
struggling to appear disengaged. 

The best part of the essay is the long paragraph opening 'Sec- 
tion n'. Here Swift speaks with only the driest irony and with little 
disguise. Again, he makes the Puritans out to be but a special case 
of a general defect in humanity: our unwillingness to let 'right 
reason' work unimpeded, our itch to distort obvious moral judg- 
ments by the self-pleasing ratiocinations of 'discursive reason'. 
Comparing Europeans (and therefore Christians) with the wild 
Indians, he argues that at least the savages, however clumsy may 
be their conceptions of good and evil, never muddle the one with 
the other, 'nor ever suffering the liturgy of the white god, to cross 
or interfere with that of the black' * The English, by contrast, 

1 Tale, p. 274. 



have so troubled these boundaries that they cannot decide 
whether Nonconformity be the issue of divine or satanic powers. 
Swift characteristically suggests it is neither, though what he 
manifestly counts as absent is indeed the hand of God : 

Who, that sees a little paultry mortal, droning, and dreaming, and 
drivelling to a multitude, can think it agreeable to common good 
sense, that either Heaven or Hell should be put to the trouble of 
influence or inspection upon what he is about? 1 

What Swift means is that any body so contentious, stupid, and ir- 
rational as the 'modern saints' must be devoid of true piety or in- 
spiration. In this proposition resides the dialectical premise of his 
argument. But it was to be a mark of his satirical style that Swift 
should, at some point in an essay, pretend to dismiss the positive 
teachings on which he based his attack. Therefore, just as, in the 
Digression on Critics, he had 'discarded' the two respectable mean- 
ings of critic^ so also in c Section i' of the Mechanical Operation, he 
had rejected the present proposition actually the foundation of 
his rhetoric as irrelevant: 

If [spirit] be understood for a supernatural assistance, approaching 
from without, the objectors have reason. . . But the spirit we treat of 
here, proceeding entirely from within, the argument of these ad- 
versaries is wholly eluded. 2 

By implication, he comes perilously near using Hobbes's tokens 
by which a true prophet is known : c One is the doing of miracles ; 
the other is the not teaching any other religion than that which is 
already established'. 3 Not only the discussion at the beginning 
of 'Section n' but the whole web of the Mechanical Operation 
seems involved with Hobbes's materialistic doctrines concerning 
'Christian polities', spirit and inspiration, and prophets. 4 It is 
from the clash between, the disillusioned psychology which these 
doctrines support, and the stern morality which the author in- 

1 Tale, p. 276. * Ibid.> p. 271. 

3 Leviathan ra. xxxii (cd, A. R. Waller, p. 271). 

4 Ibid. ra. xxxii, xxxiv, xxxvi. 

R [2433 


vokes, that the energy of Swift's present discussion arises. As Pro- 
fessor French comments, 

[Swift] was a basically philanthropic man convinced against his 
will that Hobbes 9 s Leviathan is truer to human nature than the Ser- 
mon on the Mount. As a result, he often intellectually accepted 
what he instinctively and emotionally disliked. 1 


Except for these passages, however (and some fine aphorisms), 
Swift, in the Mechanical Operation, repeats, in a halting, crude 
form, what he had already and better expressed in A Tale of a 
Tub. The use of breaks and asides is neither witty nor meaning- 
ful. The puns, in the absence of comedy, seem more harsh than 
hilarious. The pervasive word-play smells stale because Swift dis- 
covered little of it. As Empson says, "spiritual words are all de- 
rived from physical metaphors. 52 The associations with the voca- 
bulary of Hobbes further undermine the irony, since in that tra- 
dition Swift's words have already lost the figurative half of their 
meaning. The tone of the satire is obsessional; there is much 
heavy sarcasm, little comic irony, almost no humour. 

One throws no lustre upon Swift by discovering, in the 
Mechanical Operation 'anticipations of Freudian theorems about 
anality, about sublimation, and about the universal neurosis of 
mankind'. Neither does it help to add that e Swiftian psycho- 
analysis differs from the Freudian in that the vehicle for the ex- 
ploration of the unconscious is not psychoanalysis but wit.' 3 Giv- 
ing an important message to a work which as literature has been 
misconceived does not improve its literary design; a noble sub- 
ject cannot make a poor speech into great oratory. This would be 
so even if Swift's adumbrations of Freud were deliberate. But 
they are not; they are unconscious. It would be more accurate to 
say Swift exemplified Freud's theorems than that he taught them, 

1 'Swift and Hobbes A Neglected Parallel', p. 243. 

fi William Empson, Some Versions qf Pastoral, 1938, p. 60. 

8 Norman O. Brown, Life against Death t 1959, p. X6. 



for what might appear to be demonstrations are in fact methods 
of ridicule: Swift employs 'anal' imagery as an aggressive, not 
a didactic, device. His meaning is not that anal preoccupations 
supply the energy for our loftiest principles but simply that the 
holder of lofty principles must not suppose they free him from the 

The province of literary meaning belongs to the public do- 
main. The unconscious, uncalculated aspects of literature are 
those which lump it together with the non-literary uses of langu- 
age. Any angry man yelling chamberpot sarcasms at his enemy 
is anticipating and illustrating Freud. 

Nevertheless, while Swift's violence distracts one from his 
meaning, it does not destroy that meaning; and a sympathetic 
critic will not misunderstand him. Swift never attacks the Non- 
conformists for having bodies, passions, faeces; he attacks them 
for confusing these with religion, for describing themselves as in- 
spired when they are windy, or as charitable when they are lust- 
ful. On the other hand, he never argues that charity is a form of 
lust : when he says so ironically, he is employing a clumsy style of 
wit to ridicule false religion and false charity. To claim that what 
he says in irony is true in fact, neither makes Swift wise nor re- 
deems his work. 

However, this interpretation goes far to explain why men who 
can enjoy large doses of obscenity and blasphemy in other forms 
certain commercial entertainments, party jokes, army humour 
find works like the Mechanical Operation revolting. Swift's ia- 
tensity and ingenuity bring readers so close to the Freudian truth 
that if they have already spent considerable energy dodging this 
truth in their own minds, the essay frightens them; and they find 
it not tedious or repetitious but noxious and unbearable. 

Swift's originality in the Mechanical Operation lies where he sets 
it, in the treatment of Puritan zeal not as irresistible possession or 
as scheming hypocrisy, but as a deliberate method of obliterating 
one's perceptions and suspending one's intelligence so that one's 
fancy and passions may be free. With his natural guides removed, 
the practitioner of the exercises may confer upon himself such 



grace and spirituality that his body no longer appears to weigh 
him down. While in his outward acts he may then behave like a 
beast, his ecstasy persuades him that God is his inspirer. 

To Swift, such zeal or enthusiasm seemed to blend a natural 
leaning with an artificial training, either of the two elements be- 
ing the germ of the disorder. He recognized both those 'whose 
once conscious art in the creation of ecstatic states had now be- 
come a sincere and involuntary part of their nature' and those 
'who were naturally enthusiastic but who employed artifice to 
arouse themselves and others' : 

The idea that artifice became nature seems to have been original 
with Swift; in fact, his subtle distinction between types of enthusi- 
asm is by far the most discerning of any of the seventeenth-century 
classifications. 1 

Although The Mechanical Operation of the Spirit is one of Swift's 
failures, it does suggest how much of his force as a writer origin- 
ates in the fight between his reformer's temperament and his dis- 
illusioned, pessimistic philosophy of morals. While the c fra-^ 
rnent' is cast as a satire against the hypocrisy of the DissentersJ the 
Dissenters are treated in it as natural products of the English 
character ; and the English are treated as no better than what one 
might expect of diseased humanity. Like the author of the Digres- 
sion on Madness, this author must count himself among those pos- 
sessed by corruption; and the last paradox, which nobody faced 
more directly than Swift., was how one of the damned themselves 
could retain sufficient integrity to recognize the general con- 
dition. Normally, Swift's resolution is ironical. By classifying 
himself among the accused, he forestalls the accusations of 
others ; he also intimidates the reader into assuming the opposite ; 
for one intuitively attributes virtue to the prophet who even 
ironically chastises himself. 

1 Clarence M. Webster, *S\vift and Some Earlier Satirists of Puritan Enthusiasm.% 
PMLA., XLVH (Dec. 1933), 1150. 


Chapter Six 

To appreciate Swift's hopes while he was writing his first 
masterpieces we are helped by noting the situations of some 
contemporaries who were to share with him the vicissi- 
tudes of friendship and the honours of literature. Matthew Prior, 
three years older than Swift, belonged to a far humbler family; 
but he came of a generous father, had grown up near Whitehall, 
and had, from boyhood, drawn the interest of noble patrons. He 
could therefore go to the Westminster School and to St John's, 
Cambridge fair parallels to Kilkenny and Trinity; and he took 
his B.A. degree the same year as Swift. The great Charles Mon- 
tagu, later Earl of Halifax, went to school and university -with 
Prior; and together they composed the burlesque of Dryden's 
Hind and the Panther which first gave Prior a reputation as a poet. 
After a few years in a college fellowship, Prior found himself 
chosen, through Montagu and other friends, as secretary to one 
of Temple's successors at The Hague: Lord Dursley, soon to be- 
come Earl of Berkeley. During the years when Swift sometimes 
saw William III at Moor Park, Prior was attending his majesty 
on official duties between Loo and Cleves. He played a small role 
in negotiating the Treaty of Ryswick, and it was he who carried 
that treaty from the Congress halls to London. By 1698 Prior not 
only was secretary of the English embassy in Paris, as well as 
secretary to the lords justices of Ireland (which he never visited) ; 
but he also was accustomed to the special protection, conver- 
sation, or correspondence of Montagu (now Chancellor of 
the Exchequer), the Earls of Dorset, Berkeley, Portland, and 
Jersey, and the Duke of Shrewsbury. He was well known to the 



king, to the main diplomatic corps of Europe, and to the officers 
of the English establishment. Yet he was only thirty-three years 

Swift's school- and college-mate, Congreve, fled from Ireland 
about the same time as Swift, who was his senior by three years. 
Coming to London as a student of law, Congreve had still a 
father to support him, and close relatives some, persons of rank 
and fashion near at hand. 1 His charming manners and mani- 
fest genius carried him easily into the smartest set of Dryden's 
London. When his little novel Incognita appeared, he was just 
twenty-two. A year later, his brilliant first play received unusual 
applause. By 1698 he held a small sinecure from the government; 
seemed the climbing sun of comic and tragic drama; and had 
sufficient power to patronize newer-comers in the theatre. 
Jeremy Collier, in fixing him among the central objects attacked 
in A Short View . . . of the English Stage (1698), may have left Con- 
greve 's plays to suffer for decades under that censure 3 ; but he also 
revealed the conspicuous height to which fame had so early lifted 
Swift's friend. 

It was Congreve, according to Steele, who first brought Mon- 
tagu and Addison together. With an ancestry as ecclesiastical as 
Swift's, Addison enjoyed the advantage of a scholarly, clergy- 
man father, whose tastes outran his decent income. From 
Charterhouse, Addison had gone to distinguish himself at Mag- 
dalen College, Oxford; and he was still there, as tutor, and ac- 
complished Latin poet, when he caught Dryden's notice with 
some flattering couplets. He soon appeared, like Congreve, in 
Tonson's series of miscellanies; and his contribution to the 1694 
volume was a survey of English poets which included one pane- 
gyric of Congreve and another of Montagu. While Swift was 
serving his term in Kilroot, Addison produced a poem addressed 
to the king but dedicated, in. verse, to Somers (then Lord Keep- 
er). The great man saw it; and so, *at the age of twenty-four, Ad- 

1 John G. Hodges, William Congreve, the Man (New York 1941), pp. 33~4- 
* Emmett L. Avery, Gangrene* s Plays on the Eighteenth-Century Stage (New York 
P- I- 


dison had gained personal access to the two powerful political 
patrons who better than any others in that age could forward a 
young man's career'. 1 The most splendid literary event of 1697 
was the publication of Dryden's Virgil; for this Addison supplied 
the preface to the Georgics; and in a postscript Dryden hand- 
somely praised his talents as a translator. Upon the Treaty of 
Ryswick, the same year, Addison wrote a Latin poem celebrating 
pax Guglielmi, and dedicated it to Montagu. It seemed in the 
order of things, no doubt, that such a youth should succeed to a 
college fellowship. But he could thank his prudent prosody that 
the Treasury granted him two hundred pounds for travel, and 
that Montagu persuaded Magdalen to continue Addison's sti- 
pend while the young poet went on a grand tour of the Continent. 
Meanwhile, Addison's friend Steele, five years younger than 
Swift, was still an officer in the Coldstream Guards. He had been 
born in the parish of St Bride's, Dublin, to which Swift's uncles 
Godwin, William, and Adam belonged : both Godwin and Adam 
were vestrymen at the church where Steele was baptized, and 
Steele's father was admitted an attorney at the King's Inns two 
years after Swift's. Not only must the families have been ac- 
quainted with one another, but Godwin may have helped the 
elder Richard Steele to obtain preferment from Ormonde 2 ; how- 
ever, since the great (first) duke's secretary was the elder Steele's 
brother-in-law, no further pressure may have been needed: cer- 
tainly it was the duke himself who had young Richard (fatherless 
from the age of five) placed upon the foundation of the Charter- 
house, where he first met Addison. From that school, Steele, with 
the aid of his influential uncle, went on to Christ Church, Ox- 
ford; but instead of staying long enough to collect a degree, he 
enlisted in the young (second) Duke of Ormonde's regiment of 
guards. Now he advanced his career by dedicating to Lord Cutts 
(Colonel of the Coldstream Guards) a poem on the death of 
Queen Mary; for Cutts soon made him an ensign in his own regi- 
ment and a member of his household. For a while Steele acted as 

1 Peter Smithers, The Life of Joseph Addison (Oxford 1954)* P- 33- 
* G. A. Aitken, The Life of Sir Richard Steele (1889), pp. 13-15- 



his confidential secretary, performing services much like those 
which Swift, then in his last years at Moor Park, was performing 
for Temple. 

If these men understood what elements had to be added to 
their natural endowment for them to gain the independence and 
recognition they desired, Swift was surely competent to draw the 
same inferences. He had something like their aims a respect- 
able income, an entrance into the great world, literary fame; he 
had a similar education, a similar power to charm those whom he 
wished to impress, and a superior genius. His 'backing' was equal 
to the average of theirs ; his difficulties were no greater : Gongreve 
and Steele came from Ireland; Prior's family and connections 
were notably meaner than Swift's; Addison was shy; Steele was 
an orphan. Unfortunately, Swift chose to work through the dying 
order of Temple's circle rather than through the instruments 
which belonged to the future. So in 1697 as we have seen he 
approached the Earl of Sunderland, whose great gift was for sur- 
viving disgrace. In spite of his notorious treacheries and long re- 
tirement, Sunderland had been appointed Lord Chamberlain in 
April. But e a House of Commons not distinguished for idealism, 
[or] even probity', displayed such mutinous distaste for his name 
that he resigned at the end of the year. 1 Swift commented, 'My 
lord Sunderland fell and I with him,' 2 

He immediately looked elsewhere, but we are not sure whom 
he approached. 'There have been other courses', he told Winder, 
* which if they succeed, I shall be proud to own the methods, or if 
otherwise, very much ashamed.' 3 As Craik indicates, this is one of 
those hypocrite-in-reverse expressions by which Swift charac- 
teristically twists an interpretation of motives against himself: 
6 Its probable meaning', says Craik, 'is, that though he is sure of 
the honesty of his means, he will still be ashamed of having tried 
at all, if these means do not end successfully.'* We may guess 
from what happened a year later that Swift directed one applica- 
tion to Suixderland's handsome, middle-aged uncle, Henry Sid- 
ney, who, though now Earl of Romney, was still closer to Sir 

1 Ogg it. 441 . Ball t. 34. 8 Ibid. 4 Craik I. 76, n. a. 



William Temple than anybody else outside Moor Park; and 
we may further guess that Swift aspired to some well-endowed 
preferment in the church, such as the prebend of Westminster 
held by his own uncle Thomas's friend, Dr South. How his hopes 
were met, we shall see shortly. 


Although a career in England offered vistas which nothing in 
Ireland could match, the Established Church promised as little 
serenity in the one kingdom as in the other. On the contrary, the 
schisms now troubling the English clergy were soon to spread 
across the Irish Sea, so that wherever Swift settled he would be 
involved in them. Since these splits were both doctrinal and in- 
stitutional, they opened the way for a heavy flow of polemical 
composition. In addition to the growing inroads of Noncon- 
formity and the declining menace of Roman Catholicism, a new 
threat was looming: the congeries of rationalist attacks upon 
Anglican theology which are fuzzily associated with the rise of 
deism. In the adumbrative form of the Trinitarian controversy, 
this movement touched the institutional side of the church, since 
it supplied a weapon for the aggressive party in the quarrel over 

Between 1693 and 1695 the old system of licensing books be- 
fore publication was allowed to die, and men could write boldly 
upon religious arguments that once would have exposed them to 
prosecutions. A government which necessarily distinguished its 
own policies from the intolerance of James II and Louis XIV 
found it difficult to set limits upon freedom of the press. Yet this 
was the time when Socinianism was spreading, Unitarianism was 
emerging, and orthodox churchmen were put to it to defend the 
Trinity without stumbling into Tritheism. Stephen Nye, a Hert- 
fordshire parson, made propaganda for the 'Unitarians, called 
also Sociriians' ; William Sherlock, undertaking to vindicate the 
doctrine of the Trinity, found himself accused by Robert South 
of Tritheism; whereupon, Nye charged South with Sabellianism. 


The king intervened in 1 695, and prohibited further speculations 
on the subject; but he only succeeded in shifting the focus of the 
incandescence to Arianism. 1 

In 1697, with the appearance of A Letter to a Convocation Man, 
the Trinitarian controversy joined streams with another. The 
war had deepened the usual breach between court followers and 
country landowners: while those in the king's employ could 
enrich themselves with places and perquisites, the agricultural 
gentry felt crushed by the land tax which was paying for the 
armies. Toward the middle of William's reign, furthermore, a 
natural reaction had turned the 'Country' and 'Church' parties 
against the principles which had inaugurated his reign. One 
Parliamentary group, with Robert Harley at its head, tried to 
limit the king both by reducing grants to the crown and army and 
by advocating the 'Place' and 'Triennial' bills, familiar Country 
programmes. Other opposition groups, led by the Earl of Ro- 
chester and the Earl of Nottingham, argued that the new order, 
through its tolerance, was undermining the Established Church. 
Since William, after the elections of 1695, relied increasingly 
upon 'Court Whigs' like Somers and Montagu, the forces of the 
opposition had good reason to draw together against him; and 
the counter-Revolutionary swing of the nation gave them sup- 
port. 2 As part of their strategy, Harley, Nottingham, and Ro- 
chester encouraged sympathetic clergymen to write attacks upon 
the ecclesiastical policies of the government. 'From 1697 there is 
to be found the coming together of a recognizable set of divines, 
steadily encouraged by the anti-ministerial opposition and de- 
voted to the task of bringing King, ministers, and their ecclesias- 
tical appointees into popular disrepute.' 3 

When John Toland and Stephen Nye brought out some 
unusually provocative pamphlets teaching Socinian doctrines 
(1696, 1697), the clerical critics of the ministry used these among 
other excuses to demand a Convocation of the Canterbury Pro- 
vince of the Church of England. Although the machinery for 

1 Norman Sykes, William Wake (Cambridge 1957)* n. 153-4. 
* G. V. Bennett, White Kennett (1957), p. 26. * lbid.> p. 29. 



such a gathering existed, and although the York province met 
regularly (and quietly), no Convocation of Canterbury had as- 
sembled since 1664, apart from a brief and explosive sitting in 
1689. Now Francis Atterbury, aided by associates, 1 produced a 
specious, ingenious alarum, A Letter to a Convocation Man. In Hol- 
land, he said, 'Socinians and other anti-Trinitarians' were claim- 
ing the English church for their camp; and in England there 
seemed to be c an universal conspiracy' among Deists, Socinians, 
Latitudinarians, and other deniers of mysteries, 6 to undermine 
and overthrow the Catholic Faith'. A secular Parliament was in- 
competent to strangle these monsters: the bishops (the upper 
house of Convocation) and other clergy (represented in the 
lower house) must meet to fight them. 

That such a meeting would be turbulent, both sides knew. The 
court had regularly been giving bishoprics to co-operative men; 
and the nonjuror schism had widened the normal scope for such 
appointments. Among the old-fashioned lower clergy, however, 
distrust of the government had been growing, and they were 
baying for a heresy hunt. Obviously, whoever brought the two 
groups face to face would be inviting a melee. 

Swift's exact sentiments are unknown; but it is safe to con- 
jecture that while he would have supported the move to call a 
convocation, and would have welcomed any bridling of heretical 
or anti-clerical pamphleteers, he would have opposed a project 
to define the mysteries for dogmatic purposes, and would have 
discouraged an attempt to turn the clergy against the govern- 
ment. However, the issues were to travel much farther before he 
was compelled to face them directly. I notice them at this stage as 
important and provocative aspects of the environment in which 
he made his next decisions. If he persisted in seeking an ecclesi- 
astical career, it would not be for thinking England an easy field 
for ghostly cultivation; and if he sought preferment without 
emulating Atterbury or Kennett (who answered Atterbury's 
book and was made Archdeacon of Huntingdon), it was not be- 
cause their methods appeared foreign to their profession. 
1 Sykes, William Wake i. 8i-a. 




Meanwhile, although Swift continued to exchange letters with 
Varina, he could still be reproached Tor the want of gallantry, 
and neglecting a close siege'; and he told Winder (April 1698) 
that some people in Ireland had written to him 'censuring 
[Swift's] truth in relation to a certain lady'. 1 Yet when Winder 
himself alluded to e a dangerous rival for an absent lover'. Swift 
responded as tranquilly as to a weather forecast : *I must take my 
fortune. If the report proceeds, pray inform me. 32 

He could afford such serenity because he owned other re- 
sources; a new and more steadfast star had risen for him : 

at sixteen 
The brightest virgin of the green.* 

This was the girl, nearly fourteen years his junior, whose moral 
code and literary taste he had modelled upon his own. She was, 
like Miss Waring, the eldest child of a widowed mother; and 
through the first years of Swift's friendship her health too was 
fragile. Not long before Esther Johnson's birth, Temple's only 
daughter the 'child he was infinitely fond of* had died of 
smallpox. Temple's sister says he enjoyed the company of child- 
ren, whose 'imperfect language and natural and innocent way 
of talking' delighted him. 4 So he must have felt drawn to the little 
half-orphan of his late steward. In a document dated 1690, 
Temple acknowledged an indebtedness to Bridget Johnson of 
14.0; and there are annual receipts, signed by her, for interest 
payments of six or seven pounds. 5 This may possibly have been 
his method of investing a nest-egg for mother and daughter and 
insuring it against his death. 

On Swift's return in 1 696 to Moor Park, he found Hetty John- 
son to be 'beautiful, graceful, and agreeable*. e She had black, 

1 Ball i. 19, 22. a Ibid. t p. 30. * Poems u. 721. 
* Temple, Early Essays, pp. 15, 21, 29. 

5 The documents are nos. 9 and 10 of the Temple family papers in the collection 
of Mr James Osborn. (In 1695 Mrs Johnson withdrew twenty pounds of her 

6 T.Scott xi. 128. 



black hair and the enchanting gift of docility. 1 Almost twice her 
age, Swift could regard his own attitude as harmlessly tutorial. 
Even confined to the narrow dimensions of the same household, a 
priest in his fourth decade of life and a maid still in her teens 
might converse freely and frequently without inviting suspicion. 
When she had a mother, brother, and sister to live with, and he 
was studious, quippish, and chaste, what intrigue could follow ? 
Swift's parental address to Esther, teasing when not didactic, 
must have misled him, her, and their friends. It was a quieter 
game than Heloi'se and Abelard but not less tricky. With such 
pleasures in his reach (and rarely out of view), Swift was not 
languishing for Varina. 

The closest friend Miss Johnson had besides Swift also be- 
longed to the Temple circle. She was Rebecca Dingley, Sir Wil- 
liam's spinster cousin. Although much older than Esther, Mrs 
Dingley had no adequate fortune, and probably relieved Lady 
Giffard of some housekeeping duties ; she certainly was a com- 
panion and in emergencies waited on her ladyship. The status of 
'Hetty' and 'Dingley' is sketched in. Lady Giffard's masterly ex- 
position of the servant problem; her ladyship has just dismissed a 
new maid for being too dainty to work properly : 

Papa [i.e., Temple] desires she may goe for fear he should be in love 
with her. she owns never to have wash'd a room in her life, & when 
she rose next morning asked if she must make her owne bed, com- 
plain 9 d yt she should not be able to wait & worke in a room with out 
a fire & you know I have no other, but ye cruel thing of all was dine- 
ing with the common servants, & yt she said she had never reckon'd 
upon & doubted she should not be able to bear. Indeed I belceve 
nobody ever had a greater damp at heart yet when I spoke to her 
to day (how sorry I was to have put her upon what I doubted she yt 
had alwayes lived soe much better must thinke very hard, & she 
could not be more uneasy then I should be every time I did it, but 
twas such a servant I wanted) , she said she had rather venter upon 
any service then live longer out & in yt I beleeve told ye true reason 
but she had lived I find in great places & bin an absolute gentle- 

1 We may doubt the report that when her remains were dug up in the nineteenth 
century her teeth were judged to be 'perhaps the most perfect ever witnessed in a 
skull' (JV. & Q. t i a Aug. 1 87 1, p. 124). 



woman & I dare say is one, by her name & her friends for I have 
had a great deal of talke with her her journey shall be pay'd for & 
Mrs Bradley yt came downe with her will goe up with her agin a 
Monday Nanny lagger who I had orderd to goe yt day says she is 
very happy to stay longer & soe I am going in search of my pussle 
again, 3 gentlewomen had bin a little too much state; for I make 
use of my cousin Dingley when ever I am in want, Hettys place 
being ye heigh [t] of her [i.e., Dingley's] ambition. 1 

Of course, friends and relatives were always visiting the 
family; and a grandson of Temple's brother later recalled to 
Swift how the young parson was sometimes drawn into those 
occasions : 

When I was [a schoolboy] , I remember I was committed to your 
care from Sheen to London. We took water at Mortlake, the com- 
mander of the little skiff was very drunk and insolent, put us ashore 
at Hammersmith, yet insisted, with very abusive language, on his 
fare, which you courageously refused ; the mob gathered ; I expect- 
ed to see your gown stripped off, and for want of a blanket, to take a 
flight with you in it, but ... by your powerful eloquence you saved 
your bacon and money, and we happily proceeded on our journey. 3 

Expeditions to London were not Swift's only errands for Temple. 
He kept the household accounts, acted as amanuensis, paid out 
moneys, and supported conversation. Of course, as Lady Giffard 
remarks, the great man was subject to periods of spleen and de- 
pression in his last years; and Swift felt their effect. 'Don't you 
remember', he wrote to Esther Johnson years later, c how I used 
to be in pain when Sir William Temple would look cold and out 
of humour for three or four days, and I used to suspect a hundred 
reasons?' 3 But there was cosiness as well; and when Swift at 
forty-five, visiting the Lord Treasurer, was given a shilling to 

1 B.M. MS. Egerton 1705, ff. 17-18: letter of 30 Dec. [?i697], printed inaccur- 
ately in Longe, p. 215. A sign of Esther Johnson's status is that in the bills for 
mourning to be worn upon Temple's death she is entitled *Mrs. Hetty* suggesting 
a very young gentlewoman whereas the ordinary servants' surnames, Christian 
names, or both, appear with no honorific. (See nos. 19 and 21 of the Temple 
family papers in Mr James Osborn's collection.) 

* Ball rv. 66-7 (18 Mar. 1729). * Journal 4 Apr. 1711. 



stake in a family game, he wrote, C I was playing at one and thirty 
with him and his family tother night, he gave us all 12 pence 
apiece to begin with: it put me in mind of Sir W T.' 1 

Advancing years did not sharpen Temple's sense of responsi- 
bility to Swiftj whose pursuit of knowledge and friendship left 
him as far as ever from a happy settlement. Glimpses of Moor 
Park during this time are given by Lady GifFard's letter to her 

I have sent him [i.e.. Swift] with another compliment from Papa 
[i.e., Temple] to the King, where I fancy he is not displeased with 
finding occasions of going. . . Thank God Papa is not very bad, I 
hear him just now going down stairs with a lame knee. . . Papa 
wishes you here too, at a piece of roast beef at dinner which he eat 
for the first time with a very good stomach and I must tell you he is 
now a great deal better, whatever comes after it. . . We are got 
hither [i.e., to Petworth] at last, and Papa I thank God very well, 
and insufferably pert with winning twelve guineas at crimp last 
night. The Duke of Somerset says he never remembers seeing him 
better. 2 

The Petworth visit, September 1698, was one of Temple's last 
holidays. Exhausted by what was called the gout, he had already 
begun his final struggle with illness. Swift, in a paper which he 
titled Journal d'Estat de M* T devant sa Mort, kept a register of 
his dying friend's fluctuations, from i July until 27 January fol- 
lowing, when, a closing note recorded his immediate reaction to 
Temple's end: e He dyed at one o'clock in the morning and with 
him all that was great and good among men/ 3 When he had time 
for a fuller appreciation, Swift's language was even more ex- 

He was a person of the greatest wisdom, justice, liberality, polite- 
ness, eloquence, of his age and nation; the truest lover of his coun- 
try, and one that deserved more from it by his eminent public ser- 
vices, than any man before or since: besides his great deserving of 
the commonwealth of learning; having been universally esteemed 
the most accomplished writer of his time. 4 

1 Journal 9 Oct. 1712. a Longe, pp. 216, 198, 202, 227. 
8 Lyon, preliminaries, f. 6 V . * W. Scott i. 43. 



Besides such formal though honest tributes. Swift expressed 
less gracious sentiments upon his master's death. These appear 
on an odd leaf headed, c When I come to be old 1 699 '. Although 
Swift does not in fact indicate their occasion, the resolutions 
sound like the suppressed conclusions of a young man's close, pro- 
longed observation of one or perhaps several old men. Most of 
the animadversions could reflect upon Temple; others may al- 
lude to visitors like Romney. 

Not to marry a young Woman. 

Not to keep young Company unless they realy desire it. 

Not to be peevish or morose, or suspicious 

Not to scorn present Ways, or Wits, or Fashions, or Men, or [ ? War] , 


Not to be fond of Children, for let them come near me hardly! 
Not to tell the same Story over & over to the same People 
Not to be covetous 
Not to neglect decency, or cleanlyness, for fear of falling into 

Not to be over severe with young People, but give Allowances for 

their youthfull follyes, and Weaknesses 
Not to be influenced by, or give ear to knavish tatling Servants, or 

Not to be too free of advise nor trouble any but those that desire 

To fPCompell ^desire/ some good Friends to inform me wch of 

these Resolutions I break, or neglect, & wherein; and reform 


Not to talk much, nor of my self. 
Not to boast of my former beauty, or Strength, or favor with Ladyes, 

Not to hearken to Flatteryes, nor conceive I can be beloved by a 

young woman, et eos qui herdeditatem [sic] captant f Pomnes 

?del odisse ac vitare 
Not to be positive or opiniatre 
Not to sett up for observing all these Rules, for fear I should observe 

none. 1 

1 The MS. (often printed) is in the Forster Collection of the Victoria and Albert 
Museum. I have queried doubtful readings and put them between square brackets, 
used half-brackets around cancelled words, and used half strokes around inserted 



Obligatory attendance upon Sir William would provoke the 
resolutions against seeking the company of young people, against 
peevishness, against scorning modern manners, against repeating 
stories, against too much talking and boasting. We know how 
vain Temple was and how much he enjoyed his own anecdotes; 
we have seen that he was notably old-fashioned, sometimes mo- 
rose, and extremely fond of children. Temple's pride in his gal- 
lantries may account for the January-May theme, or it may be 
due to the conversation of rakish guests. Other motifs are prob- 
ably casual thoughts on old age, called up by the common as- 
sociations of the subject. Some cautions smell of literary sources: 
the Latin warning about legacy-hunters, the desire for friends' 
admonitions. 1 Finally, the attitude towards servants suggests 
that Swift, having to deal directly with the humbler members of 
the establishment, came to the traditional view of life below 

More significant than the particular opinions expounded is 
the act of writing them down. Only to certain minds would such 
a compilation appear a wise method of regulating one's conduct. 
Swift's taste for this sort of list is characteristic of him. Not only 
manners but morality, he thought, could be codified in sets of in- 
structions. Although the habit belongs to an epoch, and we find 
Descartes, Locke, and Franklin among the manufacturers of 
such programmes, we expect a great moralist to have reserves of 
insight that would line Swift's dicta with irony. Here again is 
evidence that his nobility was intuitive, that his soaring moral 
indignation starts from another spring than his explicit, didactic 
propositions. So I think the reverently admiring tone of the eulo- 
gies of Temple reveal more of Swift's essential nature than the 
naivet of his resolutions. 

In Temple's rosary of virtues, however, Swift could not count a 
zeal to further the interests of a dependent. Instead of providing 
his secretary with an income or a start in public life, Temple left 
him a modest acquaintance with court circles, the disposition of 
several unprinted works, and a legacy of a hundred pounds. Of 

1 Gf. Gil Bias vn. iu-iv. 

s [259] 


these advantages, the only one mentioned by Temple in his -will 
is the money, c to Mr. Jonathan Swift, now dwelling with me 3 , 
Bridget Johnson, who continued to work for Lady GifFard, re- 
ceived twenty pounds and half a year's wages; young Esther, 
identified as "servant to my sister', received the valuable lease of 
certain lands in Ireland. Except for Swift, all the legatees were 
servants or relatives ; nothing went to Thomas Swift, although he 
witnessed the main part of the will. 1 

Swift stayed on briefly at Moor Park, settling Temple's affairs 
and paying off legacies. A batch of receipts is preserved, endorsed 
by him, for such disbursements as a journey by himself to Lon- 
don, and a waistcoat and breeches for himself; books bought by 
Temple; funeral clothes for Rebecca Dingley, Mrs Johnson, 
Swift, and others; a quarter-year's land tax; and the interment 
fee for Temple's burial in Westminster Abbey. 2 

Next, he had to look after his own fortune ; and on this step we 
have no guide but Swift's words. He still hoped the king might 
present him to a prebend. Romney seemed a proper person to 
put William in mind of Mr Swift's merit. But the noble lord 
either failed or never tried to do the business. After months of 
profitless waiting, Swift took an unattractive post as private chap- 
lain to a new lord justice of Ireland. (This at the very time when 
Matthew Prior, though still in Paris, held the place of first secre- 
tary to the lords justices of Ireland.) Swift's account has the ring 
of bias: 

Upon this event [i.e., Temple's death] Mr Swift removed to Lon- 
don, and applyed by petition to King William, upon the claym of 
a promise his majesty had mad[e] to Sir W[illiam] Tfemple] that he 
would give Mr Swift a prebend of Canterbury or Westminster. The 
Earl of Rumney who professed much friendship for him, promised 

1 Because of the many deaths in his family, Temple drew up several wills, of 
which this was the last. Dated 8 Mar. 1694/5, it in turn had to be emended when 
one of his brothers died; so he added a codicil dated 2 Feb. 1697/8. Thomas Swift 
witnessed only the original version of 1695; Jonathan Swift's legacy is defined in 
the codicil (Courtenay n. 484-6), 

2 Documents in the collection of Mr James Osborn. Since the books may have 
been read by Swift while he was writing the Battle qfthe Books, I have listed them 
in Appendix G. 



to second his petition, but, as he was an old vitious illiterate rake 
without any sense of truth or honor, said not a word to the King : 
And Mr Swift after long attendance in vain; thought it better to 
comply with an invitation given him by the E[arl] of Berkeley to 
attend him to Ireland as his chaplain and private secretary; his 
lordship having been appointed one of the Lords Justices of that 
Kingdom. 1 

No prebend of either cathedral had in fact fallen vacant during 
the time of suspense. Of course the king could have promoted one 
of the prebendaries to free a stall for a petitioner. But Swift had 
no right to consider himself a correct object for such distin- 
guished care. Furthermore, it is just possible that Romney in- 
troduced him to Berkeley. Temple and his sister were indeed 
familiar with the earl; but that they were on visiting terms, or 
friendly enough for a recommendation from either to succeed 
with him, is not certain. Romney, however, had been a lord 
deputy of Ireland (though a blundering one) 16923; and he 
might have directed Swift to the modest opportunity when it 
arose. Berkeley was fifty years old at the time. Short, fat, and 
gouty, he had waded through a mediocre cursus honorum of lord 
lieutenancies, commissions, and envoyships, but never held a 
post of ministerial rank. In a record of minor diplomatic missions, 
his greatest state had been as Envoy Extraordinary at The 
Hague, 1689-94, where his secretary for most of the time had 
been Prior. With Berkeley, Swift left for Dublin, sailing from 
Bristol in August 1699. 


Why did Swift not push his chance more shrewdly ? He published 
nothing dedicated to men in real power; he performed no service 
that might be rewarded by the ministry; he attached himself to 
no one besides Temple ; he won no college fellowship ; he took no 
parochial living. Perhaps his ineptness reflects a half-conscious 
wish to prolong the sollicitae iucunda oblivia vitae of Moor Park 

1 Autob., f. 9 V . 



'which no time', he said eight years afterward, 'will make me 
forget and love less 5 . 1 To the excitement of Hetty's presence and 
the amenities of Temple's household, he might add the pleasant 
landscape, the nearness to London, the advantage of meeting 
distinguished company. When he was an old man, his memory 
could still dwell upon the grounds fronting the mansion : 'The 
tree on which I carved those words, factura nepotibus umbram, is one 
of those elms that stand in the hollow ground just before the 
house.' 2 However, there are so many signs of his longing, at 
thirty, for advancement, that a contrary wish should not be 

Perhaps Temple advised him foolishly. No doubt. But why did 
he take the advice ? Retired courtiers, like exiles, always under- 
estimate the decline of men whom they knew in place and the rise 
of those they did not. For Swift to have laid his money on such 
sway-backed jades as Sunderland and Romney, when all the 
world could see that Montagu, Somers, Wharton, and Portland 
were rising, bespeaks either the most cloistered innocence or 
darkening counsel. 

We must ask how Swift might have preferred this guidance to 
the judgment born of his own observation. He had seen Atter- 
bury, long before the Phalaris controversy, gain a pulpit in Lon- 
don and a royal chaplaincy for defending their church against 
Roman Catholic polemics 3 ; he knew the details of Congreve's 
success; he heard of others'. Obviously, to find glory or prefer- 
ment, one should first have identified oneself with the leader of an 
active party in literature, the church, or the government, and 
should then have studied to make oneself appear indispensable. 
If one neither contributed to controversy, nor brought lustre to a 
patron's name, nor did draught-horse work for a party project, 
nor caught the affection of a great, great personage, the road up 
was an, impasse. Swift could have written on the Convocation 
affair; he could have defended the Trinity; he could have glori- 

1 Ball x. 57. * Ibid. v. 41 7. 

* I assume that Atterbury's lectureship at St Bride's (1691) was largely a reward 
for his Anglican apologetics of 1687. 



fied the Peace of Ryswick; he could have dedicated to somebody 
other than the king. But he chose not to. 

From Swift's actions and remarks in the decade which fol- 
lowed, we may believe that he had ample intuition of all this. He 
should also have had enough experience of Temple to distrust the 
baronet's precepts supposing they were in fact what Swift re- 
lied upon. I think he left too much to Temple ; or, in a practical 
not moral sense, expected too much from him. Ignoring ten 
years of evidence, Swift was pretending to believe that Sir Wil- 
liam would act as the father Swift would like to have had. In the 
disappointment which followed, he poured over Romney's char- 
acter the acid which ought to have fallen on Temple's. This trans- 
fer of blame may suggest that Swift realized, underneath, that his 
own miscalculation, and not Temple's inertia, was the effective 
source of his disappointment. A quarter-century later, in a 
spasm of anger, he adjusted the implicit balance between Rom- 
ney and Romney's old friend : 

I own myself indebted to Sir William Temple, for recommending 
me to the late King, although without success. . . [But] I was at his 
death as far to seek as ever, and perhaps you "will allow that I was 
of some use to him. This I will venture to say, that in the time when 
I had some little credit I did fifty times more for fifty people, from 
whom I never received the least service or assistance. 1 

In view of this outburst, I think it fair to suspect that Swift had 
originally sacrificed his small faith in Romney to save his deep 
admiration for Temple. It was an excellent bargain. 

On one account, however, Swift possessed little freedom to 
choose a path; for he had already done what few of his contem- 
poraries dared to try before their careers were established : he had 
risked the effects of independence. Prior clung, without swerving, 
to Montagu; Congreve never broke with Dryden or the court; 
Addison and Somers kept up, to the end, a cordial, if sometimes 
thin, friendship; even the quixotic Richard Steele only left Lord 
Cutts after a dedication, to Lady Albemarle won him a captaincy. 

x Ballra. 301. 



But Swift had not only followed the evening rather than the 
morning star: he had abandoned Temple's shelter without the 
resource of another patron. In his own native city, however, he 
had needed Temple's good word to admit him even into a course 
which Temple himself had discouraged. At Kilroot the desola- 
tion of Swift's life had refurbished the appeals of Moor Park. 
When the offer of renewed 'acquaintance with greatness' arrived, 
Swift seized it. By returning to his master, however, he gave up 
the power to act without Temple's approval. He locked himself 
within a circle which the old man's death would have to break. 

Ultimately, moreover, the questioner is flung back upon an 
interpretation of Swift's own character. Whether or not he took 
advice from Temple, Swift certainly committed faults of strategy 
which were corrected in succeeding years. I think that he was 
fundamentally so unsure of himself, though aware of his gifts, at 
this era, that he held passively to the appearance of calm and 
strength. 'Little disguises', he wished to believe, 'were infinitely 
beneath persons of [my] pride . , . paltry maxims that they are, 
calculated for the rabble of humanity.' 1 Without appeasing the 
usual gods, he was offering up the usual prayers. He looked for 
the blessings of prudence but hugged the words of his Muse : 

Stoop not to inf rest, flattery, or deceit; 

Nor with hir'd thoughts be thy devotion paid? 

Unfortunately, if c dirty paths where vulgar feet have trod* 3 are 
not the way to Parnassus, they are, for a man in Swift's case, the 
way to a thousand pounds a year. Yet he could hardly disdain 
others for taking it unless he refused it himself. It was to be his 
lesson for the next ten years that genius, charm, and integrity, 
all together, could not exempt him from the common doom. 

1 Ball i. 20. * Poems I. 55. * Ibid. 9 p. 49. 




The Family of Swift's Father 

These are the children of Thomas Swift, vicar of Goodrich, Hereford- 
shire, as their baptisms are recorded in the MS. parish register. The 
original entries are in Latin. Where the English form of the name is not 
perfectly certain, I give the Latin as well. It is possible but unlikely 
that other children of Thomas Swift were baptized elsewhere. The 
Abraham Swift (died in Dublin 1686) listed by Ball (vr. 215) among 
the sons of Thomas Swift of Goodrich does not seem to belong to this 
family. He is not among the children baptized at Goodrich ; he is not 
named in Thomas Swift's will ; and he is not mentioned in Jonathan 
Swift's autobiographical essay, or in his correspondence or works. 

Mary ('Maria') : 12 July 1626 

Godwin: 23 January 1627/8 

Dryden: 15 November 1629 

Emily ("Aemilia") : 9 July 1631 

Elizabeth: 4 November 1632 

Thomas: 8 December 1633 

Sarah: 7 February 1635/6 

William: 13 August 1637 

Katherine ('Katherina') : 24 March 1638/9 

Jonathan: 24 May 1640 

Adam: 21 March 1641/2 

The following events, relating to Thomas Swift*s family, are also 
recorded in the Goodrich register: 

4 May 1658: Thomas Swift buried. 

5 May 1662 : Thomas Vaughan and Elizabeth Swift married. 
28 December 1664: Jonathan, son of Thomas Vaughan and Eliza- 
beth his wife, baptized. 

26 December 1666: Godwyn, son of Thomas Vaughan and Eliza- 
beth his wife, baptized. 

25 September 1670 : James, son of Thomas Vaughan and Elizabeth 
his wife, baptized. 



The Will of Thomas Swift of Goodrich 1 

Thomas Swift clerpcus] 

In the name of God amen. I Thomas Swift of Goodridge in the 
county of Hereford beinge sick in bodie but of sound and good me- 
morie (blessed be the allmighty God for it) doe make and ordaine this 
my last will and testament in manner followinge : ffirst I giue and be- 
queath my soule to allmighty God trustinge to be saued by the mer- 
ritts of my redeemer Jesus Christ And my body to be buried at the 
will and discrecon of my welbeloued wife and my executor herein 
afternamed And as for that smale personall estate wherewith god 
hath blessed mee I giue and devise the same in manner following 
Vizt) I giue and devise all my corne now groweinge vpon all or anie 
of my landes togeather with all my householdstuffe as plate, brasse, 
pewter, linnen, of all sorts bedds and beddinge woodden vessells, 
tables chaires stooles and other the necessaries of the house and all 
provisions of all sorts as bacon, beefe, butter, cheese, wheate, rye, 
barley, mault, and beere of all sorts to my deare and welbeloued wife 
Item I giue and devise vnto my foure sonnes Thomas William 
Jonathan and Adam, and to my foure daughters their seuerall por- 
cons beinge already paid or secured to them) the summe of fiue 
shillings a peece Lastly I giue and devise that my eldest sonne Godwin 
shall pay all my debtes of what kinde soever and to enable him here- 
vnto I giue vnto my said sonne Godwin all the rest of my personall 
estate as horses, oxen, sheepe, lambes, swine hopps in the house ready 
sacked woods cutt and corded on my lands, waines, carts ploughes, 
plough timber and other implements and necessaries of husbandry 
and other my personall estate not herein before bequeathed and I doe 
alsoe make and appointe my said sonne Godwin executor of this my 
last will and testament In witnes whereof I haue herevnto sett my 
hand and scale the foure and twentieth day of Aprill one thousand 
six hundred ffiftie eight: Thorn: Swift: Sealed and published. In the 

1 Principal Probate Registry, Somerset House, 1658, f. 697, Wootton. 



presence of Elizabeth Swift the marke of Edward Lambert the marke 
of Jane Howells the marke of Mary Williams Katherine Swift. 
This will was proved at London the sixth day of the moneth of 
December in the yeare of our lord God one thousand sixe hundred 
ffiftie and eight: before the judges for probate of wills and grantinge 
administrations lawfully authorized by the oath of Godwin Swift the 
naturall and lawfull sonne of the said deceased and sole executor 
named in the said will : to whome was committed administration of all 
and singular the goodes chatties and debtes of the said deceased : hee 
the said Godwin Swift beinge first sworne in due forme of lawe by 
vertue of a commission well and truly to administer the same. 



Swift's Maternal Grandfather 

The identity of Swift's mother's father has been an object of investi- 
gation at least since John Nichols took up the search while working 
upon his History of Leicestershire. Until recently, however, little progress 
was made beyond what Nichols himself suggested. Though his state- 
ments are not all consistent with one another, the drift of the argument 
is that Abigail Erick was sister to Thomas Errick, vicar (1664-81) of 
Frisby-on-the-Wreake, Leicestershire, and that her niece, Thomas 
Errick' s daughter, married John Kendall, vicar (1684-1717) of 
Thornton, Leicestershire. 1 If we can substantiate these conclusions, 
we can identify Abigail Erick' s father, since it is now known that 
Thomas Errick was the son of James Ericke, vicar (162734) of 
Thornton, Leicestershire. 2 By showing her to be James Ericke's 
daughter, we can account for two further puzzles which have plagued 
Swift's biographers : first, why no record of Abigail Erick's baptism 
has been found in Leicestershire; secondly, how a woman from that 
county happened to marry, in Dublin, a recent immigrant from 
Herefordshire. For it will appear that James Ericke emigrated to 
Ireland toward the end of 1634. Although Swift described his mother 
as c M ra Abigail Erick of Leicester-shire', 3 her wedding licence 
describes her as *of the city of Dublin spinster*. 4 According to the 
record of her burial, she was seventy years old at her death in 1710, 
and therefore born in i64O. 6 If her father moved to Ireland in 1634, 
and she was born there in 1640, no record of her baptism would of 
course be found in Leicestershire; and Swift's father would have met 
her after he arrived in Ireland, and not before. Since, however, she 
moved to her ancestral county when Swift was still a child, and he re- 

1 Nichols, Leicestershire n. ii. 6sso-i and n. 3. But cf. m. i. 261, n. 10; rv. ii. 982, 
n. 7, and 985. 

* Venn. Autob., f. 6^. * Ball rv. 475. 

6 Parish register of St Martin's, Leicester, 27 Apr, 1710. 



peatedly visited her in Leicester, he naturally thought of her as native 
to Leicestershire. 

James Ericke certainly married Elizabeth Imins, daughter of 
William Imins of Ibstock, Leicestershire, in the church of Thornton 
1 6 October iGzj. 1 His sister Anne married Nicholas Berkett in St 
Martin's, Leicester, 1 6 January 1626 /j. 2 James and Elizabeth Ericke 
witnessed the will of William Imyn of Ibstock on 21 October iGag. 3 

We have thus established a connection of three families. The name 
of each appears unfortunately in many variant spellings: Erick, 
Eyricke, Herrick, etc.; Birkhead, Berket, Burkett, etc.; Imminges, 
Irnin, Imaine, etc. These names corroborate the tie between James 
Ericke and Thomas Errick ; for when the administration of Thomas 
Errick's estate was granted to Jane, his widow (October 1681), the 
sureties named were William Immyn and Josiah Birkhead. 4 

Nichols's statement that Abigail Erick's niece Thomas Errick's 
daughter married John Kendall is corroborated, though not 
proved, by a few facts. The marriage is so described in W. G. D. 
Fletcher's Pedigrees of Leicestershire. 5 Although Fletcher was probably 
following Nichols's Leicestershire, he was a wide searcher; and it is sig- 
nificant that he found no evidence to contradict Nichols. Kendall's 
wife certainly bore the Christian name of her putative mother, Jane. G 
Furthermore, a book has been noticed which was signed 'Jane Ericke, 
her book, 1677' on the flyleaf, with 'John Kendall, his book' written 
below. 7 It also seems significant that Swift's sister was named Jane: 
for though Swift had many female relatives in his father's family, none 
was so called ; but if Nichols is right, this sister was the namesake of 
both an aunt (Thomas Errick's wife) and a cousin (John Kendall's 
wife) in Swift's mother's family. To clinch this argument, Swift, in a 
letter to John Kendall, addresses him as 'cousin', and directs the letter 
to be left 6 at Mr. Birkhead's'. 8 Finally, and corroboratively, when 

1 Thornton parish register. In the marriage bond, i Oct. 1627, she is described 
as 'Elizabeth Imins of Tiltorx' (Leicester City Museum, Archives i D 41/38/111/81). 

2 St Martin's parish register. 

3 Leicester City Museum, Archives 14 D 57/1 12/9, p. 36. 

4 Leicester County Record Office. 

fi Leicester, 1886, p. 160, 'Kendall of Twycross'. 

* She is so named in the records of his children's baptisms in the Thornton parish 
register, and she is so named in his will, 27 June 1717 (Leicester County Record 

7 Transactions of the Leicestershire Archaeological Society, vi, pt. i (1884), lo-zx. 

8 Ball i. 6. 


Elizabeth Jones, whom Swift described as his mother's cousin, was 
married at Thurcaston, the surety was Josiah Birkhead. 1 

James Ericke was made vicar of Thornton in i6sy. 2 The following 
baptisms of the children of 'James Ericke minist" are recorded in the 
parish register : 

25 July 1628 Thomas 

30 March 1630 James 

28 December 1631 William 3 

Although no further record of his family appears here, and there is a 
gap in the original register after 1634, the bishop's transcripts give 
further baptisms: 

8 October 1633 Grace 

1 8 September 1634 Mary 4 

Mary is described as the daughter not of the minister but of 'M* 
James Ericke', and the bishop's transcript for 1634/5 is signed not by 
Ericke, as in earlier years, but by c john Sommerfeild Vicar', his suc- 
cessor. There are bishop's transcripts annually until March 1641/2, 
still preserved, but no mention, in them, of Ericke or his family. 

In the induction mandate for his successor at Thornton, the bene- 
fice is said to be vacant *per cessionem derelictionem sive depriua- 
tionem Jacobi Ericke'. 5 What had happened is made clear elsewhere. 
In January 1633/4 Ericke was accused of having held an unlawful 
conventicle 3 and he confessed that *in the house of Nicholas Birkhead 
who married his sister ... in the presence of some of his neire kinsfolk 
he did then and there repeat the sermon which he had that morninge 
preached at St Maries in Leicester and that they did singe a psalme 
before and one after the said repetition' 6 . He was prosecuted in the 
Court of High Commission, and one of the decisions is recorded under 
the date 8 May 1634: 

This day after severall motions made by the Gounsell of both sydes the 
Court at last resolued to referre the finall hearing and ending of this cause 

1 Ball iv. 55-6. 

* Induction mandate dated as July 1627 (Leicester City Museum, i D 41 / 28/373) 
8 Possibly the alleged uncle of Swift identified as William Herrick of Stam- 
ford, Lincolnshire (JV. & Q. 9 4 Apr. 1885, pp. 264-5; 30 May 1885, P* 435)* 

* Leicester City Museum. 

6 Leicester City Museum, Archives I D 41/28/442. 

6 Helen Stocks and W. H. Stevenson, edd., Records of the Borough of Leicester 
1603-1688 (Cambridge 1923), pp. 271-2. In this text 'Richard* Birkhead is an 
error for *Nicholas* > occasioned by a misreading of the contraction *Nich*. 



to S r John Lambe Knight Gomissary of Leicester yet w th this Reservation 
that the Articles shall still remaine in Court and the Cause not as yet to be 
finally dismissed in reguard the Court resolued to make tryall of him for a 
tyme to see how and in what manner he did demeane and Carry him selfe 
in the execution of his ministry and so for the present he was lycensed to 
depart paying first the costs of the suite. 1 

In the autumn, during a visitation by Archbishop Laud, reports were 
made on the three parishes of which Ericke's benefice was constituted : 
Stanton, Bagworth, and Thornton. At Stanton, 15 September 1634, 
the chancel of the church was found to be 'out of repaire* and *Mr. 
Jacobus Herricke Rector* was ordered to repair it. 2 At Thornton, 
three successive notes were made : the presentment of the church- 
wardens (in English) ; the sentence of the court, added later and dated ; 
and a final note of four words in a slightly different hand : 

Mr. James Errycke vicar there presented for not readinge prayers on 
Wedensdaies and Frydayes and Sonday Eves, 

15. September 1634: contra Mr. Eyricke quern dominus monet ad per- 
legendum preces publicas diebus predictis etc et ad certificandum etc in 
proxima omnium sanctorum 

vertit solum in Hibcrniam 3 

The final four words ( e He has emigrated to Ireland*) are paralleled 
in the report on Bagworth, 22 November 1634: according to this, not 
only is the church 'uncovered and out of repaire', but also 'The 
Minister is absent and gone into Ireland and they have noe prayers 
there nor divine service.' 4 It would thus appear that Ericke remained 
in Thornton until after his daughter Mary was born, September 
1634, but left for Ireland before the date of the Bagworth entry, 
22 November. 

If my account is reliable, later events may be inferred from the 
wills of some relatives of Elizabeth Ericke. To see the pertinence of 
the material, we must recognize a connection among the Tarlton, 
Hooker, and Immings families. This is established by a will dated 
21 January 1687, in which Samuel Tarlton of London names as 

1 P.R.O. SP 16/261, f. 6 V . I have preserved the spelling, etc., of the MS. 

8 A. Percival Moore, *The Metropolitical Visitation of Archdeacon [sic] Laud% 
Associated Architectural Societies' Reports y xxxx, pt. n (1908), 493, I have expanded 
contractions, and spelled the name as in the MS. (Leicester City Museum, Ar- 
chives i D 41/13/61, f. 32 V ). 

8 Leicester City Museum, Archives i 041/13/61, f. 33 v (contractions expanded), 

4 A. P. Moore, p. 493, corrected from MS., as above (f. a8). 



cousins i.e., relatives of some sort William Immyn of Maresfield, 
Leics., and John Hooker of [? Mass worth], 1 Leics. 

These hints are repeated and enlarged by a will dated i January 
1654/5, in which John Hooker of Maresfield, Leics., names as cousins 
William Immings and also Samuel Hooker, student in New England ; 
he names as well other members of the Hooker and Immings families ; 
and he describes Frances Tarlton of the City of London as his sister. 
His executor is William Immings of Maresfield ; the will was proved 
126 November i655. 2 For us, the most important remarks in this will 
are two bequests: a hundred pounds to *Cosin Elizabeth Errick' and 
five pounds to *her daur. Abigail'. From the concatenation of names, 
places, and dates, it seems that this Elizabeth Errick was James 
Ericke's wife, and that Abigail was Swift's mother. 

Furthermore, in a will dated 1 6 March 1 652/3, John Tarlton of the 
parish of St Olave, Southwark, Surrey, and citizen and brewer of 
London, names Frances Tarlton as his wife; he is therefore probably 
the brother-in-law of the John Hooker whose will we have just looked 
at. He also names his sons, among them Samuel, who is probably 
testator of the will of 1687 seen above. One of John Tarlton's bequests 
is three pounds to * Cozen Elizabeth Erricke widow of Leicester'. Since 
this will was proved 23 February 165 3/4, 3 we may assume that James 
Ericke died before 1654. 

Finally, an 'Elizabeth Errick' was buried 4 September 1663 * n St 
John's church, Dublin 4 ; this was probably James Ericke's widow, and 
Swift's grandmother. Her death would presumably have hastened the 
marriage of her daughter, Abigail, the following year; and it may be 
significant that her son, Thomas, took his final orders as a priest in 
December 1 663, three months after her death, although his Cambridge 
B.A. degree is dated 1648 (and his deacon's orders, 7 August i66a). 5 

1 Probably a misreading of 'Maresfield*. 

a Somerset House, Registry of Wills, P.C.G., Aylelt, 403. Ibid., 346. 

4 W. H. Welply, in TLS. 9 17 Jul. 1959. The ^te Mr Welply, with great gene- 
rosity, gave me abstracts of the Tarlton and Hooker wills, the discovery of which 
was due to his patient and exact research. 

6 Venn. 



[St George Ashe's Speech to Lord Clarendon 
25 January I685/6 1 ] 

While the whole nation aemulosly crowds to kis your excellencyes 
hands, and the most differing parties agree in joynt congratulations, 
loud expressions of joy for your wished for arrivall, and the happiness 
they all promise themselves under your auspicious government ; shall 
philosophy alone (how stoicall soever) remain insensible? and the 
[poets] and muses continue silent ? who are proud to call you theire 
own, and glory in hauing form'd your mind to that rare pitch of wis- 
dome and knowledge which is admired by all men. And from whome 
indeed, but persons of your excellencyes illustrius character and in- 
quisitiue genius can philosophy expect its advance and propagation ? 
whose larger souls and more refm'd parts are most fitt to comprehend 
and cultivate the vastest truthes, whose more liberall education, in- 
nate grandeur of mind, and exalted station in ye world, gives 'em a 
more extended prospect, a farther veiw [sic] over the whole intel- 
ligible world ; To such therefore it belongs (whose plenty secures from 
the sordid considerations of gain and profit, and whose usuall 
generosity makes them disdain to bound their thoughts wholly by 
others precepts) to reason and philosophize above the common rate 
of man-kind, at least to protect those that do so, by their authority and 
example. 'Tis true, knowledge was of old for the most part only the 
study of the sullen and the poor, who thought it the gravest peice of 
science to contemn the use of man-kind, and to differ in habit and 
manners from all others; It was heretofore condemn* d to melancholy 
retirements, kept as a minor under the tuition of ambitious and arro- 
gant guardians, buried in cloysters, or the more dark obscurity of 
affected jargon and unintelligable cant; Antiquity too was ador'd 
with such superstitious reverence, as if the beauty of truth, like that of 
a picture cou'd not be known or perceived but at a distance, as if 

1 T.G.D. MS. Molyneux i. 4. 17, no. a. The date is given in the minutes of the 


theire eyes, like ye praepostrus animalls, were behind them, and their 
intellectuall motions retrograde ; No wonder then that knowledge did 
not outgrow the dwarfishness of its pristine stature, and that the in- 
tellectuall world did continue such a microcosm : for while they were 
slaves to the dictiates [sic] of their forefathers, their discoveries, like 
water, cou'd never rise higher then the fountains, from whence they 
were derived. 

But when by the generous care and noble designe of his late majty 
and our present soveraign, and the never to be forgotten zeall of your 
illustrius father, in instituting the Royall Society; Captive truth was 
rescued from its former bondage, & clouded knowledge began to 
shine more bright; when instead of words and empty speculations 
were introduc'd things and experiments, and the beautifull bosome of 
nature was exposed to veiw, where we might enter into its guarden, 
tast of its fruits, satisfy our selves with its plenty, instead of idle talking 
and wandering under its fruitless shadows ; Then philosophy was ad- 
mitted into our palaces and our courts, began to keep the best com- 
pany, to refine its fashion and appearance, and to become the em- 
ployment of the rich and of the great. 

Now, my lord, if your noble fathers patronage was so successfull in 
our neighbouring iland, how happy an omen would it be to us, cou'd 
we obtaine your excellencies tuition for our young orphan society? 
how power full an influence wou'd so great an example have upon all 
the nobility and gentry of our nation ? what a royatt of philosophers 
ought we justly expect from such encouragement? And this we pre- 
sume jxot only to beg, but even challenge, since in it is concluded the 
publick interest and good of ye kingdome, which (we are sure) 'twill 
be your excellencyes care allways to promote and advance, for tho 
fabulous tradition will have this to have been ye iland as well of learn- 
ed men as saints, and the western Athens ; Yet so few tracks and foot 
steps thereof remain' d, that 'tis not long since a mathematicion and a 
conjurer were aequivalent terms, and a telescope and quadrant were 
things as unknown as a frog or a toad. So that considering whence we 
began, our progress for so short a time has not been wholly incon- 
siderable, and some not unsuccessfull essays have been made in many 
usefull parts of knowledge ; Yet we have nothing truly to boast of, but 
the extreme honour that is done us by this noble lord's vouchsafing 
to be our president, concerning whom his presents and modesty for- 
bids me to add more, then that if every member of our company did 



contribute half the zeall and abilities for the promoting naturall and 
mathematicall knowledge, wch our worthy president does, we then 
assure our selves, that we shou'd throughly meritt your excellencyes 
protection. Thus, my lord, tho our designe seems very fair and usefull, 
at least to give offence to no one, yet even in our cradle, like Hercules, 
we have suffered persecution, and been fircely oppos'd by a loud and 
numerous; tho (God be thanked) an impudent sort of adversaries, ye 
railleurs and the witts ; But to gain their good will and do them hon- 
our, I must inform 'em, that the railleurs and philosophers are derived 
from the same originall, the confessed author as well of irony as 
philosophy being Socrates ; Methinks therefore they should be tender, 
where the honour of their comon parent is concern'd, and value their 
own friends and relations, at least above their jest; we are told by 
them that our time is spent in vulgar experiments, in empty useless 
speculations, which (suppose true) is it not necessary that many loads 
of unprofitable earth shou'd be thrown by, before we come at a vain 
of gold, yet certainly the contemplation even of flies and shells and 
the most trifling works of nature (which they so much redicule) is 
more manly then downright idleness and ignorance, and the ex- 
treams of raillery are more offensive then those of stupidity; They 
should reflect that all things are capable of abuse from the same 
topicks by which they may be comended, that (besides the ill man- 
ners in discountenancing such studies as our great master has de- 
clared himself the founder of and promoter) burlasque and [laughter] 
is ye easiest and the slenderest fruit of witt, that it proceeds from ye 
observation of the deformity of things, whereas there is a nobler and 
more masculin pleasure wch is rais'd from beholding their orders and 
beauty; Then they wou'd perceive how great a difference is between 
them and true philosophers, for while nature has only form'd them to 
be pleas'd with its irregularities and monsters, it has given the other 
the delight of knowing and studying its most beautifull works. All use- 
full enterprises among ingenious good natur'd people should find 
assistance wn they are begun, applaus [when] they succeed and even 
pitty and prais when they fail; true railery shoud only intend the 
derision of extravagant, and the disgrace of vile and dishonourable 
things; and this kind of witt ought to have the nature of salt, to which 
tis usually compared, wch preserves and keeps sweet the good and 
sound parts of our bodies, and only fretts and dries up, and destroys 
those humours which putrify and corrupt. 


My Lord, you'l pardon this digression, to which we have had 
sufficient provocation, but we shall have no longer reason to [dread] 
these pleasant gentlemen, if your excellencye will be our patron, wch 
high honour I am [comanded] here most humbly to implore; All our 
inventions and discoveries shall be consecrated to your fame, and 
'twas by such arts that the [antient] heroes and demy- [gods] gain'd 
their temples and their alters; and all studies and endeavours shall 
tend to render (if possible) the great names of Hyde and Clarendon 
more illustrious and renown'd. 



Roll of Students' Marks on the Terminal Examina- 
tions at Trinity College, Dublin, Easter, I685 1 

(Abbreviations: D=dominus [i.e., graduate], G=Greek, Z,=LatiiL, 
Log logic, tfzfl/=mediocriter, neglig =negligenter, Ph physics, 
PM=philosophie, !TA=thema.) 












Ph: male GL male Th: med 
Ph:med G:Lbene Th: negligenter 

Ph: med: GL med Th: vix med: 
Ph: male GL med Th: med 
Ph: med G:L med Th: neglig 
Ph: med G:L bene Th: neglig 
[Ph : male Th : bene 2 ] 

Ph: med GL: med Th: Med 
Ph: vix med: GLvixrned. Th Med 
Swift [i.e., Thomas] Ph: Med: GL: Med. Th: Med 
Swift [i.e., Jonathan] Ph: male G:L bene Th: Neglig: 













Ph: Med: GL Med Th: vix med 

Ph: bene G:L Med Th: med 
Ph: med: G:L med Th: vix med 
Ph : med : GL vix med 
Ph: med GL: med Th: vix med 
Ph: male. GL. med. Th: Med 

Ph: med: G:L Med The: med 

Ph:med. GL med. Th : Negligentfer] 
1 MS. in the library of Trinity College. a Line scratched through. 















[PUiseant 1 ] 

D: Marsh. 

D. Richesy 

D: Hassett 

D: Savage 














WalkCara] 1 








Ph : med. GL vix med Th : med 

Ph: med: GL Med: Th: male neglig 

Ph: male GL Med: Th: mal: neglig 

Ph: med: G:L Med Th: vix med 
Ph:male. G:L. Med Th male 
Ph: Med GL: bene Th: bene vix 
Ph : med : GL med. Th : vix med 
Ph : med. GL med Th : vix med 
Ph : med GL med Th : vix med 
Ph: bene. GL. Med: Th: Med 

Ph: pessime 
Ph: Male 

Ph: male 

Ph: pessime 

Ph: Med 

Ph: pessime 

Ph: Med 

Ph: Med 

Ph: Med 

Ph: bene 

Log bene: G:L Med Th: Med 

Log: mal: Phy: mal: GL med Th: med 

Log: mal. Phy: mal GL Med Th: Med 

1 I>oubtful reading. 





















D : Harrison 

D : Brewster 

D : Ram 

D : Warren 






Log Med Ph: med G:L: Med Th: Med 
Log & Phil: med Gr: L bene Th: bene 
L: & Phil: Med Gr:L: Med Th: Med 
Phil: Med: GL: [Pet] 1 Med Th bene 
Log: Med [Phys] 1 : med: GL: med Th: Med 
Log & Phy: Med: GL vix med: Th: Med 
Log: & Phys: vix med: G:L: Med: Th Med 

Log: & Phys: vix med: G:L: Med Ph Med 

Log: Med Phy: med: G:L: Med Th: Med 
Log: vix med: Phy: med GL: Med Th: Med 
Log & Phy: Med G:L Med: Th: Med 
Phys: med G:L: [obliterated] Th: Med 

Logmal: Phy: mal: G:L: med Th: med 

Log: Med: Th: Med 
Log: Med: Th: Med 

Log: Med: Th: Med 
Log: Med: Th: bene 
Ph: vix med: Th: bene 
Ph: vix med: Th: Med 


R[ichar]dson Ph: med: Th: Med 
R[ichar]dson Ph: Med: Th: Med 

Tarlton Ph: Med: Th: Med 

Maxwell Ph : vix med : Th : bene 

Maxwell Ph: bene: Th: Med 

Bates Ph:med: Th: Med 

Crump Ph: Med: Th: Med 

1 Doubtful reading. 




LeaCthes] 1 





















Ph: mal (Mr. Patrickson contradicts this censure) 

medioc GL & Theme 

Ph: Med: L: med. G bene Th: bene 

L: bene G: Med: Th: Med 
Ph:G:L:Med: Th: bene 

Ph: Med: L: bene. G: Med: Th: Med 
Ph: Med: L: Med: G: bene: Th: Med 

Ph: Med: L: bene. G: Med: Th: Med 
Mediocriter in omnibus 
Mediocriter in omnibus 
Ph: Med: L: Med: G. bene Th: bene 
Ph: Med: L: Med: G: bene Th: Med 
Ph: Med G:Lmed: Th: male 
mediocriter in omnibus 
mediocriter in omnibus 

Log: Med: L:G: bene. Th: Med 
mediocriter in omnibus 

R[ichar]dson Med: in ornnib[us] 

Bredy Med : in omnib[us] 

Ay ton Med: in omnib[us] 

Moor Med in omnib[us] 

Beecher Log: med: L:G: bene. Th: Med 

Touse Log. bene L:G: bene. Th: med 

Prin Log. Med: L:G: bene. Th: med 

Warring Med: in omnib[us] 

Browning Log. op time: L:G:bene. Th: Med 

Vaughen Log: Med: L:G: Med. Th: bene 

Luther Log. bene. L:G. bene. Th: Med 


Quin Ph: Med. G:L. bene Th: bene 

Garner Log: Med: G:L: bene Th: Med 

Williams log: Med: GL: bene. Th: Med 

1 Doubtful reading. 


D: Ware 
3D: Ray 

Oo [\vron3 1 
1 Ooxibtful 


Med in omnibus 

Ph: bene- Th: Med 

Ph: bene G L bene: Th: Med 

Med : in omnibus 

Ph: bene GL bene Th: Med 

bene in omnibus 

Ph: bene GL: bene: Th: 

vix med : in omnibus 

Log; bene GL Med Th: med 

vix med ; in omnibus 


Log: med: GL : ]Med: Th: M!ed 

Log : med GL med : Th : med 
log : med GL ined : Th : med 
log : med : GL. bene Th : med 

Log: Med. GL M!ed Th: Med 

G:L Med Th: med 

Log. med GL Med Th : Med 

Log. xned GL med Th : Med 

Log : med GL med : Th : med 



Disciplinary Fines of Swift and Other Students at 
Trinity College: Further Examples 

I do not repeat here those figures which have been used above, in 
Part I, Chapter Seven. The records are taken from the MSS. Junior 
Buttery Book (14 Nov. 1685 to J 4 Oct. 1687) and Senior Buttery Books 
(14 Nov. 1685 to 1 8 Sept. 1691) of the College 1 ; Swift is entered only 
in the former. Generally, the date of a week's accounts appears on 
each recto ; but it is not clear whether each verso is the same date as the 
preceding or the following recto ; I have assumed the latter, because 
terms seem to begin and to end together on facing pages. There are 
regrettably few periods when circumstances seem sufficiently parallel, 
between Swift and the other students, for him to be significantly 
matched against them. (Non-disciplinary fees or expenses are not 
counted here.) 

Two weeks, 3-16 July 1686: Swift fined is. id.; Thomas Swift, is.; 
Congreve, is. 3d. ; Thomas Wilson, 5d. (not counting 55. cancelled, 
for missing roll-call) . 

Five weeks, 1 7 July to 20 August 1686 : Swift fined 45. 4d. ; Congreve, 
is. 7d. 

Three weeks, 10 September to i October 1686: Swift fined 35. 4d.; 
Stratford, 4$. (not counting 53. cancelled, for missing roll-call) . 

Three weeks, 27 November to 17 December 1686: Swift fined as. 7d. 
(not counting 43. i id. cancelled, for missing ordinary chapel and 
services requiring a surplice) ; Thomas Swift, 55. 8d. (not counting 
35. gd. cancelled, for missing declamations and chapel) ; Stratford, 
is. (not counting 95. 8d. cancelled, for missing ordinary chapel, 
services requiring a surplice, and roll-call) . 

1 1 am deeply indebted to the officers of Trinity College for permitting me to use 
these MSS. and the General Registry. 



Three weeks, 19 February to 1 1 March 1687 : Swift fined 2s. 4d. (noi 
counting ys. zd. cancelled, for missing roll-call, chapel, and services 
requiring a surplice); Stratford, rod. (not counting 135. ad., 



Books Bought by Sir William Temple from 
Ralph Sympson in 1698 

[I have copied the items on Sympson's bill, which is no. 78 in Mr 
James Osborn's collection of Temple family papers. The dates are 
apparently of delivery. After each of Sympson's entries I have given, 
in square brackets, enough information to identify the book and, 
where possible, the number in Donald Wing's Short-Title Catalogue. 
For Brice I have used the catalogue of the Bibliotheque Nationale; 
the note on this book is in Swift's hand and is probably addressed to 
Lady Giffard. The significance of the list is of course the possible use 
of these books by Swift, who certainly read some of them (not neces- 
sarily, to be sure, in these copies) while he was writing the Battle of the 
Books. Cf. the list in the Tale, pp. liii-Uv.] 

Jul. 21, 1698 i Answer to Mullineaux book 0:2:0 

[William Atwood, An Answer to Mr. Molyneux His Case. 
1698. Wing A 4167.] 

i Essay Concerning Critical Learning o : i : o 

[Thomas Rymer, An Essay, concerning Critical and Curious 
Learning, 1698. Wing R 2425.] 

i Answer to it 0:0:4 

[An Answer to a Late Pamphlet, Called An Essay Concerning 
Critical . . . Learning. 1698. Wing A 3306.] 

Aug. 23, 1698 i Embassy from. Moscow to China 0:2:0 

[Adam Brand, Journal of the Embassy from Muscovy . . . 
into China. 1698. Wing B 4246,] 

i Hist: of Muscovy in 2 Vol 9 o: 6: o 

[Jodocus Crull, The Antient and Present State of Muscovy, 
2 vols. 1698. Wing C 7424.] 



i Boyles Answer to Bentley 3 d Edit: o: 4: o 

[Charles Boyle, Earl of Orrery, Dr. Bentlefs Disserta- 
tions, 3rd ed. 1699. Wing O 471.] 

i Wafers Discription of y e Isthmus 0:3:6 

[Lionel Wafer, A New Voyage and Description of the Isth- 
mus of America. 1699. Wing W 193.] 

i Vindication of an Essay o: o: 6 

[Thomas Rymer, A Vindication of an Essay concerning 
Critical . . . Learning. 1 698. Wing R 2434.] 

i Account of Pennsilvania o: i: o 

[Gabriel Thomas, An Historical and Geographical Account 
of. . . Pensilvania. 1698. Wing T 964.] 

i Life of Milton 0:2:0 

[John Toland, The Life of John Milton. 1699* Wing T 

i History of Standing Armies 0:0:6 

[John Trenchard and W. Moyle, A Short History of 
Standing Armies. 1698. Wing T 2 1 15.] 

for Binding Divine Plate Gal : pi? 0:3:0 

at Tunbridge o : i : 

Tunbridge. 1698. Wing A 739 or 739A.] 

i View of y e Dissertations on phalris o: i: 

[John Milner, A View of the Dissertation upon the Epistles 
ofPhalaris. 1698. Wing M 2082.] 

x Fory r i x Discription of Paris o: 2: 


[Germain Brice, Description nouvelle de la ville de Paris. 
2 vols. Paris 1698. Bibl. Nat. cat., vol. xix, p. 603.] 

i Character of a Trimer o: o: 

[George Savile, Marquess of Halifax, The Character of a i : 1 1 : i 
Trimmer, *3rd* ed. 1697. Wing H 300.] 



Abbreviated References and a Note on the 

In quotations I have followed the spelling and punctuation of the 
originals, using ellipses to indicate any deletions made within the 
passage quoted. But I have not preserved the old usage in capitals, 
italics, superior letters, or the long s ; I have not indicated omissions at 
the opening or the close of a quotation ; and I have generally spelled 
out ampersands, c ym% and a few other contractions. In some pas- 
sages from unpublished manuscripts I have followed the originals 
more closely; but several quotations, in which the substance alone 
was important, I have condensed rather freely, calling attention to 
such treatment in my notes. 

In the following list of abbreviated references, the place of publica- 
tion is London if no place is named. 


Jonathan Swift's unfinished autobiographical essay, headed by 
him, 'Family of Swift 9 . MS. in the library of Trinity College, 


The British Museum. 


Richard Bagwell, Ireland under the Stuarts. 3 vols. 1909-16. 

The Correspondence of Jonathan Swift, ed. F. Ellington Ball. 6 vols. 



John Barrett, An Essay on the Earlier Part of the Life of Swift. 1808. 
Black Book 

The Black Book. MS. in the library of the King's Inns, Dublin. 




The Bodleian Library, Oxford. 

Robert Bolton, A Translation of the Charter and Statutes of Trinity- 
College. Dublin 1749. 
Boyle's Examination 

Dr. Bentleyfs Dissertations on the Epistles of Phalaris, and the Fables 

of JEsop, Examined by the Honourable Charles Boyle, Esq;, 2nd ed. 


Gilbert Burnet, Bishop Burners History of His Own Time, ed. M. J. 

Routh. 6 vols. Oxford 1824. 

George Dames Burtschaell and T.U. Sadleir, comps., Alumni 

Dublinenses, new ed. Dublin 1935. 

Thomas Carte, The Life of James, Duke of Ormond, new ed. 6 vols. 

Oxford 1851. 

G. N. Clark, The Later Stuarts 1660-1714. Oxford 1934. 

Rosalie L. Colie, Light and Enlightenment. Cambridge 1957. 

Kenneth Hugh Connell, The Population of Ireland 1750-1845. 

Oxford 1950. 

Henry Cotton, Fasti Ecclesite Hibernias* 5 vols. Dublin 1847-60. 

Thomas Peregrine Courtenay, Memoirs of the Life, Works, and 

Correspondence of Sir William Temple. 2 vols. 1836. 

Henry Craik, The Life of Jonathan Swift, 2nd ed. 2 vols. 1894. 

The Dublin University Magazine, vol. xvin, 1841. 

The Prose Works of Jonathan Swift, ed. Herbert Davis. Oxford 

1939- - 


Davis, Drapier 

Jonathan Swift, The Drapier 1 's Letters, ed. Herbert Davis. Oxford 1935. 

Robert Dunlop, Ireland from the Earliest Times to the Present Day. 

Oxford 1923. 
Fletcher 3 Gray's Inn 

Reginald James Fletcher, ed., The Pension Book of Grafs Inn. 2 vols, 


The Letters of Jonathan Swift to Charles Ford, ed. D. Nichol Smith. 

Oxford 1935. 

John Forster, The Life of Jonathan Swift, Volume i [no more pub- 
lished], 1875. 

Alumni Oxonienses, ed. Joseph Foster. 8 vols. 188792, 

James Anthony Froude, The English in Ireland in the Eighteenth 

Century. 3 vols. 1872-4. 

John Thomas Gilbert, A History of the City of Dublin. 3 vols, Dublin 


Historical Manuscripts Commission, Reports. 

John C. Hodges, William Congreve the Man. New York 1941. 

The Journal of English and Germanic Philology. 

The Journals of the House of Commons. 

Journal of the History of Ideas. 

The Journals of the House of Lords. 

Jonathan Swift, Journal to Stella, ed. Sir Harold Williams. 2 vols. 

Oxford 1948. 




Albert Jouvin, *M. Jouvin de Rocheford's Description', in G. 

Litton Falkiner, ed., Illustrations of Irish History and Topography. 


Hugh F. Kearney, Sir afford in Ireland. Manchester 1959. 

John Keble, The Life of Thomas Wilson. 2 vols. 1863. 

Louis A, Landa, Swift and the Church of Ireland. Oxford 1954. 
Locke, Works 

The Works of John Locke, 5th ed. 3 vols. 1 75 1 . 

Julia S. Longe, Martha, Lady Giffard, Her Life and Correspondence, 


A. A. Luce, The Life of George Berkeley. 1949. 

Narcissus Luttrell, A Brief Historical Relation of State Affairs. 6 vols. 

Oxford 1857. 

John Lyon, MS. notes in a copy of J. Hawkesworth, The Life of the 

Revd. Jonathan Swift, 1755, in the Victoria and Albert Museum, 

Forster Collection, no. 579 [48. D. 39]. 

Modern Language Notes. 

William Monck Mason, The History and Antiquities of the Collegiate 

and Cathedral Church of St. Patrick. Dublin 1 820. 

Constantia E. Maxwell, A History of Trinity College, Dublin. Dublin 


Samuel McSkimin, The History and Antiquities of. . . Camckfsrgus, 

3rd ed. Belfast 1829. 

N. s?a- 

Notes and Queries. 



The Works of the Rev. Jonathan Swift, ed. John Nichols. 19 vols. 


David Ogg, England in the Reign of Charles II, 2nd ed. 2 vols. Oxford 

Ogg n 

David Ogg, England in the Reigns of James II and William III. Oxford 



Prerogative Court of Canterbury. 


Publications of the Modern Language Association. 


Public Record Office. 


Sir William Petty, Economic Writings, ed. C. H. Hull. 2 vols. 
Cambridge 1899. 


Walter Alison Phillips, ed., A History of the Church of Ireland. 3 vols. 


Memoirs of Mrs Laetitia Pilkington, ed. Iris Barry. 1 928. 


The Poems of Jonathan Swift, ed. Sir Harold Williams, and ed. 3 vols. 
Oxford 1958. 


Emile Pons, Swift, Les Annies de jeunesse et le "Conte du tonneau". 

Strasbourg 1925. * 


Calendar of State Papers, Domestic. 

Calendar of State Papers, Ireland. 
T. Scott 

The Prose Works of Jonathan Swift, ed. Temple Scott [i.e., J. H. 

Isaacs]. 12 vols. 1897-1908. 


W. Scott 

The Works of Jonathan Swift, ed. Sir Walter Scott. 19 vols. Edin- 
burgh 1814. 


The Correspondence of Alexander Pope, ed. George Sherburn. 5 vols. 
Oxford 1956. 

John Gerald Simms, The Williamite Confiscation in Ireland. 1956. 


Critical Essays of the Seventeenth Century, ed. J. E. Spingarn. 3 vols. 
Oxford 1908. 


Miriam K. Starkman, Swift's Satire on Learning in A Tale of a Tub. 
Princeton 1950. 


John William Stubbs, The History of the University of Dublin. Dublin 

D. Swift 

Deane Swift, An Essay upon the Life, Writings, and Character, of Dr. 
Jonathan Swift. 1 755. 

Sykes, William Wake 

Norman Sykes, William Wake. 2 vols. Cambridge 1957. 


Trinity College, Dublin. 


Jonathan Swift, A Tale of a Tub, ed. A. C. Guthkelch and D. Nichol 
Smith, 2nd ed. Oxford 1958. 

Temple, Early Essays 

The Early Essays and Romances of Sir William Temple, ed. G. C. Moore 
Smith. Oxford 1930. 

, Introduction to the History of England. 

An Introduction to the History of England. 1695. 

, Letters I-H 

Letters Written by Sir W. Temple, ed. Jonathan Swift, a vols. 1 700, 
, Letters in 

Letters to the King . . . by Sir W. Temple, ed. Jonathan Swift. 1 703. 


Temple, Memoirs n 

Memoirs of What Past in Christendom. 1692. 
, Memoirs m 

Memoirs. Part HI ... by Sir William Temple, ed. Jonathan Swift. 

, Miscellanea i 

Miscellanea. 1680. 

, Miscellanea n 

Miscellanea. The Second Part, and ed. 1690. 
-, Miscellanea in 

Miscellanea. The Third Part . . . by the Late Sir William Temple, ed. 

Jonathan Swift. 1 70 1 . 
, Observations upon the U.P. 

Observations upon the United Provinces of the Netherlands, 6th ed. 1693. 
, Of Poetry ; 

Essay IV. Of Poetry, in Miscellanea n. 279-341. 
-, Select Letters 

Select Letters to the Prince of Orange. . . . To Which Is Added An Essay 

upon . . . Ireland. 1701. 

John Venn and John Archibald Venn, comps., Alumni Canta- 

brigienses. xovols. Cambridge 19122-54. 

The Whole Works of Sir James Ware, ed. Walter Harris. 3 vote. 

Dublin 1739-46. 
K. Williams 

Kathleen Williams, Jonathan Swift and the Age of Compromise. 1958. 

Homer Edwards Woodbridge, Sir William Temple. New York 1940. 
Wotton, Refections 

William Wotton, Reflections upon Ancient and Modern Learning, 2nd 

ed. 1697.