Handbook to the textual criticism of the New Testament

ἊΣ

st

HAROLD B. LEE LISRARY a ig?

Βειαη YOUNG ὦ MVERSITY
PROVO, UTAH
Ἵ
en τὰ he % ἐν lea avy

Pete's Ge
BA asl

Digitized by the Internet Archive
in 2011 with funding from
Brigham Young University

http://www.archive.org/details/handbooktotextua00keny

i

: iy
‘ ri th
Bie ean

Ae Rn eae a aut n
δὶ ΟΝ “i ΝΠ! ᾿ ᾿ πὴ a
κ' ‘4 ih Ae ϊ VY a va | Pitre ahs
aN ΜΑΙ ἈΝΑ A oh ἢ Rae ah

‘i
‘

’

fi
Ἧ

co

oy Τὴν my
a oe
Nannies

“
{

Pith

Pan i Wij

% ae ΠῚ Bie ᾿

Lr
4

h ἫΝ ᾿
ΓΝ:
ΤῊ,

HANDBOOK TO THE TEXTUAL CRITICISM

OF

THE ΝΒ fhe PAMENT

MACMILLAN AND CO., LimitrEep
LONDON - BOMBAY - CALCUTTA
MELBOURNE

THE MACMILLAN COMPANY

NEW YORK - BOSTON - CHICAGO
DALLAS +- SAN FRANCISCO

THE MACMILLAN CO. OF CANADA, Lrtp.
TORONTO

pes
i ~~

*)

» \

as
“) wm

DK nt .
HANDBOOK
TEXTUAL CRITICISM

NEW TESTAMENT

BY

sik FREDERIC ὦ. KENYON, K.C.B., F.B.A.

DIRECTOR AND PRINCIPAL LIBRARIAN, BRITISH MUSEUM

SECOND EDITION
WITH SIXTEEN FACSIMILES
MACMILLAN AND CO., LIMITED

ST. MARTIN’S STREET, LONDON
1912

COPYRIGHT

First Edition, ὅσο, 1901
Second Edition, Extra Crown 8vo, 1912

HAROLD 8. LEE LISRARY
BRIGHAM ¥ hers UNIVERSITY
PROVO, UTAH

Georgia Wiyding
EPISCOPO SOUTHWELLENSI
COLLEGII P.V.M. WINTON PROPE WINTON
OLIM INFORMATORI
HUNC LIBRUM GRATO ANIMO

DEDICAT DISCIPULUS

PREPACE ΤΟ The SECOND BDELION

A NEW edition of this handbook being required, I
have endeavoured to bring the information contained
in it up to date. The principal alteration made in it
is in regard to the numeration of the MSS. The new
system proposed by von Soden is described, and
likewise the modification of the notation hitherto in
use which Gregory, after consultation with a large
number of Biblical scholars, has elaborated ; and the
latter has been adopted in this edition. A new
section is also devoted to von Soden’s theory of the
textual history of the New Testament. In other
respects, though there are many alterations in detail,
the book remains substantially as before; but the
opportunity has been taken to issue it at a lower
price, which, it is hoped, may make it more accessible
to the students to whom it is most likely to be useful.

I have to thank many friends for suggestions and
corrections, notably Prof. C. R. Gregory and Prof. J.

Hope Moulton.
BG. KB.

August 20, 1912.

vii

Wy sb At |
ye Van i it sain i)
ΠῚ ἡ" i aie i ite ui . ll

Pia! ν
ἊΨ ἡ. ae eae f age Ἦν τ

i Ἵ; ᾿ | wey i 4 ᾿ ie fo rt 4.5: ὺ
pla ile woe ibys AER

ΓΝ Be, γι
᾿ ὯΝ ite ᾿

Shay Lyra ὦ re dks alt sty ¥ 14] ip oy nm an

bee " ees iN Ἢ alee ria Hin ce dati
τόν! ἊΝ, aul ν nit 1 pert

a

db ine iB ‘ua coy + line ft " i ἱ
ἡ ὦ heen is ae te

ἐν

ΠῚ at, pyri γὴν
er Ny ne

ht

PREFACE TO) THE, FIRST) EDITION

THE object of this volume is to provide a serviceable
handbook to the textual criticism of the New Testament
for the use of students who are comparatively new to
the subject. It lays no claim to rival the standard
works of Gregory and Scrivener as a storehouse of
statistics and bibliographical information, though in
certain details it has been possible to supplement them
and bring them up to date; on the other hand, the
discussion of textual theories is somewhat fuller than
the plans of those works admitted of. It seemed
advisable to indicate to the student the present state
of the principal controversies with regard to textual
theory ; and I have tried to represent all the more
important views fully and fairly, whether I agree with
them or not.

The lists of authorities prefixed to the several
chapters are not intended to be exhaustive, but to
indicate either those works which I have chiefly con-
sulted, or those which will be most useful to the
student (especially the English student), who wishes
to push his inquiries further. Such a work as this is

in great measure dependent on the labours of others,
ix

x CRITICISM OF NEW TESTAMENT

to whom my indebtedness is, I hope, fully acknow-
ledged throughout. I have also to thank Cavaliere
F. Carta, Chief Librarian of the National Library at
Turin, for his courtesy in enabling me to obtain a
photograph of a page of the Codex Bobiensis ;
Mr. W. E. Crum for much information with regard
to) tie ‘Coptic, Versions ;) and especially: Mr. CH.
Turner, Fellow of Magdalen College, Oxford, and the
Rev. H. E. Salter, vicar of Shirburn, for their kindness
in reading the proofs and making many valuable
suggestions and corrections.
Ἐν ya

October 21, 1901.

CHAP.
. THE FUNCTION OF TEXTUAL CRITICISM .

CONTENTS

. THE AUTOGRAPHS OF THE NEW TESTAMENT

Catalogue of Papyri

. THE UNCIAL MANUSCRIPTS

. THE MINUSCULE MANUSCRIPTS

. THE ANCIENT VERSIONS ,

1. The Syriac Versions
(a) The Diatessaron of Tatian
(4) The Old Syriac
(c) The Peshitto

(4) The Philoxenian and ree Syriac .

(6) The Palestinian Syriac
. The Armenian Version
. The Georgian Version
. The Persian Versions

uw & W DN

The Coptic Versions ᾿
(a) The Bohairic Version .
(ὁ) The Sahidic Version ἱ
(c) The Middle Egyptian Versions

xi

145
147

181

164
167
171
174
175
176
181
185
192

xii CRITICISM OF NEW TESTAMENT

CHAP.

6. The Ethiopic Version
7. The Arabic Versions
8. The Latin Versions
(a) The Old Latin
(ὁ) The Vulgate .
g. The Gothic Version

VI. PATRISTIC QUOTATIONS .
VII. TEXTUAL CRITICISM IN THE PAST

VIII. THE TEXTUAL PROBLEM

. The a-text .

2) The βασι:

3. They-text’..

4. The 6-text . ;
5. von Soden’s Textual Theory

--

INDEX 11. SUBJECTS AND PERSONS .

INDEX II. PASSAGES OF WHICH THE VARIOUS READ-
INGS ARE NOTICED

PLATES

{Most of the facsimiles are reduced in scale, in order to show complete pages ;
the scale of reduction is stated in each case on the plate.]

FACE PAGE

I. Hyperides pro Euxenippo (Papyrus Roll of the

First Century, literary hand) . . 28
II. Aristotle’s ᾿Αθηναίων Πολιτεία (Papyrus Roll of
the First Century, non-literary hand) . : 30
III. Codex Sinaiticus, 4th century : : 5 60
; IV. Codex Alexandrinus, 5th century . ! : 72
V. Codex Vaticanus, 4th century 3 : ; 78
VI. Codex Bezae, 6th century . . : ; 89
VII. Codex Regius, 8th century : : 108
VIII. Codex Nitriensis, 6th century : : : {153
IX. Codex 81, A.D. 1044 : : : : 134
X. Codex 700, 12th century . ἢ ἱ 139
XI. The Curetonian Syriac MS., 5th century . sath) SESE
XII. The Peshitto Syriac (Brit. Mus. Add. MS. 14459),
A.D. 530-540. ‘ : ‘ : 161
XIII. The Bohairic Version (Curzon Catena), A.D. 889 . 183
XIV. The Sahidic Version (Brit. Mus. MS. Or. 3518),
5th century (?) 5 187
XV. The Codex Bobiensis (ἢ of the Old Diu a
century (?) : : 203

XVI. The Lindisfarne Gospels (Y of the ne circ.

A.D. 700 : 3 Se F |
xiii

CHAPTER
THE FUNCTION OF TEXTUAL CRITICISM

THE province of Textual Criticism is the ascertainment
of the true form of a literary work, as originally
composed and written down by its author. The
science owes its existence to the conditions under
which, until comparatively modern days, literary works
have been preserved. If the author’s autograph of
every book were still in existence, there could be no
dispute as to what he had written; or if printing had
been practised from the earliest days of literary com-
position, we could be sure that every book had been
handed down to us in practically unaltered form. For
authors of the last four centuries, with few exceptions,
we are in the happy condition of being certain that we
possess their works, to all intents and purposes, pre-
cisely as they wrote them. In several instances the
author’s autograph is still extant; in the rest we have
early printed editions, issued under the author’s eye.
But when once we go back into the ages before the
invention of printing, the conditions are wholly
different. Only in the rarest possible cases (the
great English chronicler of the thirteenth century,
Matthew Paris, is perhaps an example) do we possess
the authors own copy of his work; in all other
instances we have only copies made by hand at varying
Β

2 CRITICISM ‘OF “NEW (TEST ANDEN GD: ἢ:

distances of time after the composition of the book in
question. It is to this copying by hand that the
problems of textual criticism are due. Unfortunately
for our knowledge of ancient literature, the frailty of
the human hand and eye and mind is such that no
copy, except of short passages, can be trusted to be
wholly accurate; and since different copyists will
make different mistakes, it results that no two copies
of an ancient book are quite thesame. This would be
immaterial, so long as the original autograph was in
existence ; but when once that has disappeared, the
student who would know exactly what an author
wrote has to discover it by an examination of later
copies, of which the only fact certain a przorz is that all
will be different and all will be incorrect.

The function of textual criticism, then, is to recover
the true form of an author’s text from the various
divergent copies that may be in existence. The
problems presented to it are of all kinds of complexity.
If evidence is forthcoming from a period shortly after
the writer’s date, there will have been little time for the
text to have been corrupted, and common sense should
be able to detect most of the errors that have crept in.
If the interval between the composition of the work
and the earliest extant specimens be longer, much will
depend on the amount of evidence available; for
among many copies there is more chance that the
truth will have survived in some, especially if the
extant copies have no common ancestor much later
than the author's autograph. The line of textual
tradition for any given literary work is like a genea-
logical tree, starting from a single point and spreading
out as it descends to the living members of the family.
If the distance of time be great, but the extant copies
many, then the textual problem will be one of con-

oo) PERG RION OF THSTUARSCRIPICISM: ἃ

siderable difficulty, and requiring nice taste and
discernment, but it will be hopeful, because the materials
are plentiful; whereas if the extant copies be few,
there is a great likelihood that the truth will, in some
places, have been wholly lost, and is only to be
recovered by guessing—a process precarious in the
extreme, and seldom allowing any one but the guesser
to feel confidence in the truth of its results.

Now the textual criticism of the New Testament,
as it is the most important branch of the science, so
also is it the most complicated. It is the most
important branch, because it has to do with a book,
the importance of which is quite incommensurable
with that of any other book in the history of the
world; and it is the most complicated, because the
extant materials are incomparably more plentiful in
number, and more varied in kind, than in any other
instance. The difference in this respect between it
and any other ancient book can be made plain by a
few examples. The plays of Aeschylus are preserved
in perhaps fifty’ manuscripts, none of which is com-
plete. Sophocles is represented by about a hundred
manuscripts, of which only seven have any appreciable
independent value. The Greek Anthology has survived
in one solitary copy. The same is the case with a
considerable part of Tacitus’ Annals. Of the poems
of Catullus there are only three independent manu-
scripts, all of which were derived from an archetype
which was itself written no earlier than the beginning
of the fourteenth century. Some of the classical
authors, such as Euripides, Cicero, Ovid, and especially
Virgil, are, no doubt, in a far more favourable position

1 Forty appears to be the number of those that have been collated ; but
there are probably several that have not been collated. Very few, however,
contain more than the three plays which were habitually read by the Byzantine
public,

4 CRITICISM OF NEW TESTAMENT - cu.

than those who have just been named. In their cases
the extant copies of their works, or of portions of them,
may be numbered by hundreds. Yet even these do
not approach the number of witnesses for the text
of the New Testament. The number of manuscripts
of it, or of parts of it, in the original Greek, is over
four thousand; and to these have to be added a
yet greater number of witnesses of a kind to which
the classical authors offer no parallel. It is seldom
that ancient translations of the classical authors into
other languages exist, and still more seldom that they
are of any value for textual purposes ; but in the case
of the New Testament translations are both numerous
and important. It is estimated that there are at least
eight thousand copies extant of the Latin Vulgate
translation alone ; and athousand would be a moderate
estimate for the extant manuscripts of the other early
versions, in Syriac, Coptic, Armenian, Ethiopic, Gothic,
and the rest. It is therefore probably within the mark
to say that there are now in existence twelve thousand
manuscript copies of the New Testament, of which
no two are precisely alike.

The contrast in this respect between the New
Testament and classical authors may be regarded from
two points of view. On the one hand, this enormous
mass of witnesses gives good ground for supposing
that the true text cannot be wholly lost; on the other
hand, the task of selecting the true text out of all these
many and multifarious authorities is one of extreme
difficulty. Merely to examine and record the available
evidence is an enormous labour ; to estimate its value,
to distinguish between manuscript and manuscript, and
between version and version, is the hardest problem
that has ever been set to textual criticism.

In another respect, however, besides number, the

mi PORE PON OF TEXTUAL CRIBICISM 8

manuscripts of the New Testament differ from those
of the classical authors, and this time the difference
is clear gain. In no other case is the interval of
time between the composition of the book and the date
of the earliest extant manuscripts so short as in that
of the New Testament. The books of the New Testa-
ment were written in the latter part of the first
century ; the earliest extant manuscripts (trifling scraps
excepted) are of the fourth century—say, from 250
to 300 years later. This may sound a considerable
interval, but it is nothing to that which parts most of
the great classical authors from their earliest manu-
scripts. We believe that we have in all essentials an
accurate text of the seven extant plays of Sophocles ;
yet the earliest substantial manuscript upon which it
is based was written more than 1400 years after the
poet’s death. Aeschylus, Aristophanes, and Thucy-
dides are in the same state ; while with Euripides the
interval is increased to 1600 years. For Plato it may
be put at 1300 years, for Demosthenes as low as 1200.
The great Latin authors are somewhat better off.
Horace is represented by several manuscripts written
within 900 years of his death. There is an excellent
copy of Terence after an interval of about 700 years,
and portions of Livy only about 500 years after his
date. For Lucretius, however, we have an interval
of nearly 1000 years, for Catullus about 1600. Only
Virgil approaches the New Testament in earliness of
attestation. He died eight years before the Christian
era; and there is at least one nearly complete manu-
script which is attributed to the fourth century, besides
several small fragments, and two more of the fifth
century. Yet even so his text is not in so favourable
a position as that of the New Testament by nearly
100 years.

6 CRITICISM OF NEW TESTAMENG (ce

The task of textual criticism, then, in relation to the
New Testament, is to try to extract the actual words
written by the apostles and evangelists from the great
mass of divergent manuscripts in which their works
have been preserved. It is a task at once hopeful and
hopeless. Hopeful, because in so great a crowd of
manuscripts, reaching back to so early a date as many
of them do, the truth must, it would seem, somewhere
be on record ; hopeless, because the discernment of it
requires a superhuman degree of knowledge and judg-
ment, and because means do not exist for demonstrating
it corclusively. The actual extent 'to which ithe text
of the New Testament is open to doubt cannot be
precisely stated, but the estimate of Dr. Hort, whose
lifetime was devoted to this subject, is commonly
accepted as an approximate guide. He says’: “ The
proportion of words virtually accepted on all hands as
raised above doubt is very great, not less, on a rough
computation, than seven-eighths of the whole. The
remaining eighth, therefore, formed in great part by
changes of order and other comparative trivialities,
constitutes the whole area of criticism. If the principles
followed in this edition are sound, this area may be
very greatly reduced. Recognising to the full the duty of
abstinence from peremptory decision in cases where the
evidence leaves the judgment in suspense between two
or more readings, we find that, setting aside differences
of orthography, the words in our opinion still subject
to doubt only make up about one-sixtieth of the
whole New Testament. In this second estimate the
proportion of comparatively trivial variations is beyond
measure larger than in the former ; so that the amount
of what can in any sense be called substantial variation
is but a small fraction of the whole residuary variation,

1 Introduction to the New Testament in the Original Greek, p. 2 (1882).

I PUNCTION OF THATUAL CRITICISM | 7

and can hardly form more than a thousandth part of
the entire text.” It is further to be remembered that,
although some doubt attaches to the record of certain
incidents and sayings of great interest and value, yet no
doctrine of Christianity rests solely upon a disputed text.
The Christian student can approach the subject without
misgiving, and may follow whithersoever honest inquiry
seems to lead him, without thought of doctrinal con-
sequences. His researches should unquestionably be
conducted in a reverent spirit, but he may avail himself,
without hesitation or mistrust, of all the resources of
secular science.

The methods of textual criticism may be broadly
described as two in number—the comparison of docu-
mentary evidence, and conjecture. The two methods
are mutually complementary. Where documentary
evidence is plentiful, conjecture will be scarce; but
where the former is wanting, the latter will have to try
to take its place to the best of its ability. In the case
of the New Testament the documentary evidence is so
full that conjecture is almost excluded, and it is with
the principles of the interpretation of documentary
evidence that we are most concerned here. Some
statement of these is necessary, as an introduction to a
summary of the evidence itself.

The task of the textual critic is, in brief, to counter-
act the errors of the copyist; and these errors are
many,—some capable of being classified under heads,
while some resist classification. In the first place the
critic has to correct simple slips of the pen, obvious
blunders which have no meaning, and which occasion
no more difficulty than similar mistakes in the letters
of a contemporary correspondent. If the scribe of the
Codex Sinaiticus writes ποίσαι for ποιῆσαι, or ex Tov
καάλουντας for ἐκ τοῦ καλοῦντος, there is no difficulty in

ὃ CRITICISM ΟΕΥΝΕΝ ΤΕΒΨΑΕ ΕΣ

either seeing or correcting the error. A somewhat less
elementary form of blunder arises when the scribe, in
place of the word which he should write, writes one
which resembles it either in sound or in appearance.
Thus in Sophocles’ Ajax 61 some manuscripts have
φόνου, others πόνου, and the context is such as to make
the decision between them not absolutely certain ; but
whichever is wrong, the error was no doubt due to the
similarity of sound. On the other hand in Bacchylides
v. 23 the scribe of the only extant manuscript has
written φοιίβωιν for φόβῳ, an error of eye, not of ear ;
and here the metre and the sense alike make the error
obvious and easy to correct. Another common form
of error is due to the fact that in ancient manuscripts
accents and breathings were rare, and separation of
words almost unknown;? which led to trouble when
the time came for these aids to intelligence to be intro-
duced. Thus in Sophocles’ Ajax 1056 the earliest
MSS. -had, at the end of the line, the letters EAOIAOPL
Now εἰ and «¢ are constantly interchanged in manu-
scripts, and hence EAOIAOPEI was probably written as
often as EAOJAOPI. The result is that, in the margin
of the best extant MS. of Sophocles, the reading
ἐλοιδόρει is given, in place of the correct ὅλοι δορί.
Another form of error, very common in all manu-
scripts, is that of omission. This may be due to mere
unaccountable accident, and then the lost word or
words can only be recovered either by comparison
with other manuscripts or by sheer guessing. Oftener,
however, it arises from the similarity of adjoining words,
which led the scribe’s eye to slip from one to the other,
and so omit the intervening words. For instance, in
1 Early vellum MSS., from the fourth to the ninth centuries, are almost
wholly without accents, breathings, and divisions. Papyrus MSS., which are

still earlier, not infrequently have occasional accents, and, in rare instances,
the separation of words is indicated by a dot in cases of doubt.

I PUNGHONGC OP PERT UAL CRITICISM 19

John xvii. 15 the correct text runs οὐκ ἐρωτῶ iva ἄρῃς
αὐτοὺς ἐκ τοῦ κόσμου ἀλλ᾽ ἵνα τηρήσῃς αὐτοὺς ἐκ TOD
πονηροῦ, but the scribe of the Codex Vaticanus let his
eye slip from the first é« τοῦ to the second, and so
gives the passage as οὐκ ἐρωτῶ ἵνα ἄρῃς αὐτοὺς ἐκ τοῦ
πονηροῦ. Similarly in John iii. 20, 21, where the true
text runs πᾶς yap ὁ φαῦλα πράσσων μισεῖ TO φῶς Kal
οὐκ ἔρχεται πρὸς τὸ φῶς ἵνα μὴ ἐλεγχθῇ τὰ ἔργα αὐτοῦ,
ὁ δὲ ποιῶν ἀλήθειαν ἔρχεται πρὸς τὸ φῶς ἵνα φανερωθῇ
Ta ἔργα αὐτοῦ ὅτι ἐν θεῷ ἐστὶν εἰργασμένα, the scribe of
the Codex Sinaiticus has made two mistakes from this
same cause (technically known as homoioteleuton),
omitting καὶ οὐκ ἔρχεται πρὸς TO φῶς and ὁ δὲ ποιῶν

. avtov, the former owing to the double occurrence
of τὸ φῶς, the latter owing to the double occurrence of
τὰ ἔργα αὐτοῦ. Often the omissions are smaller than
these, and cause less trouble, as when a scribe writes
KATHN for KATATHN (κατὰ τὴν) or ἔπεμψε for ἐπέ-
πεμψε. But in one form or another the error is a very
common one, and has to be borne in mind constantly
in the criticism of manuscripts.

Various other classes of error exist and may be
briefly mentioned. One that is frequently invoked in
the criticism of classical authors is the intrusion into the
text of words which were originally explanatory notes
written in the margin. Sometimes the paraphrase has
extruded the original phrase, sometimes the true and
the false remain side by side. This, however, is a form
of corruption which occurs less often in the Biblical
writings than in profane authors, and even in the latter
the instances where it is proved to have taken place
are much fewer than those in which it is assumed by
some) critics. i Then there isthe .class\ of | deliberate
alterations, such as are known to have been made in
the texts of the Greek dramatists by the actors, and

το (CRITICISM ‘OF NEW ‘TEST ΜΕ Vice:

such as are suspected to have been made by the
scholars of Alexandria in the texts of the Attic authors
generally, in deference to certain supposed laws of style
and euphony. In regard to the Old Testament we
know that the text of the Septuagint was extensively
altered by Origen in order to bring it into closer con-
formity with the Hebrew text current in his day; in
the case of the New Testament there is good reason to
suppose that many of the divergences which now exist
were due to deliberate editing, intended, no doubt, to
secure the best possible text according to the materials
available for the scribe or his director, but often result-
ing in departures from the true and original reading.
In the case of religious books there is also always the
danger of deliberate alteration for doctrinal reasons,
and we know that various heretical sects had their own
recensions of certain books of the Bible; but this
danger is discounted by the enormous mass and variety
of evidence in existence for the New Testament. There
is no possibility that all the sources should be tainted ;
one or other of them would be sure to have escaped,
and when once the alternatives are presented to the
critic, there is generally little difficulty in detecting a
doctrinal perversion.

A special form of deliberate alteration, for which
the student: of the New Testament has ‘to’ be “on
his guard, occurs in the case of the Synoptic Gospels.
When the same event is recorded by two or more
writers, there was a natural temptation to scribes to
amplify one by the insertion of details mentioned in
another, or to use the phrases of the more familiar
version in transcribing that which was less familiar.
This is a form of corruption which is constantly found
in the later MSS. of the Gospels; and any one who
will take the trouble to compare the Authorised and

Mi PONG SION OF TExXTUAL CRITICISM 11

Revised Versions of the English Bible will find many
instances in which the Revisers have removed such
“harmonistic”” corruptions from the text. The identi-
fication of them, however, involves the whole question
of the origin of the Synoptic tradition; for if a
common document forms the substratum of the three
Gospels, or, as is now more usually held, the Gospel of
St. Mark and another early record were in the hands
of St. Matthew and St. Luke, it may be questioned
whether the verbal variations which now appear in the
narratives are due to modifications of the common
original by the evangelists themselves, or to the
errors of early scribes. Even, however, if the latter be
the true explanation (which is hardly probable), the
divergences certainly established themselves at a very
early date, and the removal of them in later manu-
scripts may in most, if not in all, cases be assigned
with confidence to the editorial initiative of scribes and
not to the following of primitive authorities ; and this
class of deliberate alteration must be kept constantly
in mind by the textual critic of the Gospels.

Finally there are errors of which nothing can be
said save that they are unaccountable. Every one who
has done much writing must know that now and again
he puts down words which have no meaning in the
context in which he uses them, or (if he is copying)
are wholly unlike the words which he should have
copied. His mind has strayed, and he has written
down words which some obscure train of association
has put into his head. Errors such as these are some-
times made by the copyists of manuscripts, and since
they have no traceable connexion with the true text,
they do not, as some kinds of error do, provide the
means for their own correction. The same may be
said of errors due to the defectiveness of the manu-

12 CRITICISM OF NEW TESTARENT (vce:

script from which the copy has been made. A word
may be defaced or obliterated, and the copyist must
either omit it or guess at it; and since a copyist often
has but a hazy idea of the sense of what he is copying,
his guesses are often very wide of the mark. Errors
from mutilation would arise with especial ease during
the period when papyrus was the material in use for
literary purposes. The surface was more delicate than
that of vellum, and therefore more liable to small and
local injuries, which will obscure, or wholly obliterate, a
word or a sentence. Here again the true reading is
often irrecoverable except by guessing, and even if a
guess be right, it can rarely be proved to be right ; and
an unverified guess can carry but little weight for
practical purposes. A good example of this has
recently come to light in the «sphere of. classical
literature. In a quotation from a poem by Solon,
preserved to us by the rhetorician Aristides, where the
lawgiver is depicting the miseries of his country, a
certain section of the population was described as

τοὺς δ᾽ ἀναγκαίης ὕπο

χρησμὸν λέγοντας, γλῶσσαν οὐκέτ᾽ ᾿Αττικὴν

leVTAS.

Here the words χρησμὸν λέγοντας were practically un-
intelligible, in spite of the bravest efforts of conscientious
commentators. Various emendations were suggested,
but none was generally accepted as satisfactory ; till
at last the discovery of Aristotle’s ᾿Αθηναίων πολιτεία,
where the passage is quoted, revealed the fact that the
true reading is χρειοῦς φυγόντας. The change is not
great, only seven letters being affected; but there is
no palaeographical similarity between the false letters
and the true, to account for the corruption. It is
probable, therefore, that the two words were injured in

I FUNCTION: OF sTEXTUAL CRITICISM ὃς

an early manuscript of Aristides’ treatise, and that the
scribe of the copy from which all the extant manuscripts
of it are derived wrote down two familiar words similar
in general appearance. It is instructive to observe that
one modern scholar had, in fact, guessed approximately
the right reading ; but the guess, wanting confirmation
and not supplying in itself any explanation of the
origin of the corruption, remained wholly without
authority or acceptance. The same has doubtless
happened in many of the corrupt passages of the
classical writers, but in the New Testament the number
and diversity of the witnesses render it almost certain
that, even if such an error has vitiated one group of
manuscripts, the true reading will be preserved else-
where.

These, then, are the main forms of error with which
the textual critic has to contend ; and to meet them he
has, as has been said above, the two weapons of com-
parison of documents, and conjecture. He has before
him a number of manuscripts, and in the first instance
he may (or in the case of the New Testament it may
almost be said that he must) assume that the truth lies
somewhere among them. In many cases the choice is
obvious. Errors of spelling or grammar, when con-
fronted with the true readings, must give way at once.
Where conviction does not lie quite on the surface, the
critic who bears in mind the common causes of error
enumerated above can often see how the divergence
has arisen, and which of the conflicting readings is
original. In some cases he will see that homoioteleuton
will account for an omission; in others, that the
intrusion of a marginal comment accounts for an
addition ; in others, that two or three letters have been
mistaken by the scribe for others which resemble them.
Sometimes he may suspect deliberate alteration, whether

14 “CRITICISM OF NEW TESTAMEN Dy ἐπ

with the object of bringing out a doctrine more clearly,
or to improve the literary form of the passage, or to
reconcile two divergent readings which the scribe had
before him. By these methods considerable progress
may be made in weeding out errors, and at the same
time the critic will be accumulating materials for the
second stage of his work, namely, the discernment of
the comparative merits of his various authorities. He
will learn which manuscripts are most often right, which
are closely akin to one another, which groups are
nearest in the line of descent to the original autograph.
Hence he will have some clue to guide him when the
choice between divergent readings is not evident at first
sight. In such cases it is clearly safest to follow, as
a rule, the authority which has shown itself to be most
trustworthy. The more the parentage of the several
manuscripts can be traced, the more they can be
classified into groups, and the history and origin of the
groups made clear, the better is the chance of arriving
at a sound text of the author under examination.
Examples of the use of such methods will be found in
the succeeding pages of this handbook ; for the present
it must be sufficient to describe them merely in outline.

One proposition is so often stated as a leading prin-
ciple in textual criticism as to deserve a brief separate
mention. It is that which is formulated by Bengel in
the words, Proclivi scriptions praestat ardua, or, as it is
sometimes expressed, Dzfjiczltor lectzo potior ; the harder
reading is to be preferred to the easier. Stated so
absolutely, this proposition is misleading. Many forms
of mistake produce a reading harder than the true one.
Thus χρησμὸν λέγοντας, in the instance quoted above,
is manifestly a more difficult reading than χρειοῦς
φυγόντας, but it is none the less wrong. Similarly,
errors due to homoioteleuton often produce nonsense,

I FUNCTION OF TEXTUAL CRITICISM τ

as in the case quoted on p. 9 from the Codex Vaticanus.
In fact, it may be said generally that in the case of
accidental errors the principle is not sound; but in the
case of errors due to deliberate alteration it is generally
true. A scribe or commentator fails to understand
a passage, and puts in some word which he thinks
makes it easier; an odd word is replaced by a
commoner one; a marginal paraphrase extrudes the
phrase which it was intended to explain ; an expression
which may give offence is omitted or toned down, In
all such cases the more difficult reading is likely to be
the true one. A hard reading will not be deliberately
inserted instead of an easy one; but the reverse may,
and not infrequently does, take place. The difficulty, of
course, is to determine whether a discrepancy between
two or more manuscripts is due to accidental or
deliberate alteration ; and where this cannot be dis-
cerned with certainty, Bengel’s canon must be applied
with great caution.

Of wider application and less qualified truth is
another canon, in which this of Bengel’s is included,
namely, that of two or more alternative readings, that
one is most likely to be right which most easily
accounts for the origin of the others. The “ difficilior
lectio” is preferable just because a hard reading is
likely to be altered into an easy one, not an easy
reading into a hard one. So too a scribe, writing
without any clear comprehension of the sense of the
text which he is copying, not infrequently substitutes
a familiar phrase for a strange one, even though in
reality it reduces the passage to nonsense. Even where
both alternatives make sense, one can easily be seen
to have suggested the other, while the reverse process
is impossible or improbable. Thus in another part of
the poem of Solon mentioned above, the MS. of the

16. (CRITICISM OF NEW ‘TEST Ay ΝΥ Yeu

᾿Αθηναίων πολιτεία, in which it is quoted, has πρὶν
avrapatas πῦαρ ἐξεῖλεν γάλα, while the MSS. of
Plutarch, who also quotes it, have πρὶν av ταράξας
πῖαρ ἐξέλῃ γάλα. Here it is easy to understand how
the scribe of some ancestor of the Plutarch MSS.
(copying, of course, from a MS. in which the words
were not separated) took αν to be the familiar particle
dv, not the syncopated form of the preposition ἀνά in
composition, and so altered ἐξεῖλεν into ἐξέλῃ because
πρὶν ἄν requires a subjunctive; but it is highly im-
probable that any one, with a correct reading πρὶν ἂν
ταράξας... ἐξέλῃ before him, would be dissatisfied’
with it and alter it to πρὶν ἀνταράξας ... é&etrev.
On the other hand, considerations of sense make the
miap of the Plutarch MSS. preferable to the ΡΝ
of the Aristotle MS.

It remains to ask what place is left for the second
weapon of textual criticism, conjecture; and it has
been usual to answer that in the criticism of the New
Testament it has no place at all. Where manuscript
evidence is scanty, as it is for many of the classical
authors, it happens at times that a passage is obviously
and certainly corrupt in all the extant copies; and
then the defect must be healed by conjecture, if it is to
be healed at all. But where the evidence is so plentiful
and varied as it is for the New Testament, the chances
that the true reading should have been lost by all are
plainly very much smaller. Whether, however, con-
jecture is to be absolutely excluded depends in a large
measure on the view which the critic takes of the
character of the existing manuscript evidence. As will
be shown in a later chapter, one school of critics
regards the large majority of extant manuscripts as
representing a relatively late recension of the sacred
text, and therefore considers its evidence as of little

I FUNCTION OF TEXTUAL. CRITICISM 7

value. The number of authorities which remain is
thus comparatively small, and they differ considerably
among themselves ; and hence critics of this school are
prepared to admit that, here and there, the original
readings may have been wholly lost. Thus in Col. ii.
18 Westcott and Hort (in substantial agreement with
Lightfoot) are inclined to believe that the apostle wrote,
not AEOPAKENEMBATEYON (ἃ ἑόρακεν ἐμβατεύων),
but AEPAKENEMBATEYQN (ἀέρα κενεμβατεύων), the
mistake being palaeographically very easy, and the
improvement in sense through the conjecture con-
siderable.’ It is universally agreed, however, that the
sphere of conjecture in the case of the New Testament
is infinitesimal ; and it may further be added that for
practical purposes it must be treated as non-existent.
No authority could be attached to words which rested
only upon conjecture; and a critic who should devote
himself to editing the Scriptures on conjectural lines
would be merely wasting his time. Where nothing
but questions of literary style are involved, we may be
willing to accept a reading upon conjecture, if no
better evidence is to be had; but where it is a
question of the Word of Life, some surer foundation is
required.

Putting conjecture aside, therefore, the function of
the textual critic is, first, to collect documentary
evidence, and, secondly, to examine it and estimate its
value. The object of the present volume is to show
what has been done in both these directions. In
Chapters II—VI. an account will be given of the
available textual material—the copies of the New
Testament in the original Greek, the ancient translations
of it into other languages, and the quotations from it

1 For other examples of conjectural emendations proposed in the N.T.
text, see Nestle, /ntrod. to the Textual Criticism of the N.7T., Eng. Tr. pp.
167-170.

c

18 OCRITICISM OF NEW VEST AMPND ica: a

which are found in the early writers of the Christian
Church. The materials having been thus passed in
review, an attempt will be made in Chapters VII. and
VIII. to summarise what has hitherto been done in the
way of using these materials, to discuss the principal
theories now current with regard to the early history of
the New Testament text, and to estimate the general
position of the textual problem at the present day. It
is all well-trodden ground, and each newcomer is
infinitely indebted to the labours of his predecessors ;
but it is ground which each generation must tread
afresh for itself, if it is to keep its interest alive in a
subject of such importance, and if it is to add ever so
little to the knowledge which past generations have
handed down to it. It is but a humble part that
textual criticism has to play. It is but the temple-
sweeper in the courts of the Lord; but honest labour,
even in that humble field, is not lost.

CHAPTER? If

THE AUTOGRAPHS OF THE NEW TESTAMENT

(Authorities: Sir E. Maunde Thompson, Handbook of Greek and Latin
Palaeography (London, 1893 ; a larger work is in the press) ; Kenyon,
Palaeography of Greek Papyri (Oxford, 1899), and arts. on ‘‘ Papyri”
and ‘‘ Writing” in Hastings’ Dictionary of the Bible (1902, 1904) ;
Gardthausen, Das Buchwesen im Altertum (Leipzig, 1911). ]

THE manuscript history of the New Testament covers
a space of fourteen hundred years, from the original
composition of the several books in the latter part of

the first century to the invention of printing in the ᾿
latter part of the fifteenth. It falls into three periods,
distinguished by well-marked differences in the style of
the writing employed. The first, from the middle of
the first century to the beginning of the fourth, may
be called the Papyrus period, during which copies of
the Scriptures (as of all other books) were normally
written upon papyrus, and in the style of writing suit-
able to that material. The second, from the fourth
century to the ninth, is the Uncial period, when the
material was vellum and the writing in large characters,
each formed separately ; while the third, from the ninth
century to the fifteenth, is the Minuscule or Cursive
period, in which the material is sometimes vellum and
sometimes (from the fourteenth century onwards’)

1 Paper was made and used in Europe in the thirteenth, and even in
the twelfth, century; but it only comes into anything like common use
in the fourteenth century, and never entirely superseded vellum before the
invention of printing.

19

20. CRITICISM OF NEW TESTAMEND,) cu:

paper, and the writing is in small characters, often
linked together into a running hand. Each of these
periods will have to be described in detail in this and
the following chapters.

From the papyrus period, covering the first three
centuries of the history of Christianity, no complete
copy of any book of the Bible in Greek is extant. Of
all the copies which must have been written during
that period, only a handful of fragments, mostly the
smallest of scraps, is at present known to exist, and
the majority of these belong to the Old Testament,
not to the New. There are some portions of Genesis
(xiv; 21-23; xv. 5-0, xix: 2. Ἐκ, ΤΊ τ 25-47.
XXVii. 32, 33, 40, 41) among the Oxyrhynchus papyri
(now at Oxford) ; a fragment of a Psalter (xii. 7—xv. 4)
in the British Museum; another, more doubtfully
dated, at. Leipzig. (κυ. 17-63); a few. versesyor
Isaiah (xxxvili. 3-5, 13-16) at Vienna; a fragment
of Ezekiel (v. 12—vi. 3) in the Bodleian Library ; and,
of the New Testament, small portions of St. Matthew
Gi, /1=0,) 12;..14-20), St, John (23734, Sarak ee
11-17, 19-25) and Hebrews (ii. 14—v. 5, x. 8—xi. 13,
ΧΙ. 28—xiil, 17) among the papyri from Oxyrhynchus
(now allotted respectively to Philadelphia, Oxford, and
the British Museum). The earliest of these is probably
the Genesis fragment, which may even be as early as
the second century; the latest (and also much the
largest and most valuable) the Hebrews papyrus, which
may be as late as the fourth. To the fourth century
also belongs an important papyrus of the Psalms, at
Leipzig, and several small scraps." To these may be
added, not as being manuscripts of the Bible but as

1 There are also some papyrus fragments of later dates, belonging to
the period after the introduction of vellum, but before the use of papyrus
had been entirely abandoned. For a complete list of these (so far as the
N.T. is concerned), see p. 41 ff.

x1 AUTOGRAPHS OF NEW TESTAMENT 21

more or less connected with them, two fragments of
logia or sayings of our Lord,’ discovered at Oxyrhyn-
chus, two small portions of canonical Gospels, also
from Oxyrhynchus, and the still more tiny scrap in
the Rainer collection which appears to contain an
uncanonical version of St. Peter’s denial.” But these
fragments, interesting as they are as examples of what
the soil of Egypt may yet be holding for us, are for
the most part so slight in themselves as barely to
deserve mention ; and if the papyrus period possessed
no importance beyond what they could give it, it might
be passed over very lightly and briefly indeed. Its real
importance lies in the fact that it is the period to which
the autographs of the New Testament belong, and that
by indirect means we can learn something as to the
appearance of these autographs and of the conditions
under which the Christian Scriptures circulated during
the first three centuries of their existence.

It is only within the present generation that this
knowledge has been made accessible. Before the
nineteenth century, manuscripts upon papyrus were
practically unknown, and it is only within the last
twenty years that they have been known in sufficient
quantities to provide a continuous record of palaeo-
graphy during the ages which preceded the rise of
vellum. Now, however, thanks to a succession of
discoveries in Egypt (the only country of which the
air and soil are dry enough to preserve the brittle
material), literary manuscripts upon papyrus can be
counted by the hundred, and non-literary documents
by the thousand, and we are in a position to realise
with fair accuracy the appearance of a Greek book
| hin Christi, ed. Grenfell and Hunt (1897); Mew Sayings of Jesus
1904).

See Bickell, Afittheclungen aus der Sammlung der Papyrus Ersherzog
Rainer, i. 52.

22) (CRITICISM OF #NEW TESTA BIE NE ier

during the three centuries which lie on each side of the
beginning of the Christian era.

A papyrus book was very unlike a vellum book, or
such volumes as we are accustomed to at the present
day. The material of which it was made was much
more fragile than vellum, but perhaps not more so
originally than paper, though by lapse of time it has
in almost all cases become so brittle as to require the
most careful handling. It was composed of the pith
of the stem of the papyrus plant, which grew plenti-
fully in antiquity in the Nile and its adjacent marshes.
The pith was cut longitudinally into thin strips, which
were laid side by side to form a layer, while a second
layer was composed of similar strips at right angles
to those of the first layer. The two layers were then
pressed together, probably with the assistance of a
little glue, to form a sheet, the height of which might
vary approximately from 6 to over 15 inches, and its
width from 3 to 9 inches. Several of these sheets
were then fastened together so as to form a roll, the
length of which (in the case of Greek books) rarely, if
ever, exceeded 30 feet, while it might be very much
less. On one side of the roll, as will be seen from the
above description, the fibres of papyrus lay horizontally
or in the direction of the length of the roll, while on
the other they lay perpendicularly or in the direction
of its height ; and it was the former (technically known
as the vecto) that was primarily intended to receive the
writing. The back (or verso) would normally be left
blank, though an author whose matter outran his
available stock of papyrus might occasionally be
reduced to writing upon it,’ or some one who desired

1 Thus the description of the book seen by Ezekiel in his vision (Ezek.
11. 10), ‘written within and without, and there was written therein

lamentations, and mourning, and woe,” implies a great superabundance of
matter (cf. Rev. v. 1).

nw AUTOGRAPHS (OP ONEW ‘TESTAMENT 28

a copy of a literary work for his own private use might
sometimes inscribe it on the back of a roll which
already had other writing on its recto.’

The writing on a papyrus roll was arranged in
columns, the width of which varies considerably in
different manuscripts. In the case of poetry it would
naturally be determined by the length of the lines, and
a hexameter verse, written in good-sized characters,
occupies from 5 to 6 inches, or even exceptionally as
much as 8 inches. In prose works the columns are
almost always much narrower, generally measuring
from 2 to 3 inches, though there are examples of
columns as narrow as 12 inches, and as wide as 34
inches. The only prose manuscripts at present known
in which the columns are wider than this are a few
which are not written in a formal literary hand, but in
the running hands of everyday life, and evidently were
intended solely for private use. The columns not
infrequently lean a little to the right. Corrections are
written between the lines, or, if too long for this
method, in the margins at the top and bottom of the
columns. Only very sumptuously written MSS. have
sufficient space between the columns to allow corrections
or notes to be inserted there.

Greek writing upon papyrus falls into two main
classes: the literary hand, for use in the transcription of
books, and the non-literary hand, for use in business
documents and private letters. In the former, ligatures
between the letters are rare, the characters being for
the most part formed separately. They are smaller
and less formal than the writing of the best vellum

1 The unique MSS. of Aristotle’s ᾿Αθηναίων πολιτεία and the Funeral
Oration of Hyperides are so written. The use of such volumes is
mentioned as a sign of poverty by Lucian (V2t. auct. c. 9), where
Diogenes promises his disciple, ἡ πήρα δέ σοι θέρμων ἔσται μεστή, καὶ
ὀπισθογράφων βιβλίων. Copies so written may have been articles of
commerce in a small way, but hardly as part of the regular book-trade.

“1 CRITICISM ‘OF NEW: TESTAMENT Wen,

uncials, but are carefully and gracefully written. Dated
or approximately datable examples of the literary hand
are not common, but the general sequence of develop-
ment can be made out, from the small and angular
hands of the Ptolemaic period to the larger, squarer,
and more rounded forms which characterise the period
of Roman rule in Egypt. Side by side with this goes
the non-literary hand of everyday life, the stages of
which can be fully traced by the aid of large numbers
of dated documents from the beginning of the third
century B.C. onwards. From the first (that is, as far
back as our knowledge at present extends) it is written
with the utmost freedom, showing that even persons of
quite moderate stations in life were fully able to write
a running hand with ease. For the purposes of textual
criticism the non-literary hand only requires notice
owing to the fact that literary works were sometimes
copied in it by private individuals for their own use, and
that copies of this kind may have entered into the
textual tradition of the New Testament during the
early days of Christianity. The distinction between
the literary and non-literary hands of any given period
is, roughly speaking, the same as that between print
and writing nowadays—the one carefully formed, with
separate letters of fair size, the other cursive and
irregular, sometimes large and coarse, sometimes small
and ill-formed, sometimes neat and flowing; but the
distinction, though obvious enough between average
specimens of each type, is partially obscured by
approximations in each to the style of the other, the
literary hand admitting of ligatures between the letters
to some extent, while non-literary documents sometimes
approach the care and formality of the literary type.
Aids to the reader, such as accents, breathings, and
punctuation, are not so wholly wanting in papyri as

nu, AUTOGRAPHS: OF NEW TESTAMENT 2s

they are in the vellum manuscripts of the uncial period.
It is true that non-literary documents are almost entirely
without them, and that they are nowhere supplied so
fully as they are in a modern printed text ; but several
literary papyri are partially equipped with accents, and
have a rudimentary system of punctuation. So far as
can be gathered from the extant specimens (and it
must be remembered that these generally come from
the personal possessions of private individuals in country
towns and villages, not from the great libraries of a
capital), the more carefully a manuscript is written, the
more fully is it supplied with these aids to the under-
standing. The words are not separated in any case,
but accents are placed upon the longer and more
deceptive words, as to which mistakes were most likely
to be made, and the more important pauses are marked
either by small blank spaces in the text, or by a dot
above or in the line of writing, or by a short line
(paragraphus) drawn below the beginning of the line in
which the pause occurs, or by some combination of
these devices. Capital letters, which are occasionally
used in business documents to mark the beginning of a
clause, do not occur in literary papyri; nor are lines
left unfinished at the ends of paragraphs.’

This description may serve to give some idea of
the appearance and character of the original autographs
of the New Testament. That they were written on
papyrus hardly admits of doubt. It is true that skins
had been used for the reception of writing in Palestine

1 Fuller details as to papyrus manuscripts, with facsimiles, may be found
in Sir E. Maunde Thompson’s palaeographical works (see note on Authorities
on p. 19), or in the present writer’s Palaeography of Greek Papyri (1899). A
handy collection of facsimiles (both literary and non-literary), at a very moderate
cost, is contained in Schubart’s Griechische Papyri (50 plates for 6 marks,
Bonn, 1911) ; while the fullest collection of non-literary hands is to be found
in the atlases accompanying vols. i.-iii. of the Catalogue of Greek Papyri in
the British Museum.

26. ‘CRITICISM OF SNEW TESTAMENT) Vcr

and elsewhere at an earlier date, and from the tradition
recorded in the Talmud, which required all synagogue
rolls to be so written, it is fair to conclude that the Old
Testament books were habitually written on skins in
the first century ; but this proves nothing as to the
material commonly used for ordinary writing, and the
first point to be remembered in trying to reconstruct
the early history of the New Testament books is that
originally they would be regarded as ordinary books,
and not as sacred. Skins, which might be employed
for the purposes of public service in the synagogue,
would be too cumbrous for books intended for free
circulation, and especially for letters; vellum did not
come into general use anywhere until two or three
centuries later; and there is thus every reason to
suppose that papyrus, which we know from the state-
ments and allusions of contemporary writers to have
been the material universally employed in the neighbour-
ing country of Egypt, in Greece, and in Rome, was
also commonly used in Palestine. Even if there were
any doubt upon this point, it would only affect the
books which were written in that country ; and these
can, at most, be only the Gospel of St. Matthew and the
Epistle of St. James. For the remaining books of the
New Testament, which were written in Greece, in Asia
Minor, or in Rome, we may say with confidence that
they were originally written on papyrus.

This being so, it has been shown above that we
now have adequate material for ascertaining their
general character and appearance, by an examination
of extant manuscripts of the same date. Among the
papyri already publicly known, there are over 350
precisely dated documents belonging to the first
century, and some seventy literary works (most of
them, however, small fragments), which, though not

uy AUTOGRAPHS OF NEW TESTAMENT. 27

possessing exact dates, may be assigned to this period.
It is true that all these were written in Egypt, while
the autographs of the New Testament were all written
outside that country ; but there is not much force in
this consideration. No doubt, if we had as many
extant examples of writing from each of these other
countries as we have from Egypt, it is probable that an
expert would be able to detect local types of calligraphy,
and assign manuscripts to their respective countries
on their handwritings alone, just as he is able to do
with mediaeval Latin manuscripts. But the instance
of these very manuscripts shows the extent to which
alone these local variations affect the general develop-
ment. It is not necessary to write a separate history
of palaeography for each nation. The local differences
may be discerned by long experience, but the general
development of writing is the same throughout. It is
generally easier to tell the date of a MS. from its
handwriting than its country. So, there is every
reason to suppose, it would be with papyri, and with
this further reason for expecting uniformity, that all
Greek manuscripts, in whatever country, would be
written either by Greeks or by those who had learnt
their writing from Greeks. Nor are we without means
of verifying this belief. The papyri from Herculaneum,
though they are not exactly like any of the Egyptian
papyri, yet do not differ from some of them more than
they differ among themselves. The two most recently
discovered papyri of Hyperides, those of the speeches
against Philippides and Athenogenes, find the nearest
analogies to their handwritings in some of the Hercu-
lanean manuscripts. Similarly, to take an example of
a non-literary hand, a Latin papyrus containing a deed
of sale of a slave boy, executed at Seleucia in Syria
in the year 166, has a Greek subscription in a hand

28 CRITICISM’ OF NEW “TESTAMENT? | A’tx.

essentially the same as those current in Egypt at the
same time. There is thus every reason, both a priori
and a postertort, to hold that what we know of writing
in Egypt during the first century may be applied with
confidence to writing in other parts of the Graeco-
Roman world.

We may refer, then, to the extant papyrus manu-
scripts of the first century as fairly representing the
general appearance of the New Testament autographs.
Chief among these are (1) a beautiful copy of the third
book of the Odyssey! in a rather large and graceful
hand, without ligatures, in columns about 6 inches
broad, with wide margins, written about the beginning
of the first century; (2) a large papyrus containing
three orations of Hyperides,’ less handsome than the
Odyssey, but still neatly written in a well-rounded hand
with a few ligatures, in columns about 2 inches wide
and slightly leaning to the right, written about the last
quarter of the first century ; (3) a copy of the oration
of Isocrates, De pace,” written about the same date, in
columns of about the same size, but in a much less regular
and ornamental style, more approaching the cursive
type ; and (4) the well-known papyrus of the ᾿Αθηναίων
πολιτεία οἵ Aristotle,’ written between A.D. 90 and 100
in four different hands, of which two are very small
and cursive, belonging wholly to the non-literary style
of writing, while one is a larger and less well-formed

1 Brit. Mus. Pap. 271 ; facsimile in Palaeographical Society's publications,
11: 182.

2 Brit. Mus. Papp. 108 and 115; complete facsimile in editions by
Babington ; specimen facsimiles in Pad. Soc. i. 126, and Catalogue of Ancient
MSS. in the British Museum (Greek). Plate I. in the present volume shows
the last four columns on a reduced scale,

3 Brit. Mus. Pap. 132; specimen facsimiles in Classical Texts from
Papyri in the British Museune.

4 Brit. Mus. Pap. 131; complete facsimile published by the Trustees of
the British Museum; specimen in Pal. Soc. il. 122. Plate 11. shows a
portion of the eighth column of the MS., written by the first (and _ principal)
hand.

ge ‘¢ avf OF

(‘JUSUIeISAT, MON 911} Jo sydvisoine

911 ya Areroduia}yu09

‘puevy Aresoyy ul [or snaXded jo 9] Δ πἸΌΧ 6: 1 9]809)

‘AINJUDD ISI 9181 “‘OddINAXNUH Οἷα SHCINAdAH
: . ᾿ς Pa τ ΚΣ ape a z ᾿ς pets . νωἷ-:
ΓΘ τα Ξ Ψ τῷ: . pins SS ud = =a ἀπο 4 : αὶ ‘ © i Ὡς ey, ‘ : [
eee Sas! δε ss 5 3 wit ᾿ τ Ὡς, ᾿ Ἐν a Nice Secale Na rte te [ἢ
τὰν τῇ : ws = " τὰ Ἢ ? ἘΣ ᾿ Ree ek
moe Ῥ , νὰ Fr ᾿ = fe. κι oN ᾿ ᾿" *
ao eae Sk ef Olt: ae ἐπὶ vei eres Szid Ε ἐν 11. τὰ, Ι
one ee eee Ἷ τὰν Y wt {Sak ᾿ τ"
cae. oa Sat νόος oy πῶς εν ουοῦ ἃ νου πίχ oe
Fase Be Fooe OR DOMEICOL SY “neon. τι θεν στον we Paorria τρουεθα στοῦ, οὐ
Θ᾽ ἫΝ he > Lat ae) ob emer ἱ t : πῆς TEX Diecast” 3 K jonni NeoEol ” 1: ῖ Seon μλονοιζρώλρι cf
ey Ἢ ξκαχτς, ᾿ς ae ar had, > Abr sy Nore | a BOER PERDEROCEN Ως. come “δα οικονϑύραλο
᾿ rere Ba. εἴ a= % Ay “J at eee ; “3 Boy tou AOdad; ; i ee re I IIIA. : ἢ AQL WLAR NRG Fe B45
ae 3 eee é ASS : i ΕΙΣ τς ἐξ; Ξ eS
ΞΡ Be ON thay staal re ἦ; μα AIOE eh ONE TION beKH dog sf) nebo AO GOLL MARS “2
7 Bibs we “Sire OX MRT at se SPSS NONI L Ieee eas ἘΣΤΕ τε '
os rE Se
τὶ ieee yee SEEING Ty | a7 τσ ΟΝ NU WALD
: tke LALK aOR ee ΞΡ
J ae ig ἢ τ >
Ἔ prety να Lord Es: ; : ;
: ἦτ << e ἘΠ ἧς {ΕΞ = Spee: ᾿ a ane SCOALA | : τ nie Υ Se.
3 uses prea ita soe sae. ee babes ei “4 “LIED OJ OHLRESROL Το τ Τ rHanie~erolved SS
PARSE NAONSY Fou areal ὃ ie aD τ co Raurenorangnost: = Salers Ud NOG D ESS
“ars hotter ak, τως Δ Oh ny δεὶ ᾿ PALIN NLD a Ree WOLOZ τε NOS ALOT στὰ 7
he MLD MOM ΑΓ fo 5. x : : Ὕ >" eee oe aes
ineai geese teva} kD Rtid Nomen A Ty περφολοτζοριδλορκᾷ + |, Ἰοιλονοο σωλμοννα, Ἐπ τ
" ἴδ νον ὁ, ee oe t same sies tbs oo ἜΣ WELL RENIN: ἀπ ie ae VECO ALES iy oy
τοὶ ἐστοινϑυο 5 ha CRANE που τς art ie + IOXANDIAS none * Ξ οὐβνωρστενααν, ee et βὸς
LONE WEY i Noas ; NIE INAOLO SEK, Be ee
ae a ian pS Ἶ ἡ ae πλοῦ κὸν EIN LON’ UE te
Ἐπ me LLIN: of ty ou OU ITANS NADL. : qu ᾿ ¥ α ἢ Ἰ OTOL Cet ONOLEI οὐδ
pois | Manon beg «torneo eye | i ee me Ἕ sole MINONO χα. τις
~heings AMON ML | a LOOWAOSAO NOI - a: ieee Bisco RINOLT hae -
> pee ok deve Pura tinde-ng aye 1 OUR LAN OMLIS | τοί oo} 5 BH DIO nr BOLING Nese τ
; NODLIG LYE NONLE : ΕΣ cu pieced: : ae Ἐς ἘΠ tee aie LON —-
‘ 4 uk PN ew
Oi gts2 δε συχοος Ty - na NOMEN : ᾿ Sees ᾿ coy

rad ny are Salone ty *

n AUTOGRAPHS OF NEW TESTAMENT 29

cursive, and one a fairly regular but ugly uncial,
evidently the work of an illiterate scribe. These four
manuscripts are representatives of four classes of work-
manship, to one or the other of which all the New
Testament autographs must belong. The first is the
work of a thoroughly good professional scribe of the
best type; the second is a good ordinary professional
hand; the third is the work of an educated man, not a
professional scribe, writing a careful copy of a literary
work ; while the fourth is the running hand of common
everyday writing.

To which class the original autograph of each of
the New Testament books belonged depends upon the
circumstances under which they were severally written ;
and these must remain in large measure a matter of
conjecture. To take first the Pauline Epistles, as the
earliest group among them, with the possible exception
of the Epistle of St. James. These, we know, were
not generally written by St. Paul with his own hand,
but by one of his companions, Tertius, Sosthenes,
Timothy, Silvanus, or some other, an autograph
sentence or two by the apostle being invariably added
at the end. Now, while it may be taken for certain
that these were educated men, there is no reason to
suppose that they were trained professional scribes ;
and what we have to expect from them, therefore, is
the careful writing of the educated amateur. The
epistles would be written carefully, because they were
weighty compositions, intended to be read more than
once, and perhaps circulated among the neighbouring
churches; on the other hand, they would not be
written by professional scribes, because they were not
books, but letters. One is thus led to think of the
two last of the types mentioned above, the Isocrates
and the Aristotle. The Aristotle hands, however, are

30° CRITICISM OF ΝΕ TESTAMENT*:. δα,

hardly suitable, some of them being too rapid and
cursive, with small letters and contractions, while one is
the work of a hireling, and a rather uneducated one.
The Isocrates, which is neither uneducated nor pro-
fessional, seems to come nearest to the general type of
the Pauline autographs. One little circumstance may
be taken as confirming this conclusion. In a well-
known passage at the end of the Epistle to the
Galatians, St. Paul refers to his own writing as a large
hand: “ See with how large letters I have written unto
you with my own hand.” It may be remarked in
passing that exact analogies to this may be found in
many Egyptian papyri, where the body of a document
is written by a friend or a clerk, and the principal
appends his ratification in a large hand at the end;
but the point specially noticeable here is that the
phrase implies that the body of the epistle was written
in a hand of small or medium size. It cannot have
been an uncial such as those of the great vellum MSS.
which we all know, neither can it have been such a
hand as that of the Odyssey, or even the Hyperides,
described above, which are themselves large and bold:
it must have more nearly resembled the hand of the
Isocrates. Of course this is merely suggested as the
generic type. The individual hands, no doubt, differed
in different epistles, and none would be precisely
similar to this.

Of the circumstances of the Catholic epistles we
know less, and can therefore say less. The second and
third epistles of St. John are private communications,
and would almost certainly be written in a private
hand, and on small sheets of papyrus. The first
epistle is a more formal document, and, if it accom-
panied the Gospel, may very probably have been written
in the same manner. The epistles of James, of Peter,

‘of “J 295. o7

( Ὑπϑιι[Β159.1, MIN 91 Jo
5160 .1501Π0 ay} ya A1e10durayuos ‘puey Aiviayy-uou ur 10. snakdvd jo ΘΙΟΠΙΠΈΧΉ

δ: ἡ a[vos)
᾿ΑΙΏγ119.) ISIN] 9101 ὙΠΗΠΙΥΟΙΙ NOIVNHOV ἩΊΙΟΙΘΙΣΙν

ΤῸΝ = as μὰ" hig i

wes Ἀφ so

ngs dlls ae “ ἜΘΟΥ τὸ
ἐτραιμνοα hp,

τ ρει

ἢ ἢ Peres μους

mn μρέμάμφ. ὅτ, κα fa 2 af
tine teehee RB me ow Ἢ “2 4:

o> PEM Ge i ἡ 19 Pokies ets asin εἶ

2h ete

im MoU LOGRAPHS OF NEW TESTAMENT? :31

and of Jude are formal and deliberate writings,
intended for circulation among many churches, and
must be treated as literary works, not as private and
personal correspondence. They may, therefore, have
been formally written by professional scribes, the
author’s autograph never coming into circulation at all.
One may, therefore, think of the hand of the Hyperides
or an improved form of the second Aristotle.

There remain the narrative books, namely, the four
Gospels, the Acts, and the Apocalypse. Of the publi-
eation. of, the latter, and. of tthe extent. to’ which; it
circulated, we know too little to be able to judge of
the character of its manuscripts; but it cannot be
doubted that it must have originally existed in the
personal autograph of the author. The author speaks
of himself as writing and as being told to write; and
one can hardly conceive of experiences so intimate
and personal being dictated in the first instance to a
professional scribe. With the Gospel of St. John the
case is different. Composed at the end of a very long
life, it may well have been taken down from the
apostle’s lips rather than written by his hand. One
may compare the circumstances of its first composition
with those of the first translation of it into English,
taken down by the disciples of Bede from the lips of
their dying master. The Gospel thus dictated seems
to have been finally issued by a committee of the heads
of the Church of Ephesus; and a book thus composed,
and issued in the midst of a flourishing church during
a time of toleration, can hardly have been written
except by a professional scribe, under the revision of
the committee of elders. Here, if anywhere, one may
think of an original copy of a New Testament book as
written in the best style of contemporary calligraphy.
It is only unfortunate that we have so few specimens

32. | CRITICISM "OFF NEW THSTAMENT (cu.

definitely assignable to the end of the first century to
enable us to judge what that was like. Some Homer
papyri, probably written about this period (Brit. Mus.
Papp. “του; 114, 732), in a clear, squafe, but! 'semie-
what unornamental hand, or two finely written Hesiod
MSS. at Strassburg and Berlin, may furnish a clue, or we
may think of an improved specimen of the Hyperides
type.

The Gospel of St. Matthew was probably written in
Judaea, and the same resources of penmanship would
perhaps hardly be found there ; moreover the Jews, for
whom it was intended, were a less literary and aesthetic
people than the Greeks. Whether, if the Gospel were
originally in Aramaic, its Greek form was also written
in Judaea is a question which it would be rash to
discuss and impossible to solve; but in any case the
second of the considerations just mentioned, the
Jewish character of the public for whom it was
intended, remains unaffected, so that one would not
look for any very ornamental form of writing as
representing its original appearance. The Gospel of
St. Mark was probably written in Rome, where scribes
were plentiful and good; on the other hand, the
supporters of the Church were not relatively so
important or so numerous as in Asia Minor, and it
would not command so public a circulation. In any
case the author could have been responsible for nothing
except his own autograph ; and this, as the writing of
an educated man but not a professional scribe, would
be on the same footing as the autographs of the epistles
of St. Paul, some of which may even have been written
by Mark himself. It may, therefore, be referred to the
Isocrates type, or, if it were only the author’s private
draft, to that of the principal Aristotle hand.

The Gospel of St. Luke and the Acts go together,

11 AUTOGRAPHS OF NEW TESTAMENT 33

as the works of the same author, and issued in the
same manner. Here a special feature is introduced by
the fact that both books were dedicated to a definite
person, Theophilus, and one who, as appears from the
title κράτιστος attached to his name, held high rank
in the official world.’ It is evident that one copy, and
that a carefully and even elaborately written copy, would
be prepared for presentation to this dignitary (though
Dr. Salmon is almost certainly wrong in saying that
it would be on vellum), and that another copy would
be the archetype from which the transcripts for general
circulation would be made. It is possible, if Luke
himself were a good calligraphist, that the presentation
copy would be in his own hand; but it is more
probable that it was written by a professional scribe from
the author’s autograph draft, and that the same scribe
or another made from the same draft the first copies
for circulation. Professor Blass’ theory (of which more
will have to be said in a later chapter) that the diver-
gences between the two editions of the Acts are due to
Luke having prepared two originals, one for Theophilus
and one for the public, making various small alterations
in the second, implies that Luke acted as his own
scribe; and this can hardly be assumed. On the
other hand, Sir W. M. Ramsay’s theory that the Acts
was never finished would make it doubtful whether the
presentation copy for Theophilus was ever written out
at all.

Another characteristic of the New Testament auto-
graphs to which the extant papyrus MSS. give us a
glue is the length of the’ rolls in which they ' were
contained. The shortest epistles, namely the second

1 In Egypt the epithet is applied only to the Prefect, the δικαιοδότης, and
the ἰδιολόγος, whose jurisdiction extended over the whole province, and to
the three epistrategi, or governors of the three great districts into which it was
divided,

D

34 CRITICISM ΟΕ NEW “TESTAMENT οἷ;

and third of St. John, might be written in a single
rather broad column on a sheet of papyrus about
11 inches high by 6 inches broad,—a size of which
there are many extant examples among the recently
discovered papyri of the first and second centuries.
The epistle of St. Jude and that to Philemon could
be written in two moderate columns on a slightly
broader sheet. 2 Thessalonians would be a roll of
about 15 inches in length, arranged in some five
columns. ‘The other epistles of St. Paul would range in
length upwards from this to about 11 feet 6 inches in
the case of the epistle to the Romans. The Apocalypse
may be estimated at 15 feet. Of the longer books,
the Gospel of St. Mark would occupy about 19 feet,
that of St. John 23 feet 6 inches, St. Matthew 30 feet,
the Acts and St! Luke's Gospel about’ (31) or 32'feet
In all these cases a copy written with especial elegance
would require a somewhat greater length."

1 For the purpose of comparison, the dimensions of some of the extant
papyrus MSS. may be stated. The Hyperides MS. mentioned above
(containing the oration against Demosthenes and those for Lycophron and
Euxenippus), must have measured, when perfect, about 28 feet; a MS. of
the last two books of the //zad, about 25 feet ; the Mimes of Herodas, about
the same ; the Isocrates mentioned above, 14 feet ; and the longest of the rolls
of the Aristotle, over 7 feet. In height the Hyperides measures 11# inches, the
Isocrates and Aristotle τὰ inches, the Homer 9# inches, and the Herodas
(which is in the nature of a pocket volume of poetry), only 5 inches. The fine
Odyssey papyrus measures 13 inches in height, and, supposing it to have con-
tained the first three books of the poem, would have measured about 24 feet.

An ingenious theory with regard to the New Testament autographs was
formerly proposed by Mr. Rendel Harris (‘‘ New Testament Autographs,” in
the American Journal of Philology, No. 12, supplement). Observing that the
columns of the Codex Vaticanus consist of 42 lines, and that the ends of the
several books very frequently fall in or about the 14th, 28th, or 42nd line,
he argued that the columns of the archetype from which it was derived
probably consisted of 14 lines of similar length. On like grounds he
concluded that the columns of the Codex Sinaiticus, which consist of 48 lines,
represent four original columns of 12 lines each, ‘Thus, since each page of the
Vaticanus contains three columns, and each page of the Sinaiticus four
columns, it appears that they respectively contain three times three, and four
times four, of the supposed original columns; and this he supposed to be
the meaning of the well-known passage in Eusebius, where it is said that the
copies of the Greek Bible, made at Constantine’s order for the churches in his

1 AUTOGRAPHS OF NEW TESTAMENT 35

One important deduction from the figures which
have just been given is that no complete copy of the
New Testament in a single volume could exist during
the papyrus period. Such a copy, even when written
in a small hand and with narrow margins, would
occupy a roll more than 200 feet in length, which is
far in excess of even the largest Egyptian papyri
(which, being intended less for reading than for show,
are often of great length), and is seven or eight times
the length of an average Greek papyrus. It would
not even be possible to include all the Gospels in a
single roll. Each book must have existed and circu-
lated separately, and if any given individual possessed
one of them, he would not necessarily possess the rest
also. Further, it will be understood (and this is a
point which applies also to vellum MSS.) that it was
not very easy to verify references, when divisions of
the text were few and a numeration by chapters and
verses unknown. The earliest known division of the
sacred text into sections does not reach back into the
papyrus period. Hence ignorance of any particular

new capital, were sent ἐν πολυτελῶς ἠσκημένοις τεύχεσι τρισσὰ Kal τετρασσά.
The theory is very ingenious, but unfortunately it is inconsistent with the
evidence derived from the extant papyri, as stated above. (1) Columns of
the size supposed by Mr. Harris imply rolls of papyrus only 5 or 6 inches in
height ; and these are never found except in the case of the Herodas MS. and
one or two other pocket volumes of poetry (e.g. Berlin Pap. 10571, Brit. Mus.
Pap. 1824). Such a format would be as unsuitable for a Gospel as that of
the Temple Shakespeare for a Bampton Lecture. (2) Mr. Harris argues that
the early Christians, being poor, would use papyrus of very moderate height,
because it was cheaper ; but this argument restsona misapprehension. Pliny
gives the dimensions of various kinds of papyrus, stating that the largest was
the most expensive, while the smallest was the cheapest ; but his dimensions
apply, not tothe height of the papyrus roll, but to the width of the sheets out
of which a roll was composed. ‘The papyri discovered in Egypt show that
even the poorest people used papyrus measuring 9 or ro inches in height,
and upwards. (3) Mr. Harris’ theory would require rolls of excessive length
to contain the principal books of the New Testament. The longer epistles,
such as Romans and 1 Corinthians, would become portly rolls of 25 feet in
length, while the Gospel of St. Luke and the Acts would reach the impossible
dimensions of 70 feet or thereabouts.

360 | CRITICISM GP NEW) TESEAMENT.) οἷ

book on the part of an early Christian writer does not
necessarily imply the non-existence of that book at
that time; and inexactness of quotation is to be
expected rather than wondered at.

It is to its fortunes during the papyrus period that
the New Testament owes its peculiar textual history.
All other works of ancient literature which have come
down to us were written avowedly as literary works,
and were formally copied by professional copyists from
tlie) veryiearliest, times of their existence. “Thevexts
which we now possess of Sophocles or of Virgil are
linked to the authors’ original autographs by a con-
tinuous chain of formal transcripts, each of which was
intended for public use—in most cases, it is probable,
for preservation in a public library. We cannot say
the. same. of ‘the text of the New Testament. The
several books were not composed in the first instance
as literary works, nor by men whose profession was
that of letters. They were not copied for their literary
skill, but for the substance contained in them, the
record of facts of intense importance or the admonitions
of a revered and authoritative teacher. Many of the
early copies were no doubt made by private individuals
for their own use, few or none as ordinary items in the
book-trade. From time to time the owners of them
were the objects of persecution, and the sacred writings
themselves were condemned to be destroyed; and in
such cases the official copies preserved by the churches
would be the most likely to suffer, while the private
copies would escape more easily and would be the
sources from which, when the storm of persecution was
past, the stream of tradition would be re-established.

Under these circumstances it was only natural that
great divergences should spring up in the text. Even
the classical texts, which have been handed down from

xz AUTOGRAPHS’ OF NEW ‘TESTAMENT 7

library to library, have suffered much in the process
—how much, we cannot always tell, for want of a
sufficient variety of authorities ; and with the Christian
Scriptures it was inevitable, humanly speaking, that
they should be much corrupted during these early
centuries, when in some cases verbal accuracy was little
thought of, in others it was unattainable by reason of
the position of the Church, and when the copies made
by private individuals for their own use played an
important part in the textual tradition. What such
copies might be like we learn from some of the
papyrus MSS. of ‘classical authors which have been
discovered of late years. Many of these MSS. were
evidently private copies, or, at least, did not come from
any of the great museums, but from the little libraries
of ordinary provincials ; and it is a recognised fact that
the texts of these papyri are less pure than those of
the best vellum MSS., although they are often a
thousand years older. Of all the papyri of Homer
which have been discovered (and they are many), none
is so good as the great vellum Venetian MS. of the
tenth century.:/ It, is’ δ΄ representative’ of \the purer
channel of tradition, which ran through the great
libraries ; while the papyri (some of which are ex-
-cessively corrupt) show what might happen in the case
of copies made at a distance from these centres of
scholarship and accuracy. Hence if we find great
divergences among the copies of the New Testament
as far back as the extant evidence takes us, we need
not be surprised ; nor need we be discouraged, since it
will be found that the divergences do not seriously
affect the substance of the record which these books
enshrine.

After describing, as has been done above, the
character and appearance of papyrus manuscripts at the

28 CRITICISM OFONEW, TESTAMENT) \0 cu:

time when the autographs of the New Testament were
written, little need be said of the rest of the papyrus
period. During the second century, the general
characteristics of the writing in current use remained
the same. With the assistance of a very large number
of dated documents (several to each year) it is possible
to trace the development of the non-literary hand with
some minuteness ; but for the literary hand the extant
specimens are too few to enable nice distinctions to be
drawn, and the first two centuries of our era must be
erouped together as a single palaeographical period,
within which a precise accuracy of arrangement is not
to be expected until further evidence is available. In
the third century two changes are observable. The writ-
ing, which hitherto has usually been upright, now fre-
quently develops a sloping formation; and experiments
are made in the use of the codex or modern book-form
of volume, instead of the roll. The former is a point
of merely palaeographical interest, useful chiefly as a
means of determining the dates of manuscripts ; but the
latter paves the way for the revolution of the fourth
century, when papyrus was superseded by vellum, and
when complete copies of the New Testament for the
first time became possible.

The history of the codex or book-form goes back to
the wax tablets which were commonly employed as
note-books by the Greeks and Romans from the earliest
days of their habitual use of writing. These tablets
consisted of thin rectangular plates of wood, with raised
rims to retain and protect the wax, in general appear-
ance much like a school-boy’s slate; and a number of
them could be joined together by passing strings or
thongs through holes bored in the rims on one side.
Several sets of tablets with such holes in their rims
exist to this day ; and these were the earliest codices.

11 AUTOGRAPHS OF NEW TESTAMENT 39

Next, from about the first century B.C., vellum was used
in the same way, but as yet only for notes and rough
drafts, not as a rival to papyrus for books intended
for circulation. Then, the advantages of this method
of book-formation becoming evident, an attempt was
made to use papyrus in the same manner; and from
the third and following centuries several examples of
papyrus codices have come down to us. In some cases
(and these may be supposed to be the earliest) the
writing is only on one side of the leaf,’ as though the
scribe was too much accustomed to the older manner
of writing upon papyrus to think of using both sides of
it; but normally it is on both sides, as in an ordinary
vellum codex. Usually the leaves are narrow in pro-
portion to their height,” which indicates that the
papyrus of which they are composed is not of the best
quality, so that they were probably cheap and inferior
copies.2 But as Christians in the early days of the
Church were rarely rich, it is likely that many copies
of the Scriptures existed in this form; and the extant
evidence points in this direction. Nearly all the
Christian papyri of the third century are codices,
while for non-Christian writings the roll form is still
predominant. Examples may be seen in the earliest
portions of the New ‘Testament now extant in
manuscript, the Oxyrhynchus fragments of Genesis and
of St. Matthew and St. John, mentioned above. These,
like the fragments of a collection of sayings of our Lord
found in the same place, are probably parts of pocket
volumes, written for private use rather than general
circulation ; and no doubt many such existed among

1 #.g. Brit. Mus. Pap. 126 (containing part of Homer, //. ii.—iv.).

2 E.g. the Homer papyrus mentioned in the preceding note, Brit. Mus.
Pap. 46 (a papyrus containing magical formulae), and a similar magical
papyrus in the Bibliotheque Nationale at Paris; and the same may be said
of the Oxyrhynchus ‘‘ Sayings”’ and Thucydides.

3 Thus the above-mentioned Homer is the worst of all the papyri.

δ, CRITICISM OF (NEW) TES TARE NT Υ ox:

the humble circles in which Christianity at this time
chiefly flourished. Valuable as such copies may be,
on account of their age, we cannot look to them with
any confidence for purity of text.

The papyrus period, then, may be summarily
characterised as the period when the textual problems
came into being, which we have to try to solve with
the help of the evidence afforded by the later periods.
During that period the Christian Scriptures were
scattered over the face of the accessible world, wherever
the Roman Empire spread, or wherever, beyond the
boundaries of that empire, Christian missionaries were
able to make their way. But as yet they spread and
were reproduced without the aid, without the sanction,
and often in defiance of the express mandates, of the
civil power. At certain times and in certain places
Christianity was allowed to do its work unchecked, and
was able to command the services of competent scribes
to multiply its sacred books; but oftener ἐξ was dis-
couraged, or was followed only by the humble and
poor, while from time to time it was the object of direct
persecution, in which its books were sought out and
destroyed by order of the Government. The full history
of the New Testament text during this confused period
can never be known; but it is the function of textual
criticism Ὁ trace it as far as possible);(\Uip) to" the
present time, no evidence worth mentioning is extant
which comes from within this period itself. We can
only see the results at the end of the period, and try
to work back to the causes. It is far from impossible,
or even improbable, that Egypt, which has given us
sO many precious manuscripts of early date, may yet
bring to light a Gospel or an Epistle written in the
second or third century. Such a discovery would be
full of interest, and might possibly go far towards

wy AUTOGRAPHS OF NEW TESTAMENT ἃ

settling some outstanding controversies ; but it would
have to be received with caution, and its character would
need careful examination. As has been shown above,
it might contain a text inferior in quality to that of
some existing manuscripts. It would be in the broad
characteristics of its text, rather than in its precise
details, that its value would be most likely to consist ;
and it might raise as many problems as it laid. In
any case, the best preparation for dealing judiciously
with such new testimony is a sound knowledge of the
evidence already in existence ; and it is with the state-
ment and examination of this evidence that the follow-
ing chapters will be concerned.

By way of conclusion, however, to the present
chapter, it may be useful to give a list of the extant
fragments of the New Testament on papyrus. Until
quite recently most of them had not found a place in
the standard lists of New Testament manuscripts, and a
provisional catalogue of them was given for the first
time in the first edition of the present work; but in
Gregory’s latest enumeration (described at greater length
in the next chapter) they receive official symbols in a
category by themselves, which it may be hoped will
be generally accepted. The following list accordingly
follows Gregory’s numbers.

P. [Soden ε or]! Matthew i. t-9, 12, 14-20. One
leaf ofa book. Third century. Found at Oxyrhynchus
in Egypt in 1896, and published by Messrs. Grenfell
and Hunt in Oxyrhynchus Papyri, Part I. (1898).
The variants of this fragment are of small importance
in themselves, but so far as they go they tend to
support the oldest vellum uncials, the Codex Vaticanus
and the Codex Sinaiticus. It was the property of the

’ For the system of enumeration employed by von Soden, here appended
to Gregory’s nomenclature, see below, p. 52.

42) ) CRITICISM (OFS NEW) TESTE INT) Miron

Egypt Exploration Fund, but has now been presented
to the University of Pennsylvania.

P~2 John xii. 12-15 in Greek on the verso, with
Luke vii. 18 ff in Sahidic onthe veczo.,\) Fifth or.sixth
century, in book-form, in the Museo Archeologico at
Florence. Edited by E. Pistelli (Papirz evangelici,
Florence, 1906).

p2 Luke vii. 36-43, x. 38-42; in the Rainer
collection at Vienna. Sixth century, in book-form.
Described by Wessely (Liihver durch dite Ausstellung
der Papyrus Erzherzog Rainer, no. 539).

Pt [Sod. « 34]. Luke i. 74-80, v. 3-8 (both these
portions very fragmentary), and v. 30-vi. 4. Discovered
in Egypt, attached to a MS. of Philo; now in the
Bibliotheque Nationale at Paris. Fourth century, in
book-form. Edited by Scheil (7émozres de la Mission
archéologique francaise au Caire, tom. 9, 1893).

P5 [Sod. € 02]. John i. 23-31, 33-41, xx. II-17,
19-25. Nearly the outermost sheet of a single quire
of some twenty-five sheets, containing the whole Gospel
of St. John; a quite unparalleled form of book.
Third century. Found at Oxyrhynchus by Grenfell and
Hunt, and published by them in Oxyrhynchus Papyrz,
Part II. (1899), no. 208; now Brit. Mus. Pap. 732.
Its text agrees generally with that of the Codex
Sinaiticus.

ps John xi. 45, in the University Library at Strass-
burg. Published by Gregory, without estimate of its
date, in Dze griechischen Handschriften des N.T., p. 46.

p. [Sod. ε 11]. Luke iv. 1, 2, in the Archaeo-
logical Museum at Kieff. Mentioned by Gregory,
zbed.

P2 (Sod. a 8]. Acts iv. 31-37, v. 2-9, vi. 1-6, 8-15,
in the Berlin Museum (P. 8683). Fourth century.
Mentioned by Gregory, zdzd.

1 AUTOGRAPHS OF NEW TESTAMENT 43

De it Poh, ty.) 11:13.) TSsb7.)) Fourth) of \fith
century, in book-form. Found and published by
Grenfell and Hunt, O2zyrhynchus Papyri, Part III.
(1903), no. 402 ; now in Harvard University Library.

P.° [Sod. a 1032]. Romans i. 1-7. Early fourth
century ; written in a rough uncial hand, apparently
as a schoolboy’s exercise. Found and published by
Grenfell and Hunt, of. cet. Part II. no. 209; now in
Harvard University Library.

Poe) DOr, 1 ἡ χ-:20., Vi. 3-08, Vile 5. 4, τοῦτ,
In the Imperial Library at St. Petersburg (formerly
in the Uspensky collection). Fifth century. The first
papyrus fragment of the New Testament to be brought
to light, having been brought by Bishop Porphyry
Uspensky from the East, and read by Tischendorf.
See Gregory, Tertkritik des Neuen Testamentes, p. 119.

P.2 [Sod. a 103 3}... Hebrews. 1... τως. Io the Amherst
Library. Third or fourth century, written in the
margin of a letter from a Roman Christian. Published
by Grenfell and Hunt, Amherst Papyrz, Part I. (1900),
moO. 2. ἢ

D8 Hebrews ii, 14—v. 5, x: 8—xi. 13, xi, 28—xii
17. Late third or fourth century ; written on the back
of a roll, on the vecto of which is an epitome of Livy.
Found and published by Grenfell and Hunt, Oxyrhynchus
Papyri, Part IV. (1904), no. 657; facsimile of two
columns in New Palaeographical Society, Pl. 47. Now
Brit. Mus. Pap. 1532. The longest New Testament
papyrus at present known, and important as including
parts of the Epistle to the Hebrews which are wanting
in the Codex Vaticanus, to which the papyrus is akin
in textual character.

ps ποῦς. L036]. 1 Corn 25-27,; 1: 6-8, 111. 8:16]
20. In the monastery of St. Catherine on Mt. Sinai.
Fifth century, in book-form. Discovered and edited

44 CRITICISM OF NEW TESTAMENT cu. ἢ

by Rendel Harris, Bzblical Fragments trom Mt. Sinai
(1896), no. 14.

ADP oriGor wat 180i, 4) Phe Gam ie Bee
Two leaves from a papyrus book, discovered at
Oxyrhynchus, Fourth century. Published by Hunt,
Oxyrhynchus Papyri, Part VII. (1910), nos. 1008, 1009.

Pp.° Rom. xii. 3-8. Purchased in Egypt for
Lord Crawford, now in the John Rylands Library,
Manchester. Probably an extract only, the verso being
blank. Late sixth or seventh century. Published by
Hunt, Rylands Papyri (1911), no. 4.

ee Tite ac LIE 5. 1 358.00) Part or a leat: froma
papyrus book, purchased in Egypt for Lord Crawford,
now in the John Rylands Library, Manchester. Third
century. Published by Hunt, of. cz¢. no. 5.

p18 Heb. ix. 12-19. Part of a leaf from a
papyrus book, discovered at Oxyrhynchus. Fourth
century. Published by Hunt, Oxyrhynchus Papyrt,
Parte Vill! (ror πο 107.6:

PP. Rev. i. 4-7. Written on the verso of a roll
containing a copy of Exodus; found at Oxyrhynchus.
Late third or fourth century. Published by Hunt, of.
cit; NO. 1079.

CHAPTER gil

THE UNCIAL MANUSCRIPTS

[Authorities: Gregory, Prolegomena to Tischendorf’s Movum Testa-
mentum Graece (Leipzig, 1884-1894), Zextkrztizk des Neuen Testa-
mentes (Leipzig, 1900-1909) and Die griechischen Handschriften des
N.7. (Leipzig, 1908); Scrivener, A Platn Introduction to the
Criticism of the New Testament (4th ed., by E. Miller, London,
1894); H. von Soden, Dze Schriften des N.7., vol. i. parts i.-iv.
(Berlin, 1902-1910); Thompson, of. cz¢. ; Tischendorf, Wotctia editionis
codicts Bibliorum Sinazttict (1860), and other works; Westcott and
Hort, Zhe New Testament in Greek, 1881; Nestle, /xtroduction to
the Textual Criticism of the Greek N.T., Eng. Tr. from 2nd ed.
(London, 1901); Lake, Ze Zext of the New Testament (London,
1902) ; and special works on the individual MSS., mentioned in the
course of the chapter. ]

IN the early years of the fourth century we reach an
epoch of the greatest importance in the history of the
New Testament Scriptures. Several causes combined
to alter completely the circumstances under which they
were copied and transmitted ; and it so happens that
it is from this very period that the earliest concrete
evidence, in the shape of actual substantial copies of
the Scriptures, has come down to our times. In the
first place, the conversion of the Emperor Constantine,
and the consequent recognition of Christianity as the
state religion of the Roman Empire, led to a great
multiplication of copies of the Scriptures, and also
enabled them to be made with due care and with all
the resources of ordinary literary production available

45

46) CRITICISM GF NEW TESTAMENT ) ice

at that date. Next, the adoption of vellum instead of
papyrus as the material employed for the best copies
of books rendered possible the combination of all the
sacred writings within the compass of a single volume ;
while at the same time the increased publicity given to
Christianity and the growth of doctrinal controversies
led to the precise definition and demarcation of the
canonical books which such a volume should contain.
It is true that this definition did not take place once
for all at any given moment or by any authoritative
decision, but was a gradual process spread over a con-
siderable time, as our earliest extant manuscripts testify
by their inclusion of sundry other books in addition
to those which we now recognise as canonical. The
process was not wholly completed before the end of the
century ; but a practical unanimity had been reached
in most quarters by that time, and the process is a
prominent feature in the history of the New Testament
during this period.

These causes acted and reacted on one another in a
manner which makes it difficult to say precisely which
is cause and which effect. Thus the adoption of
vellum, and therewith of the codex form in place of the
roll form, made it possible to unite all the Scriptures
in a single volume; while the needs of the churchmen
and of the lawyers alike demanded some form of book
in which their saored writings or their authorised codes
could be kept together without danger of loss or
separation. The formation of a canon of authoritative
Scripture was both necessitated by the development
of controversy, and itself in turn facilitated and pro-
moted such controversy. But whatever the exact play
and counter-play of cause and effect may have been,
the concrete phenomenon which marks the fourth
century in the textual history of the New Testament

ΠῚ THE UNCIAL MANUSCRIPTS 47

is the appearance of complete copies, either of the New
Testament or of the whole Greek Bible, written upon
vellum and arranged in the form of a codex.

The use of vellum, as has been briefly mentioned in
the last chapter, was not a new thing at this date.’
For some centuries it had been used for note-books or
for cheap copies of literary works; but it is only in the
fourth century that it ousts papyrus from the post of
honour as the recognised material for the best copies in
general circulation. For a time, no doubt, the two
materials overlapped, and there are examples of
papyrus books (but in codex form, not as rolls) as late
as the seventh century, or, in the case of Coptic works,
still later.2 But on the whole the victory of vellum
was won early in the fourth century ; and the proof of
it is found, not only in the decline of literary papyri,
both in number and in quality, at this point (though
non-literary papyri exist in immense numbers up to
the seventh or eighth century), and in the appearance of
vellum codices, but in two definite pieces of external
evidence. On the one hand there is the statement of
Eusebius (Vz¢. Constant. iv. 36) that the Emperor
Constantine about the year 331 ordered fifty copies of
the Scriptures om vellum for the churches in his new

1 See Palaeography of Greek Papyri, ch. v.; Thompson’s Handbook of
Greek and Latin Palacography, 35-37.

2 There are (besides many smaller fragments) a Coptic MS, containing
Deuteronomy, Jonah, and Acts, in the British Museum, and two Greek
collections of magical formulae, one in the British Museum and one in the
Bibliotheque Nationale, apparently of the fourth century; a Hesiod at
Vienna, of the fourth century ; an Aristophanes at Berlin, and two MSS. of
Menander, at Cairo and Geneva, of the fifth ; a Nonnus at Berlin, of the
seventh ; portions of a Psalter, and of the Ascension of Isaiah, in the Amherst
Library ; a Psalter (containing Ps. xi. 2—xix. 6, xxi. 14—xxxv. 6) in the British
Museum ; part of Zechariah and Malachi at Heidelberg ; part of Cyril of
Alexandria De Adoratione at Dublin and Paris; all of the seventh century ;
a large Coptic codex of the Psalms of the seventh century, and an equally
large one of homilies, perhaps rather later, both in the British Museum.

There are also said to be many late literary papyri in the Oxyrhynchus
collection, which have not yet been published.

48 CRITICISM GF NEW TESTAMENT | cx,

capital ; on the other, there is the statement of Jerome
that the (papyrus) volumes in the library of Pamphilus
at Caesarea were replaced by copies on vellum through
the efforts of Acacius and Euzoius (crc. 350).!

The change of material was accompanied by a
change in the style of writing. Vellum is, of course,
a far stronger substance than papyrus, and therefore
admitted a firmer style of writing, with thicker and
heavier strokes. Moreover, economy of space being
less necessary, now that the roll form had _ been
abandoned, larger letters could be employed; and the
general result is that early writing upon vellum is ona
larger and more handsome scale than most writing
upon papyrus. Further, the scribes who wrote it seem
to have cast back for their models to the best ages
of the papyrus hand. The papyrus manuscripts of the
third century are mostly written in a sloping hand
which is not very graceful in appearance; but the

vellum manuscripts of the fourth century recall rather .

the best hands of the first and second centuries, which
have a strong claim to be regarded as the finest
specimens of calligraphy in the whole papyrus period.
At the same time we are able wholly to put out of
consideration the non-literary styles of writing. During
the papyrus period we have abundant examples of the
non-literary hand of everyday use, side by side with
the literary hand which was reserved mainly for book-
production ; and in the case of the New Testament we
have seen reason to believe that the non-literary hand
played a considerable part in the preservation and
transmission of the sacred books. But with the rise of
vellum, and the simultaneous emergence of Christianity
into the position of an authorised and established

1 It may be added that the records of the persecution of A.D. 303 in
Africa mention both rolls (22472) and codices (but especially the latter) among
the books sought out and destroyed (Routh, Re//. Sacr. iv. 289 ff. and 322 ff. ).

ΠῚ THE (UNCIAL’ MANUSCRIPTS 49

religion, the spheres of book-production and ordinary
writing became more sharply separated. No doubt
a non-literary hand continued to exist, since it was
required for the common affairs of daily life; indeed
up to the first quarter of the eighth century we have
plentiful evidence of its existence upon papyrus ; but it
ceased to affect the channels by which the text of the
New Testament was handed down. Thenceforward,
through all the centuries until the invention of printing,
we have only to do with the formal book-hand in the
various modifications which it assumed in the course of
successive generations.

Writing upon vellum, as has been briefly indicated
above (p. 19), falls into two main periods. In the first
of these the book-hand consists of capital or uncial?
letters, formed singly and separately, without ligatures,
and usually of a considerable size. Examples may be
seen in Plates III.-VIII. of the present volume. The
earliest of them, in the fourth century, show evident
traces of their descent from papyri. The writing is
little, if at all, larger than that of the best papyri; it is
arranged in narrow columns, three or four to the page ;
and it is wholly devoid of ornamentation. There are
no accents or capital letters. The most that is done
in the way of making a new paragraph is to begin

1 The term is derived from a passage in Jerome’s preface to his Latin
translation of the book of Job, where he inveighs against the extravagant
style in which many books were written in his day, ‘‘ uncialibus, ut vulgo
aiunt, litteris.”. The word apparently means ‘‘inch-long” letters, but it
occurs nowhere else, and it has sometimes been supposed to be merely
a misreading of z7z¢za/¢bus, which closely resembles it and is actually found
in many copies of the passage in question. But no one would be likely to
use the term “initial” to describe a particular type of writing, nor is
the word itself so unfamiliar as to need the apologetic ‘‘ut vulgo aiunt,”
which, on the other hand, is quite in place as a qualification of the exagger-
ated phrase ‘‘inch-long.” The characters of some early MSS. are quite
large enough to justify the phrase as a pardonable exaggeration; e.g.
in the New Testament MS. known as N (see below) the letters measure
about five-eighths of an inch, with capitals nearly twice that size.

E

50. CRITICISM (OF NEW TESTARIEN fT) cr.

a new line and slightly draw back the first letter into
the margin, but without enlarging it. In the fifth
century the writing grows rather larger and the columns
broader, so that there are not more than two to a page,
and sometimes only one. Enlarged initials mark the
beginnings of paragraphs; or if, as often happens, the
new paragraph does not begin a new line, but follows
on ἢ ‘the same line as the end: ‘of the previous
paragraph, with only a short blank space to mark the
break, the enlarged letter is the first in the first full
line of the new paragraph. In the sixth century the
writing is generally heavier and thicker, and often
larger ; while in the seventh century, instead of being
upright, as hitherto, it shows a tendency to lean to the
right, the letters being at the same time compressed
into more oval and pointed shapes. This tendency
increases through the eighth and ninth centuries,
culminating in the heavy and angular forms of what
is known as the Slavonic hand, which (with certain
exceptions and modifications, and occasional reactionary
attempts to revive the earlier upright and rounded
style) predominates throughout that period; until in
the tenth century a new form of writing, which had
arisen in the previous century, finally prevailed over
this large and cumbrous style, and the uncial period of
Greek palaeography comes to an end,’

The uncial period, then, covers a space of some six
centuries; and it,.is to. this. period) that the most
valuable of the extant manuscripts of the New Testa-
ment belong. We come now to the consideration and
description of these in detail; but before commencing
this task, it will be convenient to explain the system
which regulates the numeration and nomenclature of

1 For a fuller sketch of the development of Greek uncial writing, see
Thompson, Greek and Latin Palaeography, 149-158.

ΠῚ THE UNCIAL MANUSCRIPTS 51

New Testament manuscripts. The method in general
use hitherto goes back to the edition of J. J. Wetstein
in 1751-52. He first conceived the scheme of denot-
ing the uncial MSS. by the letters of the Roman
alphabet, and the minuscules by numbers. Since,
however, very few MSS. contain the whole New Testa-
ment, it was considered convenient to classify the
books in four groups, each having its own numera-
tion of manuscripts. These groups are (1) the four
Gospels, (2) the Acts and Catholic Epistles (which
generally go together), (3) the Pauline Epistles, (4)
the Apocalypse. For about a century after its adoption,
this system answered its purpose without difficulty ;
and it reached its full development in the two great
catalogues of New Testament MSS. which were re-
cognised as the standard authorities in the latter part of
the nineteenth century, those of Scrivener (/utroduction
to the Criticism of the New Testament, 1st ed. 1861,
4th ed. 1894) and Gregory (Prolegomena to the eighth
edition of Tischendorf’s Novum Testamentum Graece,
1884-1894; repeated and continued in his Zeatkritzk).
In these lists, uncial manuscripts are indicated by the
capital letters, first of the Latin alphabet from A to Z,
next of the Greek alphabet (so far as that differs from
the Latin) from [ to Q, and finally of the Hebrew
alphabet, the resources of which have not yet been
exhausted. Since, however, each group has its own
numeration, a letter which denotes a certain manuscript
in one group will often denote a different manuscript in
another group. Thus, the great Codex Vaticanus,
which is known as B in the Gospels, Acts, and Pauline
Epistles, does not contain the Apocalypse; hence in
the case of that book B denotes a different MS.
Similarly D of the Gospels and Acts (Codex Bezae)
does not contain the Pauline Epistles, and the letter

52 CRITICISM (OF NEW TESTAMENT © cx.

is consequently given there to another MS. (Codex
Claromontanus). In order to avoid confusion, it is
usual to distinguish these second holders of a title by a
subsidiary mark, either B,, D,, or B??% DP") and the
like. Minuscule manuscripts (2.4. manuscripts written
in the smaller kind of hand which succeeded the
uncial type) are indicated by numbers, each group of
books having, as before, a separate numeration, the
complications of which will be explained in the next
chapter.

The drawbacks of this system, in face of the
constantly increasing stream of discoveries in the latter
half of the nineteenth century, are obvious, The letters
of the three alphabets could only be made to suffice by
grouping a large number of fragments under a single
letter (notably O, T, W, ©); and the unfamiliarity of
the Hebrew alphabet made it ill suited for the purpose
of common use by New Testament students. Further,
the practice of assigning the same letter to different
MSS. in different groups was found to be confusing,
and the methods of distinguishing them cumbrous.
Accordingly, H. von Soden, in the great edition of the
New Testament, of which the prolegomena have taken
eight years to print, while the text and apparatus are
still to come, devised a wholly new system of numeration.
All MSS., whether uncial or minuscule, are included in
a single list, or rather group of lists. Each manuscript
is indicated by a number, to which is prefixed the
letter ὃ, ε, or a (for διαθήκη, εὐαγγέλιον, OY ἀπόστολος)
according as the MS. in question contains the whole
New Testament, the Gospels, or the Acts and Epistles,
with or without the Apocalypse. The numbers, how-
ever, are not assigned simply in a regular succession ;
an attempt is made to indicate by them the approximate
date of each manuscript. The method is as follows.

τὴ THE UNCIAL MANUSCRIPTS 53

Manuscripts of the Gospels, earlier than the tenth
century, are denoted by the numbers from I to 99;
those of the same period containing the entire New
Testament or the Acts and Epistles, by the numbers
1 to 49; the proper letter being of course prefixed in
each case. If these numbers should not suffice, the
same figures are to be used again with a o prefixed.
In the tenth century Gospels MSS. are indicated by
the numbers from 1000 to 1099, those of the other
groups by the numbers 50 to 99. Manuscripts of the
eleventh century are denoted by the numbers from 100
to 199, followed by those from 1100 to 1199; those
of the twelfth century by the numbers from 200 to
299, followed by 1200 to 1299; those of the thirteenth
century, 300 to 399 and 1300 to 1399; those of the
fourteenth century, 400 to 499 and 1400 to 1499;
those of the fifteenth century, 500 to 599 and 1500
to 1599. If even these figures are not sufficient (as
happens after the eleventh century) the numeration
of the twelfth century MSS. continues from 2000, the
thirteenth from 3000, and so on.

Further complications are introduced to indicate
whether MSS. of the ὃ or α categories include the
Apocalypse or not. In each hundred, the first half
(ze. the numbers from 100 to 149, 200 to 249, etc.)
indicate MSS. containing the Apocalypse, the second
half MSS. which omit it. Finally, MSS. containing
only the Acts and Catholic Epistles are denoted by the
numbers 1000 to ΙΟΤΟ (tenth century or earlier), 1100
to 1119, and so on, MSS. containing only the Pauline
Epistles by the numbers 1020 to 1069, I100 to
1169, etc., and MSS. containing only the Apocalypse
by the numbers 1070 to 1099, 1170 to 1199, etc.
Manuscripts containing commentaries in addition to
the text have a separate numeration, with a letter

54 CRITICISM OF NEW TESTAMENT occu.

or abbreviation prefixed to indicate the particular
commentary included.

To give some instances of the working of this
system, the student who has succeeded in mastering
it will know, simply from the numeration employed,
that (6 1,means/a MS. of ‘the N,T. earlier, than A.D.
900% εἰ 1015, .a.MS. of the Gospels of the tenth
century ;e 215 or € 1115,.a\ ΜΙ ΒΟΥ the’ Gospels of
the eleventh century ; a 235, a MS. of the ἀπόστολος
including the Apocalypse, of the twelfth century; a
375, a MS. of the ἀπόστολος omitting the Apocalypse,
of the thirteenth century ; a 1416,a MS. of the Acts
and Catholic Epistles, of the fourteenth century ; a 1540,
a MS. of the Pauline Epistles, of the fifteenth century ;
and a 1680, a MS. of the Apocalypse, of the sixteenth
century.

The ingenuity of von Soden’s system is obvious,
and a student who works with it habitually would, no
doubt, learn eventually to use and interpret it without
difficulty. On the other hand, a student who only
wished to refer intermittently to a textual apparatus,
or one who merely wished to look up the description of
some particular MS., would find it intolerably com-
plicated. It is bewildering to look at von Soden’s
catalogue, and find, for instance, ε 1075 preceding
e 107, € 1166 before e 257,and soon. Such difficulties
and complications would only be justifiable if the
benefits of the new system were great and certain ; and
it is to be feared that this is not the case. The dating
of Greek minuscules is far from being a certain science,
so that the information which this nomenclature pur-
ports to give is not to be depended on. Nor is the
information itself of great value. The period during
which the date of a MS. is important is the first
millennium; yet all the MSS. prior to A.D. 900

nt THE UNCIAL MANUSCRIPTS a

are lumped together indiscriminately. After that date,
the exact age of a manuscript is of quite minor
importance for the textual criticism of the New
Testament. It is important to know that a manuscript
belongs to the fifth or sixth century rather than the
eighth or ninth; but it matters little whether it is of
the tenth or the thirteenth.

Another objection to von Soden’s system is the
confusion which it introduces into the study of textual
criticism by its rejection of the nomenclature to which
scholars are accustomed. Not only is it difficult for
the student brought up on Gregory and Scrivener to
recognise well-known MSS. under a new symbol (to
substitute 6 2 for x, 6 4 for A, or e 19 for N), but it
will be a serious drawback to students brought up
on von Soden to find that the great works of the past
on textual criticism, such as those of Westcott and
Hort or Sanday, are almost unintelligible on account of
the change of symbols. To change a well-established
system is an evil, only to be justified if the gain is sure
and great ; and that, as we have seen, is not the case.

While, however, the adoption of von Soden’s method
of numeration, in what promises to be an edition of
the first importance for textual criticism, is nothing
less than a disaster, the fact remains to be faced that
the system initiated by Wetstein is breaking down by
reason of its inability to deal satisfactorily with the
constantly increasing number of uncial MSS. and
fragments of MSS. To meet this difficulty, Gregory,
after prolonged consultation with a large number of his
fellow-students in all parts of the world, has devised
a modification of the Wetstein system, which will give
the required elasticity with the least possible disturbance
of the traditional nomenclature. The principles of his
revised scheme are as follows :—

56° (CRITICISM OF NEW TESTAMENT | cu:

1. Papyri are denoted by an “antique” P (P)
followed by a number (a method already adopted in
the Oxford text of Homer).

2. Roman and Greek characters are retained for
uncials, but Hebrew characters are abandoned, except
in the case of x. With the exception of a few well-
known MSS., such as DP) Kacts HPals and the like,
one letter is not assigned to more than one MS.,’ and
the long series grouped under the letters O, T, W, ©
are abolished. Sufficient symbols are thus furnished
for forty-five uncial MSS., including all those at present
known which are of much importance, all of which
retain their familiar symbols. Other uncial MSS. are
denoted by numbers, printed in thick type (technically
known as “clarendon ” type), with 0 prefixed, beginning
with 046; the numbers 01 to 045 being assigned
as alternative designations to the forty-five MSS. which
are distinguished by letters.

3. Minuscule MSS. are indicated, as before, by
numbers in ordinary type; but the four series (for
Gospels, Acts, Pauline Epistles, and Apocalypse) are
thrown into one, so that each MS., however many of
these groups it contains, only has a single number.
The method adopted is that each MS. which includes
the Gospels retains throughout the number which it
had in the Gospels series, while those of the other
groups follow on at the end (or fill accidental gaps)
with new numbers.

This system, though in one respect it sacrifices
symmetry to practical utility, is simple and workable,
and involves the least possible breach with tradition.
It is free from the elaboration of von Soden’s system,
which breaks down through attempting too much.

1 Only eight letters, Ὁ E F GH K L P, have more than one connota-
tion. See the list which follows in the present chapter.

ΠῚ tne UNCIAL MANUSCRIPTS 57

After all, a scheme of numeration is only intended
to provide a handy symbol for each MS., not to take
the: ‘place of a’ description’ of it)’ Di Gregory’ has
secured the assent of an overwhelming majority of
New Testament scholars to his system, and it is much
to be hoped that it will be generally adopted. It has
already been taken into use in the edition which is
at present the handiest in existence for working pur-
poses, namely, the new Oxford text with select critical
apparatus, edited by Professor Souter. It will accord-
ingly be taken as the basis of the following catalogue,
though references will also be given to the earlier
notations of Gregory and Scrivener (where they differ
from this) and to that of von Soden.

According to the latest catalogue of New Testament
manuscripts, the total number of uncials at present
known to exist is 168." The large majority of these,
however, consists of fragments, sometimes extremely
small. Of the 168 manuscripts above mentioned, fifty-
seven may be reckoned as containing some substantial
part of the New Testament, while the remainder are
mere fragments. Only one uncial manuscript (x, the
Codex Sinaiticus) contains the New Testament com-
plete. Four more (ABCYW) originally contained
the whole New Testament, and still contain most
of it, but have suffered some mutilation. Nine
(RBKMSUWOQ, 01414) have complete copies of the four
Gospels ; seven (RABP, 049, 056, 0142) of the Acts ;
nine (NABK,L,P, 049, 056, 0142) of the Catholic
Epistles ; seven (RAD,G,P, 056, 0142) of the Pauline

Epistles ; and four (NAP, 046) of the Apocalypse.”

1 Gregory, Die gr. Handschriften des N.T. (1908), supplemented by
Textkritik, part ili. pp. 1082-3, 1368-72, 1484 (1909). The total becomes 171
if two transcripts of D, and a leaf at Damascus (Soden ε 49), which Gregory
excludes, are reckoned.

2 Small mutilations, affecting only a few verses, have been ignored in this
computation.

58 CRITICISM: OF NEW TESTAMENT): ca:

With regard to the dates of these manuscripts, it is
only very rarely that the exact year in which they were
written is known ; and the few to which precise dates
can be affixed are among the latest. For the rest, their
dates must, as a rule, be fixed on palaeographical
evidence alone, with slight assistance in a few cases
from collateral testimony of various kinds. Hence there
is inevitably some divergence in the dates assigned
to these manuscripts, especially in the case of the small
fragments. The following estimate may, however, be
taken as approximately correct. Two great manuscripts,
the Codex Sinaiticus and the Codex Vaticanus, the
one complete and the other nearly so, may be assigned
with considerable confidence to the fourth century,
together with about six small fragments. The Codex
Alexandrinus and the Codex Ephraemi, both originally
complete Bibles, though the latter is now much mutilated,
and Mr. Freer’s Gospels MS. (W), belong to the fifth
century, together with about twenty-two fragments.
Thirty-five manuscripts (mostly small fragments) may
be assigned to the sixth century, twenty-five to the
seventh, twenty to the eighth, forty-three to the
ninth, and twelve to the tenth. These figures, though
not to be relied on as absolutely accurate in detail, will
at least serve to show approximately the chronological
distribution of the uncial evidence.

The examination and classification of the textual
evidence must be reserved until the several witnesses
have been individually described; but a provisional
distinction may be drawn which will be useful for the
purpose of this description. Speaking very generally,
it may be said that the manuscripts of the New Testa-
ment fall into two great classes,—those which support
what is known as the Textus Receptus, and those
which depart from it. The Textus Receptus is that

a i a i ar ne, αὐδουι μος

ee, ee

—— ΕἸ ρυ χα

ΠῚ ΤΗΝ UNCIAL ΜΑΝΌΘΒΘΟΝΙΡΥΘ 59

type of text which, having been adopted in the earliest
printed editions of the New Testament, has continued,
with only slight modifications, to hold its own as the
standard text in ordinary use. It is found in our
ordinary editions of the Greek Testament, and in an
English dress it is familiar to us in the Authorised
Version ; and it is supported by a vast numerical
majority of manuscripts. On the other hand, there is
a type of text which (especially in the Gospels) often
departs very markedly from the Textus Receptus.
The evidence for it is very various, being collected from
a few manuscripts, some of the versions, and some of
the Fathers; but the remarkable feature about it is
that it includes the earliest testimony in each of these
branches, and consequently its weight is far greater
than its merely numerical following would show. The
result is that this type of text has been adopted in most
of the modern critical editions of the Greek Testament,
and a form of it appears in the Revised Version of the
English Bible. It is a main purpose of the present
volume to estimate the value of these two great types
of text; but it would be premature to discuss the
question now. For the present, it will be best simply
to call the Textus Receptus type the a-text, and its
opponent the §-text; though it must be remembered
that within each of these classes (and especially in the
second) there are considerable divergences of detail.
The classification is of the broadest, and is merely
adopted as a rough basis for inquiry. In the nomen-
clature here adopted there is the advantage that the
chief representative of the a-text is (in the Gospels)
the manuscript known as A, while the great champion
of the B-text is the manuscript known as B; and
we avoid begging the question as to the relative merits
of the rival types. A subdivision of the §-type, of

60 CRITICISM OF ' NEW) FEST AREN® ἐπ!

which the principal Greek representative is the manu-
script known as D, may be entitled the 6-text. Its
most obvious characteristic is a very wide departure
from the a-text, with apparently arbitrary modifications
and not infrequent additions to it; the character of
these variants will form an important subject of inquiry
later on.

It now remains to consider the several uncial manu-
scripts in detail. Ihe more important ones will need
examination at some length, while for the less important
reference may be made to the standard catalogues
of manuscripts.

x, Codex Sinaiticus [Sod. ὃ 2] (Plate III.).—First
in the list (not strictly in alphabetical precedence, since
the Hebrew alphabet properly follows the English
and Greek, but in practical usage) stands this well-
known manuscript, notable alike for its intrinsic value
and for the circumstances attending its discovery. The
story is familiar, but cannot be passed over in any
history of the New Testament text.” In 1844 the
well-known German Biblical scholar, Constantin Tisch-
endorf, was travelling in the East under the patronage
of King Frederick Augustus of Saxony, in search of
manuscripts ; and in the course of his travels he visited
the monastery of St. Catherine on Mt. Sinai. There,
in a waste-paper basket containing a number of leaves
of various manuscripts, destined to light the monastery

1 Scrivener, i. 90-188, Gregory (Prolegomena to Tischendorf's Movum
Testamentum Graece, 8th ed.), 345-450, ZTextkritik, 18-123, 1018-1083,
1363-1372, 1484. The numbers according to von Soden’s system are
appended in brackets.

2 It is told several times by Tischendorf in the various publications to which
his discovery gave rise, viz. Notitia editionis codicis Bibliorum Sinaitict
(1860), Bibliorum Codex Sinaiticus Petropolitanus, 4 vols. (1862), Novum
Testamentum Sinaiticum (1863), Novum Testamentum Graecum ex Sinattico
codice (1865), Die Sinatbibel (1871); besides a few controversial articles. Ido
not see that any sufficient cause has been shown to question the truth of Tisch-
endorf's story or the good faith of his dealings in the matter, as has sometimes
been done (cf. Gregory, Textkritik, pp. 23-29).

PLATE 11].

at τ
ae LAT KON
Fal = Ἣ
Pee
ες
δον
τιν

_FRRLAT AN Rp eTAC rOKsoy>
SPAKRONCARRATSS £ ASN IOYAENE!?
SSERONTAMENITON τι ManeEsboROy NM
MN HM KAN καλὰ “Porn pb νπετροίωον
BS FOCVOY HAN OY a ayy pore Jatin
eRe ἸΞἈΡΞἈ ORI ΡΟ. Sey RUE Sie oe

[

ΤῸ
5

EXY PRO Ficatiasy : :
Ἀγ. ! τὰ
MIQON εἰ ΚΘ ῪΣ ξ : é

“Ὁ PSTOYMNHMeEp : εν :
Ms fovekhianenrrc} Oc BN Ee
ει SATOEED POY CANA. 2 = rhs

RON NGAMPOKOR
KAG FI RTC NO MIERS
eb TWOICAE SR Ce pt
BELA MO HORE fe
: πὸ πνεῖν KEINE RPE
᾿ δ ΩΝ RIOT?

¥

oS One NC PEIAYETAIC poo
CD BRAS KOMMEG CE. > :
THEO YN rer
WHET GC Ay +
POOR Lhe te ΓΕ
ΟΥ̓ Ope Crabs
CES Aa FM eee Pree
CL gary © CPE DDS
poe cae = ee a Ἀε τες
REAM IIE peer -. SS
TAA PETS Oh &®, τς 11
ΘΕ ΧΊΟΣ Ἢ Foye ᾿,
“λεὼ bree foe
POPOV ery > .
PRBS GRC TERS 8,
PEONSIAMARE COLA, = ὁ
MY VOPR ἢ κυ ττε :

oe

4
POS TAB YORE ae

τ, ἊΝ εἈχνώλος: PEER Hae
f oy “yee 4
eels eeN Oi CMI

Bei y Prt Ri 2.
ΡΟ ΚΙ Se eee ooo
VATA Fay rs
ΡΟ ὦ. fry

Scale 3 : 8.
of St. Mark being omitted.

“6 MAKEIKYARCMENS e PATEL
YO ΟΝ ΟΣ PAP a τοῦ ἴκῸςς
ΜΈΓ᾽ OAR , PIE NTA Re rd?
PeBAIEICEXKBOYENICl 9 τ΄
ππ JOMNHMGOneh= Ὁ

CODEX SINAITICUS.

The page shown contains Mark xvi. 2-Luke i. 18, the last twelve verses
This is one of the pages written by the corrector,
whom Tischendorf believed to be identical with the scribe of the Codex Vaticanus. )
To face p. 60.

TKN λον ROS

KR CHELATE PIO
1 ey tEX EVE CAN BH
“TAZACOAIAFEIFT CIA
PIE TIVO Ἢ ΤΊ ΓΙ.
poxpoprimeniarns
ENEULMIBIEIPAP MAR
RAOUDEOTISPCADC DH
ΡΤ ΜΟΙ ΡΥ
“TOLPPALRALY TIALS
‘TA ΓΕ ΟΕ ΘΟ ys
ROVOTEAOSE RAM
SAAPH oO MOrvarre ΤΙ
* ANWBOENPEALI NA
ἘΕΡΕΡΟΣ SAO SMES
COMPA ἈΠ ΡΈΕΙ:
αν ΥΡ Π ΟΣ
ΡΝ ΟΣ
PAST AES ERS ERED ως: Ἕ ke,
Me EP ERIC TIOGA
"FUE EAGT κυνῶν

Ps
μὲς PE RCKSIR,
VD AEE GPA OP Tee ov

MERE Pes PESTER DR

Rights BE. ἐπ
ee ἔτ ΦΑΑΤΊ
riches ζω: Ἐν
Ἔν Fiber Pion
"ELEY EXY Teh Gs

PROG TACALE PRI
ERP POARICKLIAIKA
CLUTCH PP pO ey >,
MEMIN POLK MOY
αὐ ΟΥ̓ Seno ΕΣ μλρ᾽
: Ae peat cad
ἘΠΊ» PRIA
“τοτοϊ γοκεμρίνον
PECOM RICH MEP:
AS UCD PIICRE
CPECREVURECEM Tate
PREC pe rN ey sorte
Pepe geryiiced
BAD PERRI TY ees

Pelee τε ἜΤ
γι τ ἘΣ 2 τ Ἐν;
DEOL p Ee lig

CALCIO ORLEACD MEK
“LON BS AOR TOM hoy
PA PIANE POE LADO

ΕἸΜΙ OYAROY FI Por 4“
OMEN ONES OV Trhe

fA POY Oy ri pAanie
CEDOHMAG RY Thi 1"

AGE ROY ἐς POD CL b>

SOM po yay op Aele
ΠΟΥ Forpery Mi Amr
Ἃς VANCES ΘῈ δ
fe ARPIAC TA Di rs

SIZ OCG ENMECESS
ἐστὶ

Cir hier hes Pg ie oa
BS Se tate ee a Ἐ dF ὸ
ἜΑΡ Στοῦ

Boye erect
GAIL PORES POPS Li
Oor αἴσιος;
is oasis an 1.
ΕΣ ΕΣ AEP AY

ep ee | eae 2

® Pout εἴ j
tt Sa | EEN SEE,
ΣΕΥ ῬΑ: +

PUPAE

eto yee BPr bey eee
HICUSY tact $y rs
HOMER OVE. εν
BAVTT ENDL καὶ τ) δε τον
CPE
ΠΕ ΟΣ ΚΕ atta
er bear Ὁ

SAS SPS AY tic Ft
ς fale ee oe

Fourth Century.

+

Ct eee

3
5
;
3
2
4

mR

ΠῚ THE UNCIAL MANUSCRIPTS 61

fires, he chanced to notice several leaves of vellum
bearing Greek writing of an extremely early type,
which on examination proved to be part of a copy of
the Septuagint version of the Old Testament. No
objection was offered to the visitor’s appropriating the
supposed waste paper, and he was informed that much
similar material had already met with the fate for
which these sheets were intended. He was also shown
further portions of the same manuscript, containing the
books of Isaiah and the Maccabees; but these he was
not allowed to appropriate, and he was obliged to con-
tent himself with warning the monks that such things
were too valuable to be used as fuel. Returning to
Europe with his spoil, which consisted of forty-three
leaves from the books of 1 Chronicles, Jeremiah,
Nehemiah, and Esther, he deposited it in the University
library at Leipzig, where, under the name of the Codex
Friderico-Augustanus (from King Frederick Augustus,
the discoverer’s sovereign and patron) it still remains ;
and in 1846 he published its contents. In 1853 he
returned to the East, and revisited the monastery of
St. Catherine, hoping to acquire the rest of the manu-
script ; but this time he could neither see nor hear
anything of it, though (as subsequently appeared) it
had meanwhile been shown to the learned Russian
bishop, Porphyry Uspensky, and to an English officer,
Major Macdonald. Believing that the treasure must
have already found its way into Europe, Tischendorf
_resigned his search and abandoned his hopes; but in
1859 his work took him back to Mount Sinai, this time
under the patronage of the Tsar Alexander I]. His
stay was only of a few days’ duration, but shortly
before its end he happened to be conversing with the
steward of the monastery on the subject of the Septua-
gint, copies of his recent edition of which he had

62 CRITICISM OF NEW TESTAMENT ¥) cw

brought with him to present to the monastery. The
steward observed that he too had a copy of the Septua-
gint, and presently produced it, wrapped in a napkin,
the only protection of its loose and often mutilated
leaves; and there the astonished visitor beheld the
very manuscript which for the last fifteen years had
been so much in his thoughts. The prize was even
greater than he had hoped; for not only was much of
the Old Testament there, but also the New Testament
intact and in excellent condition, with the addition of
the Epistle of Barnabas and much of the Shepherd
of Hermas—two early Christian works which hovered
for a time on the edge of the Canon before being ulti-
mately excluded from it. Tischendorf’s delight may
be imagined, and could not be concealed. He obtained
leave to take the precious manuscript to his room ;
and that night, thinking it sacrilege to sleep, he spent
in transcribing the Epistle of Barnabas, of which no
copy in Greek was previously known to exist.

After discovery, diplomacy. Without much diff-
culty, apparently, Tischendorf persuaded the monks to
let him have the manuscript sent to Cairo, where a
first transcript was made; but some time elapsed
before he could induce them to part altogether with
their property. Ultimately, however, he prevailed upon
them to present it to the Tsar, the great protector of
the Greek Church and his own immediate patron; and
in October 1859 he had the satisfaction of carrying
his treasure to St. Petersburg, where, in the Imperial
Library, it has since found its permanent home. As
an acknowledgment of the gift, the Imperial Govern-
ment made a present of 7000 roubles to the convent
of Mount Sinai and 2000 to that of Mount Tabor. A
preliminary account of the MS. was issued by Tischen-
dorf in 1860, and at the end of 1862 the complete

ΠῚ TEE ONCIAL’ MANUSCRIPTS 63

MS. was published in facsimile type.” A critical
edition of the New Testament for general use followed
in May 1863. Collations of the MS. have been pub-
lished more than once; but its definitive publication
(so far as the New Testament portion is concerned)
was delayed until 1911, when the Oxford University
Press published a complete photographic facsimile from
negatives taken by Professor and Mrs. K. Lake, with
an introduction by the former. By this admirable
publication (to be completed, it may be hoped, by the
addition of the Old Testament portion) the series of
facsimile editions of the great New Testament codices
is satisfactorily rounded off.

The Codex Sinaiticus, when complete, evidently
contained the whole of the Greek Bible; but much of
the Old Testament had disappeared before Tischendorf’s
discovery of it. Part may have perished only shortly
before his first visit to Sinai, through the fate from
which he rescued the leaves which now form the Codex
Friderico-Augustanus at Leipzig; but the original
mutilation was of much earlier date, since fragments
were found to have been used in the bindings of other
MSS? The details of the Old Testament portion of
the MS., however, do not concern us here ; they may be
found in Professor Swete’s /utroduction to the Old Testa-
ment in Greek (pp. 129-131). The New Testament is
intact, and includes in addition the non-canonical books
known as the Epistle of Barnabas and the Shepherd of
Hermas, the latter being incomplete. It is written on
very thin vellum of excellent quality, prepared from the
skins of antelopes (as Tischendorf suggested) or of
some animal of similar size, arranged (as a rule) in

1 Some specimens of the edition were sent to the Exhibition of 1862 in
London, several months before the appearance of the complete work.

2 These were discovered by Bishop Porphyry on his visit to Sinai in
1845.

64 CRITICISM Of NEW TESTAMENT | ca:

quaternions of four sheets (=eight leaves or sixteen
pages). Each page measures 15 inches by 134, and
there are signs that they were originally larger, but
have suffered from the binders’ shears. Each page
contains four narrow columns of writing,’ about 24
inches in width, and consisting of 48 lines. There are
no enlarged initials, but the first letter of a paragraph
projects slightly into the left-hand margin. There are
no accents, and punctuation by the first hand is rare,
but the text is divided into short paragraphs, and the
remainder of a line in which a paragraph ends is left
blank. The writing is a rather large uncial, larger than
is found in any but a very few papyri, but not so large
as uncials subsequently became on vellum.” In width
of column and in shape of writing the Codex Sinaiticus
recalls the best literary papyri of the first and second
centuries; and the resemblance is increased when,
instead of a single page, the open codex itself is seen,
since the two pages shown side by side, with their
eight narrow columns, present very much the appear-
ance of a papyrus roll.

According to Tischendorf, the original text of the
MS. was written by four different scribes, one of whom
(in addition to part of the Old Testament) wrote the
whole of the New Testament except seven leaves*® and
the “Shepherd.” The writer to whom the seven leaves
were due, besides writing the books of Tobit and Judith
in the Old Testament, also seems to have corrected the

1 Except in the poetical books of the Old Testament, where there are
only two columns to the page.

2 The complete facsimile gives the impression of a rather thicker and
heavier hand than the photographs previously accessible had led one to
erie leaves in Matthew, the last leaf of Mark and the first of Luke, a
leaf from 1 Thess. and a leaf from Hebrews, and the beginning of the
Apocalypse. The six first-named leaves form three sheets, or attached

pairs of leaves. As stated below, Tischendorf believed the scribe of these
leaves to be identical with the scribe of the Codex Vaticanus (B).

ΠῚ THE UNCIAL MANUSCRIPTS 65

New Testament throughout ; and it may reasonably be
presumed that the appearance of his hand in the seven
leaves named is due to the corrections needed in the
original pages being of such a kind or extent that he
found it advisable to rewrite them. The corrections
by this hand, together with some others which appear
to be about coeval with the MS., are indicated in
critical editions by the sign νῷ Tischendorf (and his
conclusions are substantially confirmed by Lake, the
only other scholar who has minutely examined the
original MS.) distinguishes five other correctors of some
importance: δῦ, whose labours have been directed to
orthographical matters, such as accents and stops, and
who appears to have tired of his work before he
reached the end of the first Gospel; νῷ and πῇ,
correctors nearly contemporary with one another, and
probably of the sixth or early seventh century, both
active in the Gospels (where, however, they clearly
used different texts), but only sm“ appearing sub-
sequently ; x, a slightly later corrector, who appears
mainly, if not wholly, in the Epistle of Barnabas; and
xn‘, who restored faded portions of the text, with
occasional notes of various readings, about the eighth
century. Later notes are few and unimportant.

1 Another scribe of this group (hitherto identified with x¢, but by Lake
said to be certainly different) has added at the end of Esther a note
stating that the MS. had been collated with a very early copy, which itself
had been corrected by the hand of the holy martyr Pamphilus. Pamphilus
was the disciple of Origen, co-editor with Eusebius of a text of the Septua-
gint embodying the results of Origen’s labours, and founder of a library at
Caesarea which was the centre of textual study of the Scriptures, initiated
and inspired by Origen. Copies of Origen’s works were the special objects
of Pamphilus’ zeal as a librarian (Jerome, 4. cxli.). Bousset (Zextkritische
Studien zum N.T., in Harnack and Gebhardt’s Texte und Untersuchungen,
xl. 4) has found confirmation of the conclusion that the ‘c’ group of
correctors worked in the library at Caesarea, in the fact that the corrections
of x closely resemble those of the manuscript Hz, which was also
corrected from a MS. of Pamphilus (see below). Specimens of the hands

of the several scribes and correctors are given in Lake’s facsimile edition,
Plates II. and III.

F

66 CRITICISM OF NEW VEST AMENT cx:

In connexion with the different hands of this MS.
an important remark is made by Tischendorf, to the
effect that the scribe who wrote the seven leaves above
mentioned (together with Tobit, Judith, and Hermas)
is identical with the scribe of the New Testament in
the Codex Vaticanus (B). The importance of this fact,
if it were true, would lie in the identity of time and
place which it would establish beyond reasonable doubt
for’ these two great “codices,” ‘(Inthe \absence.ofia
photographic facsimile, or of the careful examination
of the MS. by a competent palaeographer, this observa-
tion has held the field until recently, though regarded
with constantly increasing suspicion. Similarities in
punctuation and orthography, which at first sight were
impressive, lost weight when carefully examined, and
when confronted with additional evidence from Greek
papyri. The publication of the photographic facsimile
has finally demonstrated the impossibility of Tischen-
dorf’s identification, which may now be relegated to
the limbo of exploded delusions.’

Palaeographers are generally agreed in assigning
the original writing of the manuscript to the fourth
century ; but as the matter is not one upon which
demonstrative evidence exists, it may be as well to
indicate the considerations by which the dates of this
and other early vellum uncials are fixed. It is only in
a few of the latest uncial manuscripts that the date
of production is expressly stated; the earliest is the
Uspensky Psalter of the year 862. For earlier manu-
scripts it is necessary to rely upon circumstantial
evidence, and to reason back from points which may
be regarded as fairly certain to those which are less so.
A trained palaeographer will learn to distinguish the

1 For Tischendorf’s arguments (which were accepted by Hort and
Scrivener) see his Movum Testamentum Vaticanum (1867), p. xxi. On the
other side, see Lake’s Introduction to the facsimile, p. xii.

ΠῚ re INC TAL MANUSCRIPTS 67

relative antiquity of different writings; and thus, by
allowing a reasonable space of time for each stage of
development, it is possible to arrive at approximate
dates in cases in which there is nothing but palaeo-
graphical evidence to go upon. A copy of Dioscorides,
at Vienna, is known to have been written for Juliana,
daughter of Flavius Anicius Olybrius, Emperor of the
West in 472, and this supplies us with an approxi-
mately dated example of writing about the beginning
of the sixth century. Again, the Codex Laudianus
of the Acts must have been written (as will be shown
below) somewhere about the beginning of the seventh
century. But it is clear that these manuscripts are
not as early as several others which are known to
us. They are less simple, more ornamented, showing
evidence of progress and development; and therefore
the other MSS. must be assigned to earlier dates in
proportion to their degree of relative priority. Thus,
to take the four oldest copies of the New Testament—
the Sinaiticus, Vaticanus, Alexandrinus, and Ephraemi
—it is clear that all of them are earlier than the above-
mentioned Dioscorides; and it is likewise highly
probable that the two first-named, which are the
simplest and least adorned of all, and show a greater
resemblance to papyrus MSS., are earlier than the other
two. If, then, the Codex Alexandrinus and the Codex
Ephraemi be placed a generation or two before the
Dioscorides, about the middle of the fifth century or
even earlier, the Codex Sinaiticus and Vaticanus must
be thrown back into the fourth century. Collateral
considerations, derived from the non-canonical additions
found in them, the evidence of early systems of division
of the sacred text, the comparison of early Latin
manuscripts of which the dates can be approximately
ascertained, and so on, all tend to confirm this con-

68 ..CRITICISM OF NEW TESTAMENT) cm

clusion; so that, though demonstrative evidence may
not be forthcoming, there is very good reason to accept
the general belief that these manuscripts are not later
than the fourth century.

It may be asked, however, whether they may not be
earlier; and it may be pointed out that we now know
uncial hands of a type bearing some resemblance to
those of these manuscripts to have been in use on
papyrus before the end of the first century. There are,
however, various considerations which forbid us to push
back the great vellum uncials to anything like so early
a date. As has been shown in the first chapter, there
is good reason to believe that for the first three
centuries of our era papyrus held its own as the chief
literary material; and the circumstances of the Church
make it highly improbable that such manuscripts as
the Codex Sinaiticus or Vaticanus would be produced
before the recognition of Christianity by Constantine.
Further, it is certain that the Codex Alexandrinus
cannot have been written before the latter part of the
fourth century, at earliest, since it contains (attached to
the Psalter) compositions of Eusebius and Athanasius,
who died in 340 and 373 respectively ; and the
Sinaiticus and Vaticanus cannot be separated from the
Alexandrinus by any very great gap. In the case of
the Sinaiticus evidence to the same effect is furnished
by the fact that it bears in its margins the section
numbers compiled for the Gospels by Eusebius, who, as
just stated, died in 340.1 It is true that they are not

1 Eusebius divided each of the Gospels into sections, which he numbered,
St. Matthew containing 355, St. Mark 236, St. Luke 342, and St. John
232. He then compiled canons, or tables, placing in parallel columns the
numbers of those sections in each Gospel which contain descriptions of the
same event. These tables consequently serve the purpose of a harmony
of the Gospels, without the labour of transcribing all the passages at full
length. Thus one table gives, in four parallel columns, references to
incidents described in all four Gospels; three, of three columns, give those
common to three Gospels (the combination Mark, Luke, John does not

ΠῚ THE UNCIAL MANUSCRIPTS 69

inserted by the first hand ; but it is clear that they are
not of much later date, since they are omitted from
two of the leaves supplied by the corrector x* (see
above, p. 65), who was contemporary with the original
scribe. The section numbers must therefore have
been inserted throughout the Gospels before these two
leaves were re-written by the corrector; and conse-
quently the MS. as a whole must, at the very least, be
later than the date at which Eusebius devised his
system of section numeration.’ All the indications,
therefore, point the same way; and though precise
accuracy is not obtainable with the existing evidence,
it is clear that the Codex Sinaiticus cannot be much
earlier than the second half of the fourth century, while
it can hardly be materially later.

A problem of considerable interest, if only it could
be plausibly solved, is that of the place of origin of this
MS.; but it will be convenient to defer the considera-
tion of it until we come to speak of the Codex
Vaticanus. Although Tischendorf’s argument as_ to
the identity of the corrector of » and the scribe of B
has broken down, yet Lake gives reason to suppose,
from the resemblance of the superscriptions, that the
two MSS. must at least have been together at a very

occur) ; five, of two columns, give those common to two Gospels (the
combination Mark, John does not occur); while one, of one column,
gives the passages peculiar to each evangelist. These tables are very
commonly prefixed to manuscripts of the Gospels, while the section
numbers, with a reference to the number of the table to which each
section belongs, are given in the margins of the text (often with the
numbers of the corresponding sections in the other Gospels, which really
dispenses with the necessity of a reference to the tables). Eusebius’ section-
division was based on a harmony of the Gospels (now lost) by Ammonius
of Alexandria, in which, the Gospel of St. Matthew being taken as a
standard, the corresponding sections in the other Gospels were written
down in parallel columns; and hence, through a misunderstanding of the
words in which Eusebius expresses his indebtedness, the sections are
commonly spoken of as the Amsmnonian sections, while the tables are called
the Lusebian canons.
' See note by Dean Gwynn in Scrivener, i. 94, and Lake, p. xix.

σγο CRT TICISM OF*NEW TESTAMENT © cu:

early date; and in any case the similarity of text
suggests the possibility of a community of origin.
For the present, then, it will be sufficient to state
that, while Hort was inclined to assign it to Rome,
Ceriani to southern Italy, and others to Egypt, the
opinion most generally held of late years would asso-
ciate it with Caesarea and the library of Pamphilus in
that place. Lake, however, who formerly advocated the
Caesarean view, has now come round to the belief that
Egypt is its real country of origin (see his Introduction
to the facsimile, pp. x—xv); and this is the conclusion
which to the present writer has always seemed the most
probable. |
The character of the text contained in the Codex
Sinaiticus cannot be fully investigated here; but it may
be said broadly that it is one of the principal repre-
sentatives of what has been called above the #-text,
though with not infrequent traces of the influence of
the 6-text. Tischendorf’s seventh edition of the Greek
Testament was issued in 1859, a few months before his
discovery of the Codex Sinaiticus, and while the Codex
Vaticanus was still practically unknown to scholars.
This edition was consequently based mainly upon
authorities of the a-type, agreeing substantially with
the Textus Receptus ; but his eighth edition, published
ten years later, was prepared under the influence of
these two great fourth-century codices, and it has been
computed that it differs from its predecessor in more
than 3000 places." The Sinaitic and Vatican manu-
scripts are, in fact, the two great champions of the
A-text, and it is primarily (though not by any means
entirely) to their influence that the textual differences
between our Authorised and Revised Versions are due.
It may be useful to indicate a few of the more import-

1 Scrivener, ii. 283.

III THE (UNCIAL() MANUSCRIPTS 7%

ant passages in which the Codex Sinaiticus departs
from the Textus Receptus.» In Matt. i. 25 it omits
τὸν πρωτότοκον (with B and Z); in Matt. v. 44 it
omits εὐλογεῖτε τοὺς καταρωμένους ὑμᾶς, καλῶς ποιεῖτε
Tovs μισοῦντας wupas (with B); in Matt. vi. 13 it
omits the doxology of the Lord’s Prayer (with BDZ) ;
it omits Matt. xii. 47 (with BLT’; ν΄ inserts the verse) ;
it omits Matt. xvi. 2, 3 (ὀψίας . . . δύνασθε), with B
and a few other uncials; it omits Matt. xvii. 21 (with
Bye it omits: Matt. «vill, 11 >(with: BL) 5 it) has’ the
Revised Version reading in Matt. xix. 17 (with BDL) ;
in Matt. xxiv. 36 it adds οὐδὲ o vids (with BD; Ὁ
has cancelled the words); in Matt. xxvii. 49g it adds
the incident of the piercing of our Lord’s side (with
BCLI). In St. Mark it omits υἱοῦ τοῦ Θεοῦ at i. I
(here departing from B and being corrected by s*) ; at
vi. 20 it reads πολλὰ ἠπόρει for πολλὰ ἐποίει (with
BL); it omits1x.44, 46 (with BCLA), and the end
of 49 with BLA; it omits the last twelve verses of the
Gospel (with B). In St. Luke it reads εὐδοκίας for
εὐδοκία in 11. 14 (with ABD); it omits δευτεροπρώτῳ
in vi. 1 (with BL); in x. 42 it reads ὀλίγων δέ ἐστι
xpeia ἢ évos (with BC*L); in xi. 2-4 it has the
shortened version of the Lord’s Prayer which appears
in the Revised Version (with BL); in xxii. 43, 44 it has
the incident of the Bloody Sweat (with most uncials,
but against ABRT; another hand, which Tischendorf
takes to be x*, has enclosed the passage with marks
of omission); in xxiii. 34 it has the word from the
Cross, “Father, forgive them” (with most uncials, but
against BD); in xxiii. 45 it has τοῦ ἡλίου ἐκλεί-
mwovtos instead of καὶ ἐσκοτίσθη ὁ ἥλιος (with BCL) ;
in xxiv. 51 it omits καὶ ἀνεφέρετο εἰς τὸν οὐρανόν
(with Ὁ). In Johni.18 it has μονογενὴς Θεός for ὁ
μονογενὴς υἱὸς (with BCL) ; in ii. 3 it reads οἶνον οὐκ

72 CRITICISM OF NEW TESTAMENT ©) cx

εἶχον, OTL συνετελέσθη ὁ οἶνος τοῦ γάμου (without
support from any Greek MS.) ; in iii. 13 it omits ὁ ὧν
ἐν τῷ οὐρανῷ (with BL); in v. 3, 4 it omits ἐκδεχο-
μένων τὴν τοῦ ὕδατος κίνησιν (with ABCL), and the
whole mention of the angel (ἄγγελος. . . νοσήματι)
with BCD; in vi. 69 it reads ὁ ἅγιος τοῦ Θεοῦ instead
of ὁ χριστὸς ὁ υἱὸς τοῦ Θεοῦ τοῦ ζῶντος (with
ΒΟΙ1)) ; it omits the incident of the woman taken in
adultery (with all the principal uncials) ; and (alone of
all manuscripts) it omits the last verse of the Gospel.
These examples from the four Gospels may suffice to
show something of the character of this manuscript ;
the bearing and value of its testimony must be con-
sidered later, in connexion with that of the other
witnesses, who now remain to be examined.

A. Codex Alexandrinus [Sod. ὃ 4] (Plate IV.).
—In contradistinction to x», this has been the longest
and best known of the early uncial copies of the Greek
Bible. Its original home appears to have been in
Alexandria, whence it was no doubt brought to Con-
stantinople by Cyril Lucar in 1621, on his transference
from the patriarchate of Alexandria to that of Con-
stantinople. By Lucar it was offered asa gift to James I.
of England, through the intermediary of Sir Thomas
Roe, the English ambassador to the court of the Sultan ;
but James having died before the gift took effect, it
was ectually, received, :by,Charles. ἢ 1627,,\ and
deposited by him in the Royal Library, whereby, when
George; 11 in, 1757 / presented) that Library te: the
nation, it ultimately passed into the possession of the
British Museum. The Epistles of Clement, which are
attached to the New Testament (and of which at that
time no other copy was known), were promptly pub-

1 A story that the MS. was found at Mt. Athos, and thence brought
to Alexandria by Cyril, lacks confirmation.

PLATE IV.

PS Sr pee

Ὺ

ἂς

τὸ COSC τε κα γε ΌΧΘΟΊΝΝΝ

ἃ ἡ τοΥυτοΥυ ME rHrroceremtcrors

VY PHuCeTArK PC PoCE coy! Sma

ONONKEIMENOMNEH λυ σαν

ΔΊΣ NHCCPENCTOCY mora

APHEAGY AMO TP NEHACOYVES

BIOY ALN OYNC ECON POPMO MEM

SPONTOINAOZAC MN “ITOK

MLPICAPCTE mecerpret tHE CON

ANOICECYAO KIAC

MIE PENCTOCICAHIHMAOO NAT

mM DMEICTONOY NONGOEALT OS!

oe ERIOIAN OOH POIMAILNE CE Fits |
a FFPC AAKHAOY CATE ROD MG PT
τ αισουσκιινλος νι WKATLAQCDMG
Ho WMATOY VO POPCVONOC

ἃ OOKCEENUDPICC NTIMCTN

SAPTHIENOONC! ICYCANVECE SNE

Be seorricsvomapian KAt POR
( ICAP POR PO JOCRKCEMEC

NONCHTCN ENT LAOrm Vee

RENICENWPICANHICPEEONTT

~ PEXVOCPOYAAAHIOC ΒΕῚ OCYONY Fore

τα τον: ταιλιοΥ POY TOY

q RAFT FANTTPE CO IANOYCAN FCOCEON

x ee . MACAMN! VEPrrepNarsrsHroen mo

ees VETO TUN ΟΜ Οὐ μετα POC NY

see |UD PX EAT NOW Crepe)

i =e —epaPriMATREAYTACY™M KRRAAKQOYEN
nage ἐν Ὁ το κι αιλλυστιο τ ΑΙ τῖ ((

ἔν ὁ.

&

twpers

~ hl ppeEANOTFOIMAIN CG AOEAZO ©
πε γα ς eAIAINOY ει σον όσον
Sos i>. ENT PACT NOICHHKOYCAN KX}
Aer KAN τος σοκλαλεεν γοῦν,

: τοὺς Δι CEMANCOIE*
ἔ } mcprroKrarroy CPrreMcin
LAYTON IKATCIKATION εὐ ΟΜΝ

Ay TOY IC rONRA! LOC MYO Por
CAFYVEAOVI FPFOTUOV CY AAT κίον
ει. ἩΣΙΧΥ ΤΌΣΟΥ ΓΕΠΧΚΟΤᾺΤΑ

- ]XmMorce mniconcannn MC
ὯΣ SOY KAOA PIC MOVAY ΜΚ ADS
POM NOMONMCDCCOIC ANTE
‘ PAPORAYEONCICICPOCOAYM®
ir rapaci "ΟΝ λα} KAOUWIC
Ὡς ἈΡΟΚ ΛΙΧΜΟΙ ΟΜΝ ΤΥ ΓᾺΡ
. χαὐιογνεχνκ ιν κι ινληίῖνι τς 3.

ς ἸΛἐ ΥΥΧΟΟΥ ΧΙ ΟΎΟΙΧΤΙ ᾿ς ΝΣ
MOC preMOMONCrS NECIMCLKY:

CODEX ALEXANDRINUS.

. ᾽ S \ 5 ( ὡς € sy

(Scale 2 : 5.

eer ers πο ἘΣ
SS ES FE cae parent
‘ ᾿ ὅδ - “aiginme’>-

αν αι MNOMUDEKY OPUS,

ae ΩΝ
~ τὴ ᾽
ae saw
: Mis ΣῊΝ
ee
ZOVYOOC TPY EON Cina εχ ον :
; ,VOVCHUPICVTe fain Ν
ἐς κειλΟυ TEMA IOC MAP κα ε νον οὶ 4
QY MECN KWAEOAPIOCOY TOC ἍἌ,
RBI A TOC KATE VYARBEEOTVEPOCALD KO ἔξ
“ες ἈΝ IC BPP POW ERAN
ἸζΆΥ εὐ EXPERAADTONCEL Ry PON icant”
XxY4 VEIN NPPeeant FORMAN ORY FO ©
poyrer’ee POVAT HOY Mt 1rAGQINGXY
ΜΝ λα EMEPLAT ETON ICY
1c Ars tQOC NCHS HcCiCororpe:
Ix NEM PUI ECAP ΤΟΝ εὐ COP ONGIC
“POYPENTALORIITA POYRIOT FOONERY
POYE KATALONIOICMOUNONTOY
NOMONWEPCT IXY POY WARY FOC
CINESXRVORNY ες ΠΟ VAS XP IK RAKES
AY POY IKAEC YAO! HOGNT FONG
KALTCHIEPOCM MY MN TORYWECIC
ἽἼ Om ROYAONCOWVACE EEO ΝΎ ΤΆ —
POPLEMACOVENCIPELIN EO Tdi
OLODOAAMOEMOY αν ει ΓΝ
COW OF sea Ape aes &
COYHMORDE ISN PUD TOI AMD IN
LD coc ICAI IO K Ay y POOH Us
WALA OS AN gp LOY THA

QNDE EPLOOECIIOC sq)> WATE ERATE PRY ΤΟΥ
ὌΝ ΧΟ CF εὐ Ὁ τ ἈΝΧΡΥ
κι ον ΡΥ ΝΥ LOY ΣἈΙΟΥ ΚΟ
τ ΟΡ ΧΥ τ OGY RABE TONES

“recs >COCSAAMP EAS Ft 1M ΓΆΣΥ ΣΙ

ΓΧον ον POC κου ας σε οῦ"
IWATANAC ΝΟ σαι τ λλκιομεν
THEA ISANIOTCE HY MICE EOS ANTE ING
ΕΝ ΟΡ. ICRIQOVWAC NYA IC

τ τ ν ἈλτταλαΟΧΟΎ COVA fonshans ee
“ιν ANXI HORAAYPOWCEN ee

ἝΞ RIN LOARCLY IN IKA PAL CLIN AF RAO OME es
* Canriiarcnxis Ox )hiericQoynp> δ :

CPXRMNONDENCINDDY HCACH (ta ool
“$s PORCKHK! HKYINCNTIMCPA Crepe ἢ
ANIC ΖΘ ΣΝ σα ἐὰν MCF ΡῈ ἐκ
ANAPOCX! του eicrl PYPOC NCOP
KAINW UPLX ES PACD TP CINODS :
MINONEXPCCCAPUIN oye pie
CVAPOXIOTOY ΤΟΥ nine”

HATAC UICC OIE {KAT PCYOYC®
TIVE RNIKATSE IMTS FCMERNTE TE ΜΕ
“ας prc ΦΥΛΆΣ ΘΕΟΣ, ΩΣ
reorieyreul WCF AIC ARAL NC PDT J
PPO IRC PE POL §POCNG XKOMOMSE οὐ

ag

Fifth Century,

ϑ

versions, and in Origen and Irenaeus. )

To face p. 72.

"

ΠῚ THE, UNCIAL’ MANUSCRIPTS 73

lished in 1633 by Patrick Young, the Royal Librarian ;
the Old Testament was edited by Grabe in 1707-20,
and the New Testament by Woide in 1786. Pre-
viously to this date the text had been repeatedly col-
lated by various editors of the New Testament (first for
Walton’s Polyglot in 1657, then for Mill’s great edition
in 1707), so that its contents were already well known.
Woide’s edition was reproduced in a handier form by
B. H. Cowper in 1860, and in 1879 a definitive publi-
cation was reached by the issue of a photographic fac-
simile under the editorship of Mr. (now Sir) E. Maunde
Thompson, the late Director of the British Museum.’
In its present binding, bearing the royal arms of
Charles I., the Codex Alexandrinus consists of four
volumes of moderate quarto size; but originally it
formed a single volume of considerable thickness. It
consists of 773 leaves” of thin vellum (so thin that in
many places the ink has worn completely through it),
measuring 122 inches by 103 inches, and written in a
firm and fairly large square uncial hand, with two
columns to the page. New paragraphs are marked by
enlarged capital letters ; but if the end of a paragraph
falls near the beginning of a line, the succeeding para-
graph is commenced in the same line, the enlarged
capital being reserved for the first letter of the line that

1 The Old Testament was published in facsimile type, with elaborate
prolegomena, in 1816-28 by the Rev. H. H. Baber, Keeper of Manuscripts
in the British Museum; and a photographic facsimile, edited by Mr.
Maunde Thompson, followed that of the New Testament in 1881-83. It
has also been collated throughout for the Cambridge Septuagint. Speci-
men facsimiles in Pal. Soc. i. 106, and in facsimiles of Biblical MSS. in
the british Museum, Pl. 11. The fullest description of the MS. is con-
tained in Thompson’s Introduction to the facsimile publication. A handier
and cheaper photographic facsimile, on a reduced scale (about # of the
original), of the New Testament was edited by the present writer in 1909.

* Originally 820, allowing one leaf for the lost conclusion of the Second
Epistle of Clement, and five for the Psalms of Solomon, in addition to the
forty-one which are missing from parts of the Old and New Testament and
I Clement, as described below.

74 CRITICISM OFF NEW TESTAMENT ὃ

follows (see p. 50). Four different hands may be dis-
cerned in the original text of the New Testament, one
having written Matthew, Mark, and most of the
Pauline Epistles; the second Luke, John): Atcts, the
Catholic Epistles, and the first part of the Pauline
Epistles (to 1 Cor. x. 8), the third the Apocalypse and
the fourth (who also wrote the historical books of
the Old Testament) the Clementine Epistles. Several
correctors’ hands have been employed upon the MS.,
but the only alterations that are of much importance
are (1) those made by the original scribe (A'), and
especially (2) those made by the first corrector Δ.
who would appear to have been nearly or quite con-
temporary with the MS. Other corrections are very
much fewer and less important.

When complete, the manuscript contained the whole
of the Old and New Testaments, with the addition of
the third and fourth books of the Maccabees at the end
of the Old Testament, and the two Epistles of Clement
of Rome and the Psalms of Solomon at the end of the
New. The latter work has now completely disappeared
from the MS., but its former presence is proved by the
table of contents at the beginning, in which, however,
its title is distinctly separated by a space from those
of the canonical books. The Epistles of Clement, on
the other hand, are included with the canonical books,
following the Apocalypse in the list of contents with-
out break or distinction. The latter part of the (so-
called) Second Epistle of Clement has disappeared
along with the Psalms of Solomon; and one leaf of
the First Epistle is also missing. In the New Testa-
ment nearly the whole of St. Matthew is lost (as far as
chap. xxv. 6); also John vi. 50—viii. 5 2 (where it should
be noted that a calculation of the extent of space
missing shows that the section on the woman taken in

ΠῚ THE: UNCIAL MANUSCRIPTS 75

adultery, vii. 5 3—viii. 11, can never have formed part of
the: MS. )s-and):2 ‘Cory iv. 19=xil,’6;

In point of age, the date usually assigned to the
Codex Alexandrinus is the fifth century ; and unless
more precise evidence should come to light, it is not
likely that this estimate will be disturbed. A superior
limit is given by the fact (mentioned above, p. 68),
that it contains, prefixed to the Psalter, treatises by
Eusebius, who died in 340, and Athanasius, who died |
in 373. It also contains the Eusebian canons and
sections. The manuscript cannot, therefore, at the
very earliest, have been written before the second half
of the fourth century ; but this is the date which has
already been assigned on fairly good grounds to the
Codex Sinaiticus, and although it is not necessary to
separate the two manuscripts by any wide interval,
there are indications that the Alexandrinus is the later
of the two. The handwriting is firmer and_ heavier,
less reminiscent of the papyrus type; the use of
enlarged capital letters marks an advance; and the
arrangement of the text in two columns to the page is
also a later stage than the three columns of the Vati-
canus (B) and the four of the Sinaiticus. If, therefore,
we' place the Alexandrinus in the first half of the
fifth century, it is not likely that we shall be far wrong.

Its early history is partially revealed by inscriptions
on its fly-leaves. A note by Cyril Lucar states that,
according to tradition, it was written by Thecla, a
noble lady of Egypt, shortly after the Council of Nicaea
(A.D. 325), and that originally her name was inscribed
at the end of the volume, but that the page had
been lost through the mutilation of this part of the
manuscript, Ὑπὸ date\.is evidently too. early, for
reasons given above, but the rest of the tradition is
plausible enough, save that the whole MS. was certainly

χα. CRITICISM OF NEW TESTAMENT cu.

not written by one hand. An Arabic note, of the
thirteenth or fourteenth century, so far confirms it as to
say that the MS. was written by “Thecla the martyr.”
Another Arabic note, signed by “ Athanasius the
humble” (possibly the patriarch Athanasius III., who
died about 1308"), states that it was presented to the
patriarchal cell of Alexandria; and a somewhat later
Latin note assigns the gift to the year 1098, but with-
out giving reasons for this assertion. In Alexandria,
and in the possession of the patriarch, it certainly was
at a later date, since Cyril Lucar brought it thence to
Europe ; and all the evidence points to its having been
produced in or about that town. The titles of some of
the books, which have been inserted by a writer slightly
later than the original scribe, contain forms of the
letters a and pw which are characteristically Coptic, and
almost demonstrate a very early residence of the MS.
in Egypt. Further, in the Old Testament this manu-
script has a type of text which is by some scholars”
identified with that of the edition of the Septuagint
by Hesychius, which was prepared and circulated in
Egypt; while others,> who doubt the actual identifi-
cation, at least admit a close kinship. Thus, though
demonstrative proof is wanting, such indications as
there are point to Alexandria as the place of production,
while there is no evidence in a contrary direction.

In character the text of the Codex Alexandrinus in
the New Testament presents a curious phenomenon ;
for whereas in the Gospels it belongs emphatically to
the a-type, and indeed is the best representative of
that family, having the text in a purer form than that

' Professor Burkitt, however, tells me that Arabic palaeographical
science is not competent to settle the date of this note with any confidence,
so that this identification has no high degree of probability.

2 E.g. Ceriani (De Codice Marchaliano, 105, 106).

δ #.g. Cornill (Zzechzel, 67 ff.).

ΠῚ THE UNCIAL MANUSCRIPTS ἢ

which appears in our Textus Receptus, in the Acts
and Epistles, and still more in the Apocalypse, it
belongs rather to the $-type. The phenomenon is,
of course, quite easy of explanation. When codices
first superseded papyrus rolls, and it became possible
to unite all the books of the New Testament in a single
volume, the texts of different parts of such volumes
would necessarily be taken from different rolls; and it
might easily happen, as has been the case here, that
different books, or groups of books, would be copied
from rolls containing different types of text. Thus in
the MS. described below as A, the Gospel of St. Mark
has evidently been copied from an original of a
different family from that of the other three Gospels.
It must consequently be remembered that A in the
Gospels represents a different type of text from that
which it has in the rest of the New Testament.

B. Codex Vatieanus [Sod. 6 1] (Plate V.)—This
is generally held to be both the oldest and the most
valuable of all the manuscripts of the Greek Bible. Its
modern home is, as its name implies, in the Vatican
Library at Rome, and here it has been at least from
1481, since it appears (as a “ Biblia in tribus columnis
ex memb.”) in a catalogue of that date.’ It was thus
n Europe a century and a half before the Alexandrinus,
and nearly four centuries before the Sinaiticus, and yet
it was later than either of these in becoming fully
accessible to modern scholarship. During the greater

1 It is stated by Vercellone (Del? anttchissimo codice Vaticano della
Brbbia greca, 1860) and repeated by Gregory (Zextkritzk, i. 35) that this
entry occurs in the catalogue of 1475 ; but this catalogue has been printed
in full by E. Miintz and P. Fabre (La Azbliothéque du Vatican au xv®
stécle, 1887, 159-250), and it contains no Bible answering to this descrip-
tion. Vercellone’s reference relates really to the catalogue of 1481 (Bibl.
Vat. MS. Lat. 3952, f. 50), mentioned, but not printed, by Miintz and
Fabre (p. 250). It would appear, therefore, that the MS. entered the

Vatican between 1475 and 1481. I have to thank Dr. F. Spiro for verify-
ing the above reference to the Vatican catalogue.

55. GRITIGISM OF (NEW TESTAMENT cu

part of that period scholars did not care to investigate
it scientifically, and when at last they wished to do so,
permission could not be obtained. A few readings
from it were supplied to Erasmus by his correspondent
Sepulveda, but too late for use in his editions of the
New Testament. In 1669 a collation was made by
Bartolocci, librarian of the Vatican; but this was not
published, and was never used until Scholz in 1819
found a copy of it in the Royal Library at Paris.
Another was made in 1720 for Bentley by Mico, and
revised by Rulotta ; but the former, which was imperfect,
was not published until 1799, and the revision not
until 1862, while a further collation by Birch in 1780
was quite incomplete. When the manuscript was
carried to Paris by Napoleon, along with other treasures
of the Vatican, it was examined by Hug, who first
proclaimed its extreme age and great importance ; but
nothing more was done in the way of a complete
publication of its contents before it returned to Italy,
and then it was withdrawn from the use of foreign
scholars. First-rate textual critics, such as Tischen-
dorf and Tregelles, were barely allowed a sight of it,
and only a few of its readings were grudgingly con-
ceded to the world from time to time. The fact was
that the Vatican had itself taken the publication of it
in hand,’and had had the text in type’ since’ 1838 ;
but though the work was committed to a scholar of
European reputation, Cardinal Mai, he executed it in
so slovenly a way that it was held back from publica-
tion during his lifetime, while the two editions (large
and small) which were issued after his death, in 1857
and 1859, were so inaccurate and so much at variance
with one another as only to prove conclusively the
necessity of having the work done over again. In
1866 Tischendorf obtained leave to examine it in

PEATE V.

soe, FS
ROR Paseo

Do pine pe Eg

᾿ 4
;
ty '

4.79
ἘΠ eg τοι vitfes era esc ee μὰ MAK YEA Y OS [aL pandere ΐ
(aie are apes maT 1CEWDCA OB ares ACHAMN TE CAMA KE καλχμ.- |
ὝΝ =e ΜῈ ΕΥ Ἐπ πε oun te cays MEKMWOTTRO CW Reppin 3
creere: ; AMMEN He MADAKON IATA BPM AS AOBAN YEA OR rpiz- ‘
PEC RAPS RICHER CD RET QCYNMCACEHKAIFAL MENOLTH MAY TH Cine” ᾿
MOCK EMH RATAN OYAEAOZ ACTA TGar PEG ls mea oy Meek a

PEN OKO MERRY ro KOPACMENONE Noy Kqpoactucd, δὲ ρους

φν hee { MENOTO Tie
i ae LOT OAR ὝΛΔλια.
εὐ Η θειολῦ > με:
ak Terra EN MOY
PAN TAAA A PIMG yo 08

a ZOn reddy. ὦ
WAALINALCI MAIC AAA”
PLA REEMA δίατο CAP RE
SAE ne ΟἽ θνει ἐλᾷ AS
TOY πημᾶχο RAEN AT
way KY No “PONG H

Oy Kors iy χὰ ore
RAN: τε PON AGT σα
COA CERRY τανμ καλὴ
ἸΚᾺ NOT woe MICU ΕΚ ΠΥ,
«3 OK ALT RE μεσ ἢ
MACAIAKONOYCR RIN ῶς
ATROHKHC Gyr ADELA
“TOCKAAA TING Y Matar:

oe “PRN A He cha ΕΜ ΡῈ

ΠΟ Apr Kass RATOKT

ery se Ey nea ze
ongel ER RED PE?
KRTOYSEs, yer =
MATIC TEP yr icon ern
: κιϑοιούγόμδηέν AS :
Ἐηῶ ὙΒ μὴ ΑΝ ACen

PreNicarreycyroye
AE AHAGICTON down
: rYCEWCAIK rin ke
: = τ ag Gera ia
roy i es “TAP Oy kee”
Sa τανῦν ἘΞ :
WS κκοντα ταγημεν
μαγοσος ταιέρλάτη:

τὰς κακάμε FRIC pe

Pare Me PETG IHe REN
THey ner barroyouc
AOtNC el εαγτοκατα
rdyMenom ᾿λλόπβοης,
ττολλώσμλλον TOME
μομέμλοχμ ἕπου.
τε ΘΎ ἘΘΥΚΥ Κημεν
PHAATIOAA Rie.
MM pubes eo OxKK IO} ki RO
ον γς εὐ TI gerd
AY MACH TON acu
NOMaY POY po TOs
KF ῬΕ 1 arto: οὐτῶγο,
᾿ RHA ἐκ τὸ, Ἔφθαρο:
TAL POT μενοι oe
ery CAH TAN OM MATA
RYT’ NAM poet Ne sen
MEPOM HAE
wa yuma dnp ἐγεια
Ὁ ἐγ + Ῥεῖ MAKRIREAL ROM
wre Dae NETMHAN A καὶ

τ όκες ΜΝ Op τὲ NOP

"AOR APE EE AER ANG RIC
«ἦμεν

OM ANIA Am RNA :

PET NCW HOTAI δε ἧς

eres RRA RERERE ἘἘΕῚ : NR Ap
OM PAN Ay τ ει μὲ (eke
a BAAN SS GERe es Coe oe

ORM The re ἘΠΕῚ κα.
AY A ee
BONE AD "TAD κει Tate ey
ΒΗΓ,
ΜῊ τὰ εὖ ἔτ PTreka
Ry Mts AS ek Crone
MAECTUNOY τ εὐ ney.

[ae

KACO E Reasynaes
waroc
KON TECTHNAIAKONID,
TAY THN καθ HAGH
Sx MENGY KET Ka KOyur
AAAR AMEN AMEG ATR
KPYNPAT HCAIORYy MAE
MPH ILE INATOYN Tec
ENTANOY PERM ΑΕ AH
ADY NTE CTONAS ONY
SYA TH AME Pasce:
yHebanoes Sea τε
NON TEC ἐλ Toyon p
PACANCY NIA BCID
Θ᾿ wna ἐμώγηοι τὸ
ΘῪ οταλε KALE εκ κ εκὰ
AYM MEN ONTOEYar re
ἈΤΘΜΗ Μεθ κέ τε CAA
AY MEN GILCECTIN ke xa
AY Me esonrén διοόδε
“voy RIMNOOTSy-Te
Bap xmcen τὰ sees
κά λστ ἐτῶν Aniie-rus
ὌΡ αοὴν Ὁ Ala shy τρῶς
wprcugurToyéyar —
ben toy rncaos corey
Seer cacen ery
sy Py tAPeéayToyces
Joabe See ras ων
Enéay roe aL Aoy
FMDN ATE γοτοσεδαι
NMMEK EKO TOYS

CODEX VATICANUS. Fourth Century.

᾿Ξ ΘΟΕ 5. - 7.

The page shown contains 2 Cor. iii. 1-iv. 6.
show the untouched writing of the original scribe. )

To face p. 78.

τ ΤΟΥ res * f-

Cols 2; ll 35-36,

©

SP Reiainnate gee

a

es THE UNCIAL MANUSCRIPTS 79

passages as to which special doubt existed, and was
able to work at it for forty-two hours in all—a time
which would have been longer but for his own breach
of contract in copying twenty pages in full instead of
confining himself to an examination of isolated passages.
Out of all the material then available, supplemented
and corrected by his own labours, Tischendorf in 1867
produced an edition of the Vatican New Testament,
which represented a great advance on all previous
knowledge’ of the MS.; but in 1868. the Papal
authorities themselves issued an edition of the New
Testament, prepared by Vercellone and Cozza, which
was followed in successive years by volumes containing
the Old Testament. Finally, in 1889-90 all previous
publications were superseded by a photographic fac-
simile of the whole MS., which renders its evidence
available to all scholars, except so far as certain
questions as to the distinction between the hands of
correctors and the original scribe must always necessitate
a reference to the original.’

The manuscript on which so much labour and so
many searchings of heart have been spent is written on
very fine vellum, said (but without much authority) to
be made from antelopes’ skins. In shape it is square,
the leaves measuring 104 inches by 10 inches, and ten
leaves (or five sheets) making a quire. The writing is
a small, neat uncial, distinctly recalling the papyrus
hands of the first and second centuries, on which it
appears to be modelled; it has little of the heaviness
which soon came to mark the style of writing upon
vellum. Unfortunately its appearance has been spoilt
by a corrector, who thought it necessary to trace over
every letter afresh, only sparing those which he regarded

1 A specimen facsimile of a page may also be seen in Pal. Soc. i. 104,
and partial or reduced facsimiles in many handbooks to the Bible.

So CRITICISM°ORPYNEW TESTAMENT © δὰ;

as incorrect and therefore better allowed to fade away.’
There are three narrow columns to each page, a
survival from the usage on papyrus, though not so
marked as in the Sinaiticus. There appear to be no
accents, breathings, or stops by the first hand. Correc-
tions have been made by two hands—one a con-
temporary or nearly contemporary reviser (B*), while
the other (B?), who retraced the whole, is placed by
Tischendorf in the tenth or eleventh century. The
original scribe of the New Testament is different from
the scribes of the Old Testament, of whom, according
to the same scholar (though the matter is open to
doubt), there are two.

The Codex Vaticanus originally contained the whole
of the Greek Bible, but has now lost Gen. i. 1—xlvi. 28,
Psalms cv. 27—cxxxvii. 6 in the Old Testament, while
the books of the Maccabees were never included in it;
in the New Testament the conclusion of the Epistle
to’ the’ ‘Hebrews (ch. ix. 14 to the end), the Pastoral
Epistles of St. Paul, and the Apocalypse, are absent.
The Catholic Epistles follow the Acts, according to
the common order in Greek MSS., and so have escaped
destruction.

With regard to the date of the Vatican MS., there
has been substantial agreement among palaeographers
since Hug originally assigned it to the fourth century.
In simplicity of writing and the absence of ornament
or of enlarged capitals, it presents features earlier in
type than any of the other great vellum manuscripts ;
and to this must be added the fact that its divisions
of the text seem to be earlier than any other. The
Eusebian sections do not appear in it, but in place of
them there is a different division of the Gospels, found
only in one other MS. (2); of these sections there are

1 Four untouched lines may be seen on p. 1479 of the MS. (Plate V.),

ΠῚ THE UNCIAL) MANUSCRIPTS 81

170 in St. Matthew, 62 in St. Mark, 152 in St. Luke,
and 80 in St. John.’ In the Acts there are two in-
dependent sets of sections, the earlier consisting of 36
sections, the later of 69; the second of these is also
found in part in ». Similarly in the Catholic and
Pauline Epistles there are two sets of sections. Since
the Eusebian sections, when once they were introduced,
rapidly spread into general acceptance, the divisions in
B appear to be of earlier origin, and tend to show at
least that the manuscript represents an old tradition.
In simplicity of writing and absence of ornament or of
enlarged initials, the Vatican MS. is the most primitive
of all, and the nearest to the papyri. Unless future
discoveries should seriously disturb the basis of our
palaeographical knowledge, the position of the Vaticanus
may be regarded as substantially fixed.’

It is far from being the same with respect to its

1 Τὶ has been argued by A. Schmidtke that this section-division (of which
traces are also found in x and in Cod. 579) goes back to the Gospel-
harmony of Ammonius, which he also thinks (though with very slight
evidence) was the basis of a recension of the Gospels by Hesychius.
Hesychius is known to have made an edition of the Septuagint in Egypt
in the fourth century, and Bousset and others hold that he did the same for
the New Testament, and that B represents this edition.

2 An attempt has been made by A. Rahlfs (Machrichten der kinigl.
Gesellschaft der Wissenschaften zu Gottingen, 1899, pp. 72 ff.) to fix the
date of B more closely. He argues that it must have been written later
than A.D. 367, because its contents correspond with the list of canonical
books prescribed by Athanasius in his thirty-ninth Festal Letter, written
in that year, Von Gebhardt, however (Theologische Literaturzettung,
1899, col. 556), while admitting the probability of Athanasian influence
to be shown by the coincidence, argues that B must have been written
before 367, because the sharp distinction drawn by Athanasius between
canonical books (κανονιζόμενα) and books recommended by the Fathers
of the Church to be read (ἀναγιγνωσκόμενα) is not observed, the
ἀναγιγνωσκόμενα (Wisdom, Ecclus., Esther, Judith, Tobit, in the Old
Testament) being inserted among the canonical books in B without distinc-
tion (though in a single group). The New Testament ἀναγιγνωσκόμενα
mentioned by Athanasius (the Didaché and Hermas) do not of course
appear in B, since the end of the MS. is lost. The da¢a are consequently
too uncertain for any conclusion to be drawn with confidence. The most
that can be said is that the contents of B rather tend to support a belief in
its Egyptian origin.

G

85, CRITICISMPOFr NEW TESTAMENT © ‘cu.

place of origin, about which the most distinguished
critics have disagreed. The Roman editors suggest
Egypt—a suggestion also made by some of the earliest
critics ; Ceriani is in favour of southern Italy ; Hort
inclines to Rome; Mr. Rendel Harris has pleaded
hard for Caesarea, and for a close connexion, at least,
with the library founded there by Pamphilus and
Eusebius, and this view (which had already been
propounded by Canon Cook and Scrivener) has found
considerable favour of late years.

Some such connexion appears to be suggested by the
following facts. The later chapter-division of the Acts
in B (into sixty-nine chapters, made by a hand only
slightly later than the first), which is also inserted in
the margin of x by a very early hand, appears to be
based on the division adopted by Euthalius in his
edition of the book,’ made about the middle of the
fourth century. Now the earliest extant MS. of this
edition (Cod. H,) contains a colophon stating that it
was collated with a volume in the library of Caesarea
written by Pamphilus himself. This colophon was very
possibly copied from an earlier MS. (as is not unusual
with such notes); and in that case we should have
evidence that a very early copy (if not the archetype)
of the Euthalian edition of the Acts was at Caesarea,

1 Euthalius, bishop of an unknown place called Sulca, was the author
of an edition of the Acts and Epistles in which the text was arranged
colometrically (z.e. in short clauses, corresponding to pauses in the sense),
and provided with prologues and chapter-summaries. Until recently he
was supposed to have lived in the middle of the fifth century; but Dean
Armitage Robinson (Zzthaliana in the Cambridge Zexts and Studies, iii.
3, 1895) has shown good grounds for placing him a century earlier.
Kuthalius’ chapter-division in the Acts was into forty chapters and forty-
eight sub-sections, making a total of eighty-eight ; but the Dean points
out that some of the sub-section marks, being only asterisks or letters,
might easily be dropped out or overlooked, and argues that the coincidence
of the sixty-nine divisions in B with those of Euthalius is so general
(though not universal), even in some rather unlikely divisions, as to point
to a common origin.

ΠῚ THE UNCIAL MANUSCRIPTS 83

whence its system of chapter-division may have been
inserted into B and ». Another MS. of this edition
contains, in addition to the ordinary chapter-division of
Euthalius, another division into thirty-six chapters
which is identical with the earlzer chapter-division in B.
These coincidences do not frove a common residence
at Caesarea for x, B, and the Euthalian archetype, still
less that x and B were actually written there, but they
point in that direction. Moreover, as already shown,
excellent authorities believe that there are proofs of a
community of origin between B and x, and the note of
the corrector of at the end of Esther (see p. 65, above)
shows that that MS. was at Caesarea probably as early
as the sixth century. Further, the text of B in the Old
Testament is held by competent critics, such as Hort
and Cornill, to be substantially that which underlies
Origen’s Hexapla edition, completed by him at Caesarea
and issued as an independent work (apart from the
other versions with which Origen associated it) by
Eusebius and Pamphilus. Dr. Gregory, the author of
the Prolegomena to Tischendorf’s last edition of the
New Testament, is even disposed to regard B and x as
having been among the fifty MSS. prepared by Eusebius
about 331 at Constantine’s command for the churches
of his new capital (and so also von Gebhardt); but
there is not the least sign of either of them ever
having been at Constantinople. That there was some’
connexion between the MSS. and Caesarea seems \
fairly certain ; but the arguments which prove it do
not go so far as to show that this was actually their /
place of origin. The fact that x was collated with the
MS. of Pamphilus so late as the sixth century seems to ὃ.
show that it was not originally written at Caesarea; ~—
otherwise it would surely have been collated earlier
with so excellent an authority. Origen’s textual’

34° CRITICISM OF NEW TESTARIENT ‘cu

information was collected in Egypt, where he began
the preparation of the Hexapla, rather than in Caesarea
(the home only of his later life), and if B is in harmony,
with the text used by him it would be accounted for by "ἡ
the hypothesis that B was written in Egypt. Indeed,
if B and x had been written at Caesarea, it is probable
that in the Old Testament they would have contained
the Eusebian edition of the Septuagint, which is not
the case. Further, it is noteworthy that the section
numeration of the Pauline Epistles in B shows that it
was copied from a manuscript in which the Epistle
to the Hebrews was placed between Galatians and
Ephesians; an arrangement which elsewhere occurs
only in the Sahidic version, current in Upper Egypt.’
A connexion with Egypt is also indicated by the fact
that, as in the Codex Alexandrinus, the titles of some
of the books contain letters of a distinctively Coptic
character, especially the Coptic 9a, which is used not
only in titles, but also very frequently at the ends of lines,
when space is to be economised.” The resemblance of the
writing of both B and » to hands found in Egyptian
papyri cannot be pressed, since we do not know enough
about the contemporary hands outside Egypt to say if
this resemblance is worth anything; nor is there any
force in the argument derived from the fact that two
damaged leaves have been patched with papyrus, since
papyrus was still used out of Egypt in and after the
fourth century. On the whoie, it can only be said that
the evidence does not admit of a decisive verdict.
There is fair evidence of a connexion with the textual

1 Cf. Serivener, i. 57.

2 It must be remembered, however, that, although this form of mu was
undoubtedly used in Egypt, we cannot affirm, for want of evidence, that its
use was confined to that country. Still, its frequent occurrence in B is
prima facie more favourable to an eastern than a western origin for that

MS., and consequently for x too (in which it likewise occurs, though less
frequently).

ΠῚ THE UNCIAL MANUSCRIPTS 85

school of Caesarea, which does not exclude an actual
origin in Egypt, from which the school of Caesarea took
its rise; and other evidence is rather in favour of such
an origin. The evidence in favour of Rome, on which
Hort relied, has only been summarily stated, but it does
not appear to be very weighty.’

Of the character of the text of B much will have to
be said in a later chapter. Here it must be sufficient
to say that it is the foremost champion of what we
have called the 8-text, and to refer to some examples
of its readings, generally in common with x, which
have been given above (p..71):' It differs from: the
a-text especially in the Gospels, and most notably in
the way of omissions. When it was first examined, the
witnesses that supported it to any great extent were
few. Since that time they have increased in number
and in variety of testimony; but the examination of
this evidence is the central problem of New Testament
textual criticism, and must be reserved until the re-
maining manuscripts and versions have been described.

B,. Codex Vaticanus 2066.—See below, Cod. 046.

C. Codex Ephraemi reseriptus [Sod. ὃ 3].—As
the epithet of its Latin title denotes, this manuscript is
a palimpsest; that is, the original writing has been
more or less completely removed, by washing or scrap-
ing, from the surface of the vellum, which has then
been used again to receive the transcript of another

1 Hort (Zutroduction, pp. 264-7) rests his argument mainly on (1) certain
spellings of proper names, such as Ἶσάκ and ᾿Ιστραήλ, which show Western
or Latin influence; (2) the fact that the chapter-division in the Acts
common to x and B (see above, p. 81) occurs in no other Greek MSS.,
but is found in several MSS. of the Latin Vulgate, including the Codex
Amiatinus. But this argument is placed in quite a different light by Canon
Armitage Robinson (Zu¢halzana, pp. 42, 101), who connects this system
with the divisions of Euthalius, and suggests that it was introduced into the
Vulgate by Jerome himself, as a result of his studies at Caesarea. If one
basis of Hort’s belief be thus removed, the other is too slight by itself to
give it much support.

86 (CRITICISM OP UNEW TESTAMENT | ca.

work. This was a device employed not infrequently
at times when vellum was scarce, and not a few manu-
scripts have come down to us with such double layers
of writing, in which, strange as it might appear to the
authors of the intruding treatises, the earlier work is
almost invariably the one which interests us most at
. the present day. Perhaps the most valuable of such
manuscripts are those which contain the Justztutes of
Gaius and the De Republica of Cicero (discovered
respectively in 1816 and 1822), since these works were
otherwise wholly lost ; but the best known of them is
probably that which is now before us. It was brought
from the East to Italy inthe éarly days of ‘the
sixteenth century, when Greek scholars and Greek
manuscripts found so warm a welcome in the West,
and it became the property of the Medici family.
With Catherine de’ Medici it travelled from Italy to
France, and so entered the Bibliotheque du Roi in
Paris, which still (under its changed name of Biblio-
theque Nationale) is its home, The first complete
collation of the portions of the New Testament con-
tained) in it was made) for Bentley’ ini) 7716. by
Wetstein; but this was superseded by the complete
publication undertaken by Tischendorf—of the New
Testament in 1843 and the Old in 1845. It has been
questioned whether his statements as to the various
correctors who have worked upon the MS. are always
to be trusted ; but only an expert who had had some-
thing like Tischendorf’s experience in the decipherment
of uncial MSS. can speak on such a point with any
authority ; and the precise assignment of corrections is
seldom an easy task, even apart from the special diffi-
culties which attend the decipherment of a palimpsest.

Like the three manuscripts already described, the
Codex Ephraemi originally contained the whole Greek

ΠῚ ἜΝ ΝΟΙΑΤ MANUSCRIPTS 37

Bible ; but it only survives now in a sadly mutilated
form, thanks to the disaster which befell it in the
twelfth century, when the original writing was defaced,
much of the volume thrown away, and the rest used
to receive a Greek version of some treatises by St.
Ephraem of Syria. The MS. consists now of 209
leaves, of which 64 contain portions of the Old
Testament, while 145 belong to the New. Every
book of the New Testament, except 2 Thess. and 2
John, is represented in it, but none is perfect." When
complete, the New Testament would have occupied
238 leaves. The leaves measure 12} inches by 9
inches, and are composed of good, but not especially
fine, vellum. Each page contains but one column of
writing (thus showing a step in advance since the
double columns of the Alexandrinus, though the choice
of one column or two to the page continued to some
extent open till the end of the manuscript period), in a
somewhat thick uncial hand of about the same size
and character as the Alexandrinus. The superimposed
writing having been arranged in double columns, it is
often possible to get a fairly clear view of the original
hand between the two columns; but elsewhere it is
very difficult, and often impossible, to decipher it.
From its resemblance to the Codex Alexandrinus, it
may be assigned to about the same date, namely, the
first half of the fifth century. Enlarged initials are
used, and the Eusebian (or Ammonian) sections are
marked in the margins. Tischendorf distinguishes two
correctors, one of the sixth, the other of the ninth
century, the latter being responsible for the insertion of
accents and breathings.

1 For an exact list of the contents of the MS. in its present state, so far
as relates to the New Testament, see Scrivener, i. 121, note, or Gregory,
Prolegomena, p. 367. Specimen facsimile in Omont, Manuscrits grecs de
la Bibliotheque Nationale, Pl. 111.

ssh. CRITICISM -OF NEW THES TARIENG ice,

The text of C is more mixed in character than that
of any other important MS. It belongs consistently
neither to the a-type nor to the B-type. Its transcriber
(or the transcriber of some ancestor of it) must have
had texts of both kinds before him. Consequently,
though its age makes its evidence important, as
showing what readings were extant at that early date,
it is not of much value in estimating the weight of
testimony with regard to particular readings. Hort?
notices that certain displacements in the text of the
Apocalypse show that it was copied from a MS. in
which some pages had been disarranged, and that
these pages must have been of very small size, only
equal to about a quarter of a page of C itself. It
cannot, therefore, have been a volume intended for
church or library use, but must have been a small
pocket volume, either of vellum (such as the cheap
copies of literary works which we know were written on
vellum in Martial’s day, when that material was held
in less estimation than papyrus), or, perhaps more
probably, of papyrus, such as we have reason to know,
from fragments recently discovered, existed during the
third and fourth centuries.? In either case it would
not be likely to be a copy of high authority with
regard to its text, having been intended primarily for
private use, and therefore written probably with less
attention to precise accuracy of text. How far the
same is the case with some other of our earliest copies,
we are not in a position to affirm.

D. Codex Bezae [Sod. 6 5] (Plate VI.)—the most
peculiar, and in some respects the most remarkable, of
the Greek MSS. of the New Testament. Its modern

1 Introduction, p. 268.

* E.g. the ‘‘ Logia” and St. Matthew fragments among the Oxyrhynchus
papyri; and see above, p. 39.

—" ν᾿ " τ

69 'Φ anf or
(uevnidd> pue snovualy ul pue ‘suorsi0A oidoiy1y pure sIpryes 911] ‘soynosnurut Moy Ὁ UI puv “SJ SI} Ul WaTesniafl jo
{00 Π00 911 JO UOlsloap OY} Ul pellosur Θ5ΠΈ]Ο [VUOTIPpL oY} YUM Suruuiseq ‘gl-6z “ax spy ulvjuoo uMOYs sastd ay, 18: € ayeas)

‘ANID IXIS “AVZAA XAGOD

Gs π᾿ SET
ee ῷ : ice
. “4
1
roy ielnq de tuad.by Magy LN Ota στο τερι hit OO | ee Le Pauvirouy με ινα άναν ψιάνα σον , ΄
ψινεονσἦζητα ττεσδιμαα 9) ο ρα αν αν to 4 ‘eae τὰ το 9γοιθγάλομομάοον ογάγιι ᾿
Brmvicei ion SROPYWA NL νὰ ΟΝ Ο ἐτ Ἶ Bee nom ἐν γυρεογ νυν [νον τρέοι,
Ι WE ιν ζει, τεύχει τῆν α4- αι Με town Spy : τ Z τἀ Γι εν Εἰ “οὐ ννάνιαα.
' yor peg ον τ 6) AQ? οι, α Σνηννν ἢ Ψ ᾿ οα ὐζονζυς ον ον αν ἰνὰ
τς EL SGT EM / IVS QVUOP OLWONbINP WAGS3N ἢ a ee nirodAoxanmir AwADS MOL YOO.
mg LAN py fonbts mod Wo Leino σύ tod ; QHD MH τσ OUND EVOL VOILE LV OA OL
o cletocs Sek avild Sawa tsi anbi sans ἘΠ κοννολοι yas. foot νι ραν dL
Do τσ ψΎτιΣ Abo ΌΜΥηνΝ ἡνρ νυ γα ay S Ἴ ee NYY ¥Nd vdood Loo VAVILON aula
sth OE } rays 6 to So pont pvr Sodrs σ΄, -| ρος, JWI OVNI LO VL ON
bev doe IN PWNYIsAA! WS pYUwn ds 13% “ SAO. τὰ JOYNOLNEDYVOELN Odo ΜΙ Mb Sot
στη} ‘S28 το χε 9 ΓΟ ΘΓ Ρ Ἶ > Ἰργάνο χινο αν ἀοιν αι ΓΟ Ον να
ἐπῆγεν τε γι ee OI ONY χε Του vg ΤΟ p % : Yio ΧΟ αν ONO Od doy
Dob τε τ γε ) «oa τς Swan tve.ng dion Snyavd 2 αν aI eT oVVAV EVO
Jot eo MM Sasi ϑ odd S vp is ἐν SN OS’ ΐ θκλ. λυ VAD LS VOOUNDMI
y |) SosddaNeL fay ava Tews αἴ oy γι is 2 Ao ἀγινυοννι νον νον 20 VO
pice ty bry wus SoaainyiaMuinb ΚΟ DADLA SEEN WED wUEVIAQLIOd
noo Sng wid y go¥deo namo SSS ap «ἢ ᾿ς αὐ γονν αν ον OHNE ἐν SIN Y HBA YoY
rere Sndiog .webowt No Κι 9 τ wna Ἢ be ie SEON ULE XO POD LIN PORTO EL
Py se OM fewop didNostad 19 _ oe My cide tao ΝῊ
Sosa Vad ασα αι ο yx IN Ow ASN ee 2 γχοφνονώρθάοι Nyvaryen γνἀοσοννιν.
OLINVY OALINS pl do N LW © 4 ee ᾿ : A oni γον να το
γα θα ον ἀπιΝνἀιταυνι αι Sep Sis ObonbAypni : ἢ 2 σα ζομειο Aviva ovvioiwn Ἄνν γάοι
INOLLNA DIN YY ΘΒ κε ει δι τε LIV ΝΑ IS RORDLD _ : Ι ἀν νἱγιμα οταν ἡ ἐν χϑονν NON ay
«τ 450: o-oo» ep λη 9 τὠνχήήθριρον ἡ : ἐξ ΜΗΥΌΙ ΗΜ ἀν τ γ 9.11.)
7.7 ᾿αὐϑομει Δι) ῦχγ ANSI 9 17OI lbh 1a 5 i DOOHVILO LID MOLY NAL YF
Cityo fee UDVEI OLIN YEN 9, ϑο 5. " ε MVIQXOLIN YOON OB YH AY OT
Lott a glwaonvil διγήσηρ tt Stes Yeo Sp IAG ‘4 ry γοοιν νυ QO. NBA YOLLY NAN Sho ᾿
CREE ΥΩ me 33.3 vo Bd SOLMES Mt > = * 3 Boinddo wip τυ ἊΝ ιν. 4.9 ; Ἶ
' CE bat ver ye ti > Ssinortadtittovanoayg on : [ ION ood dp tv avidin 9 : ᾿
ΐ Lehto Δ fon ἔα Κ ο γου τον ἐς - Αοα ἀγοσαοα Aolminiympdy 6}
i its) ἐς τὴν ee Δεν 9... ϑῦτην : : i~nNio1oiiawood 3413

iy 2d MQ Ly AANON ab wp say nb19 PYODINIDAHOLAYD ιϑνθθενέν JOLT

τ

΄ Ξ - . < x « 3
oR 4 ον πο γι ΙΝ SAL ον TR EE a Fei ia

TAD Ly Ta

ΠῚ ERE 'UNCLAL ‘MANUSCRIPTS 89

history begins in 1562, when Theodore Beza, the
great biblical scholar of the later Reformation period,
obtained it from the monastery of St. Irenaeus at
Lyons, as the result of the sack of that city by the
Huguenots.’ It had, however, already been collated
by some unknown scholar, by whom a large number of
_ its readings were communicated to Robert Stephanus,
which appear in the margin of his Greek Testament of
1550. Since Stephanus speaks of its having been
collated for him in Italy, it has been supposed that it
was the very ancient copy of the Greek Bible which
was taken to the Council of Trent in 1546 as evidence
in support of the reading “Si eum volo sic manere”
(ἐὰν αὐτὸν θέλω μένειν οὕτως) in John xxi. 22; but
this can hardly be regarded as certain. Beza used it
slightly in the later editions of his Greek Testament,
and in 1581 presented it to the University of Cambridge,
in whose keeping it has remained ever since. Con-
siderable use was made of it by Walton, Mill, and
Wetstein, and in 1793 a complete edition of it was
prepared and issued by Dr. Thomas Kipling for the
University of Cambridge. It was again edited by
Scrivener in 1864, and in 1899 a photographic facsimile
of the whole MS. was issued by the Cambridge University
Press. The university has thus always been forward
to allow scholars the: use of its great treasure, and to
make it readily accessible to those who may not be
within reach of its library. The peculiar character of
the MS. has naturally attracted much attention, and
within the last ten years of the nineteenth century
no less than three important studies of its text were

1 Dom Henri Quentin (Revue Bénédictine, xxiii. 1, 1906) argues that it
was already at Lyons in the ninth century, on the ground of certain readings
in the Martyrologium of Adon, written at that place, which appear to have
been taken from Codex Bezae ; but it is impossible to prove that this is the
only source whence they could have been derived.

60, CRITICISM OF NEW STESTAMEN TD. ἐμ

published by Rendel Harris, Chase, and Weiss,’ to the
contents of which we shall have to refer later.

The differences between this manuscript and the
four which have been already described begin with
its external appearance. They were all copies of the
entire Greek Bible; but there is no sign that D ever
contained the Old Testament, and in its present state
it includes only the Gospels and Acts (with considerable
mutilations, notably the conclusion of the Acts, from
ch. xxii. 29 to the end), and a small fragment of the
Catholic Epistles (3 John 11-15, in the Latin version),
which originally stood between the Gospels and Acts,
not, as usual, after the Acts. The Gospels are arranged
in the order very early adopted in the Western Church,
viz. Matthew, John, Luke, Mark. But a more striking
difference is the fact that D contains a Latin text of
the New Testament as well as the Greek. The two
versions stand side by side, on opposite pages, the
Greek holding the place of honour on the left. The
pages are somewhat smaller than those of the manu-
scripts already described, measuring 10 inches by 8.
The writing on each page occupies a single column,
but is not written in continuous paragraphs but in
κῶλα, or short clauses divided according to the sense ;
in this way the corresponding words in the two lan-
guages are kept more strictly parallel. It is written in
rather large uncials, which show a curious resemblance
between the Latin and the Greek. No less than nine
correctors have been distinguished, ranging from the
date of the MS. itself to the twelfth century or later.
The age of the original writing is not easy to

1 Rendel Harris, A Study of Codex Bezae (Cambridge, Zexts and
Studies, ii. 1, 1891); Chase, Zhe Old Syriac Element in the Text of Codex
Bezae (London, 1893); Weiss, Der Codex D in der Apfostelgeschichte
(Texte und Untersuchungen, Neue Folge ii. 1, Leipzig, 1897). Specimen
facsimile in Pal. Soc, i. 14, 15.

ΠῚ THE UNCIAL: MANUSCRIPTS ΟἹ

determine. The general opinion of palaeographers
and textual students has assigned it to the sixth
century ; but recently Professor Burkitt ( /ourn. Theol.
Stud. iii, 501, 1902) has argued for a fifth-century
date. His main reason is that the principal corrector
of the Latin side of the MS. handles the text more
freely than he supposed conceivable after about A.D.
500. This, however, is obviously insecure ground ;
the individual peculiarities of correctors cannot be tied
down so definitely. On the other hand, no palaeo-
grapher would maintain that the distinction of hands
between the last half of the fifth century and the first
half of the sixth is so clearly established as to preclude
the possibility of the earlier date for Codex Bezae. It
has not the same appearance of age as A; but even
this argument is weakened by the fact that D was
plainly written in a country of which Greek was not
the native tongue. The question cannot but remain
open until further evidence comes to light.

The existence of a Latin text is in itself sufficient
evidence that the MS. was originally written in the
West. Its presence in the south of France suggests
the possibility that this was also its first home; and in
such a country, where the Church had been founded by
missionaries from Asia Minor, who spoke Greek, but
was planted among provincials who spoke Latin, the
existence of bilingual copies of the Bible is quite
intelligible. Further, it is said that the Latin text of
Codex Bezae agrees with the Bible quotations of
Irenaeus, even in obvious errors of transcription,’ which
goes far to confirm its connexion with the Church
which Irenaeus founded. The main difficulty in the
way of this theory is the uncertainty whether bilingual
manuscripts would have continued to be produced in

1 Nestle, lrz¢voduction, Eng. Tr., p. 65.

g2 CRITICISM OF NEW TESTAMENT: πε.

southern France as late as the sixth century. It is
known that the Greek language continued in use there
as late as the time of Caesarius of Arles, who belongs
to this century, but there is no evidence that it was
still employed for liturgical purposes. Mr. Brightman,’
examining the question from the liturgical point of
view, and with reference to certain lection-marks which
have been supposed to indicate a Gallican use, rejects
such an opinion altogether, declaring the lections in
question to be of the Byzantine use, and would refer
the MS. rather to southern Italy, where the Byzantine
rite was regularly followed. There is, of course, no
difficulty in supposing a bilingual manuscript to have
been produced in Magna Graecia; on the contrary, the
chief objection to this theory is that Greek was so well
known in that region that we should have expected
the Greek part of the MS. to be better written than it
is. In point of fact, the Greek has the appearance of
having been written by a scribe whose native language
was Latin; and some of the mistakes which he makes
(Ὁ writing'\7) for X ‘er \c:for Ὁ) \poitit: in the same
direction. We want a locality where Latin was the
prevalent tongue, but Greek was still in use for eccles-
iastical purposes ; for the liturgical notes are all on the
Greek side. Southern France may have been such a
locality, but clear evidence is lacking for this particular
period ; some parts of northern Italy may have been
such localities,? but we know too little about them ;
Sardinia was such a locality, as is shown by the Codex
Laudianus (E,) having had its home, and probably its
origin, there ; but southern Italy hardly seems to answer
the requirements. The problem, therefore, of the place
of origin of D must remain for the present unsolved.

1 Journal of Theological Studies, i. 446 ff.
2 Professor Sanday, for instance, has tentatively suggested Ravenna.
y- y sugg

ΠῚ CHE -UNCIAL MANUSCRIPTS 93

The text of D differs widely from that of any other
Greek MS. of the Bible, finding its chief affinities in
the Old Syriac and Old Latin versions, described
below. It contains remarkable additions to the normal
text, such as the passage which it inserts in place of
Luke vi. 5 (τῇ αὐτῇ ἡμέρᾳ θεασάμενός τινα épyato-
μενον τῷ σαββάτῳ, εἶπεν αὐτῷ" ἄνθρωπε, εἰ μὲν
οἶδας ὃ ποιεῖς, μακάριος εἶ" εἰ δὲ μὴ οἶδας, ἐπι-
κατάρατος καὶ παραβάτης εἶ τοῦ νόμου), and a long
passage after Matt. xx. 28 (to the same general effect
as Luke xiv’ S-11)) Ati the end: of St. ihukevit) has
also a remarkable series of omissions, leaving out
Euke xxii. 20 (the\ssecond) mention of the cup in).the
institution of the Sacrament of the Lord’s Supper),
xxiv. 12 (Peter’s entry into the sepulchre), 36 (“and
saith unto them, Peace be unto you”), 40 (“and when
he had thus spoken, he showed them his hands and his
feet”), 51 (“and was carried up into heaven”), besides
minor omissions. In addition there are repeated
variants, too many to enumerate here, throughout the
text of the Gospels, while in the Acts the variations
are so frequent and so remarkable as to give rise to
the theory that it represents a different edition of the
book, though equally issued by St. Luke himself.

The discussion of this theory, and of the whole
type of text of which D is the chief Greek example,
and which we have called the 6-text, must be reserved
to a later stage, when all the data of the problem have
been stated; but there is one subject of controversy
which applies to D alone, and which must therefore be
mentioned here. This has to do with the relation
between the Greek and the Latin texts contained in it.
It is clear that these are not wholly independent of
one another, so that each would have substantial value,
the one as a copy of the Greek Testament, the other as

δ. CRITICISM OF NEW TESTANEENT ce.

a copy of the Latin version. In many small details,
such as the order of words and clauses, the texts have
evidently been assimilated to one another ; and it is a
question whether this assimilation may not extend to
some of the larger variants of which this MS. is full,
and also whether the Greek has been assimilated to
the Latin or vzce versa. The older view, held by Mill
and Wetstein, was that the Greek text had been assimi-
lated to the Latin (in which case the Greek would
cease to have independent value, and it might fairly
be questioned whether this particular type of text ever
existed in Greek at all). The great critic Griesbach,
however, maintained the contrary thesis, that the Latin
had been assimilated to the Greek ; and this view has
been generally held down to the present day, when
the earlier opinion has been revived and elaborately
defended by Mr. Rendel Harris! He points out a
number of instances in which small alterations (such as
the omission or insertion of the parts of the verb εἰμί)
have been made in the Greek, in order to produce a
more exact verbal parallelism with the Latin ; and on
the same principle he explains a considerable number
of the more important variants of the manuscript. On
the other hand, Bishop Chase” refers these variants,
or a large proportion of them, to an original Syriac
influence. He would trace back the ancestry of the
text of Codex Bezae to a bilingual Greek and Syriac
manuscript, produced probably at Antioch, where Greek
and Syriac influences met; and he explains many of
the readings of Codex Bezae as due to translations
from the Syriac. This would account for one of the
remarkable characteristics of its text, namely, the

apparently aimless substitution of synonyms for words

1 A Study of Codex Bezae (Cambridge, 1891).
2 The Old Syriac Element in the Text of Codex Bezae (London, 1893) ;
The Syro-Latin Text of the Gospels (1895).

ΠῚ THE JUNCIAL MANUSCRIPTS 95

found in the normal Greek text; such as κύριος for
θεός, or vice versa, μνημεῖον for μνῆμα (Acts il. 29),
πολλούς for ἱκανούς (xiv. 21), TO μέσον THs νυκτός for
TO μεσονύκτιον (xvi. 25), TH ἐπιούσῃ for τῇ ἑξῆς (ΧΧΙ.
I), μετεπέμψατο for μετεκαλέσατο (Xx. 17), εἶπεν for
ἀπεφθέγξατο (ii. 14), εὐαγγελίζοντε (sc. -Tat) for καταγ-
γέλλουσιν (xvi. 17), καθεζόμενος for καθήμενος (iii. 10 ;
the reverse in vi. 15), and many more of the same
type. How far the larger variants are to be explained
in the same way is a point which must be considered
later; and even if Mr. Chase’s view of the history of
the Bezan text is correct, it would not remove the
possibility of later accommodations of the Greek to
the Latin, such as Mr. Rendel Harris argues for. On
the other hand, Scrivener, whose acquaintance with the
MS. was minute and extensive, maintains that the
Latin is so sensibly accommodated to the Greek as to
be deprived of all independent value. Von Soden,’ the
latest textual critic to deal with the subject, considers
the influence of the Latin version on the Greek to be
the important characteristic of the MS.

The only possible conclusion derivable from this
conflicting testimony, where the evidence on either side
is separately convincing, is that assimilation has taken
place from do¢h sides ; and if this is so, there can be
little doubt how the phenomenon is to be explained.
The process must not be confined to the scribe of D
alone, but must be extended back to some of its
ancestors. When the New Testament was introduced
into western lands, it came in its original Greek shape,
whence it was translated into Latin for the convenience
of local converts; and when copies were made in
which this Latin version was put beside the Greek, it

ΘΕ Weiss, ae Codex D in der Afostelgeschichte, pp. 18 ff.
4 Schriften des N.T. is 1323-1340.

96 CRITICISM OF (NEW (TESTAMENT) cx.

would naturally be assimilated to it in details. But as
time went on, and such bilingual MSS. were copied by
scribes to whom Latin was more familiar than Greek,
it ‘would naturally happen that the Latin version
became the predominant partner in the combination,
and that in small details, such as the exact parallelism
of clauses, the Greek was modified so as to suit it.
Codex Bezae would therefore be a Latinising descendant
of a Graecising ancestor; and (especially if the Syriac
element be admitted as an earlier ingredient) its
unsupported evidence on purely verbal points will
not be regarded as very authoritative. The larger
variants, however, and those which are supported
by other evidence, would not be affected by these
considerations.

One other detail, pointing in the same direction,
may be noticed in conclusion. A common form of
mistake in the MS. (which is very full of scribal
errors) is the use of wrong terminations, such as τήν
for τῇ (Acts xiii. 14, xiv. 20), χωρίον for χωρίου (iv.
37), Κλαύδιος for Κλαύδιον (xviii. 2), ἕκαστος for
ἕκαστον (iii. 26), αὐτοῦ for αὐτόν (xiv. 20), αὐτῶν for
αὐτοῖς (i. 26), «.7.r.. Now such mistakes might easily
be made, if the scribe were not very careful, in copying
from a manuscript in which abbreviations were used.
Such manuscripts have been found of late years among
the papyri discovered in Egypt;* and where + stands
for τήν and τ᾽ for τῆς, αὐτὸ for αὐτός, αὐτόν, or αὐτοῦ,
av’ for any case of αὐτός, and similarly with other
words, it is easy to understand how mistakes arose
when they were re-copied. Manuscripts with such
abbreviations, however, were never meant for official
or library copies, but merely for private use; and

1 Weiss, of. 277. ΠΡ 21. 25.
2 Cf. Palacography of Greek αξγγέ, p. 32 and Appendix IV,

ΠῚ THE UNCIAL MANUSCRIPTS 97

precise purity of text is less to be expected from
them. Hence it must be reckoned as a point of
discredit to the text of D if a papyrus copy of this type
is reasonably to be suspected in its ancestry ; and it
may be added that the frequent confusion of ἡμεῖς and
ὑμεῖς (and their cases) points in the same direction,
since this blunder (odd and inconvenient as it seems)
is far from uncommon in papyri of the inferior sort.
With this warning it is time to quit Codex Bezae, and
to pass on to the consideration of the remaining MSS.
of the New Testament.

D,. Codex Claromontanus [Sod. a 1026].—Since
Codex Bezae does not contain the Pauline Epistles, the
letter D is in that group of the New Testament books
assigned to another manuscript, which contains the
Pauline Epistles and nothing else. It so happens that
the manuscript to which the designation falls has
several points of resemblance to its comrade of the
Gospels and Acts. Both are Graeco-Latin manu-
scripts; both are commonly assigned to the sixth
century ; and both belonged in the sixteenth century
to the reformer Beza. The manuscript now under
notice was acquired by him from the monastery of
Clermont, near Beauvais, in northern France, and used
in his second folio edition of the New Testament (1582).
After his death it passed through a succession of
private hands, until about the middle of the seventeenth
century it was bought by Louis XIV. for the Royal
Library at Paris, in which it still remains. Some use
was made of it by New Testament scholars after Beza,
notably by Wetstein ; and it was fully edited in 1852
by Tischendorf.* In size it is slightly smaller than
Codex Bezae, each page measuring 9? inches by 7%

1 Specimen facsimiles in Pal. Soc. i. 63, 64, and Omont, Manuscrits
grecs de la Bibl. Nat., Pl. V.

H

os CRITICISM: OF NEW TESTAMENT icy.

inches ; but the writing is considerably superior in point
of beauty. Asin Codex Bezae, the Greek text occupies
the left-hand page, the Latin the right, and the text
is divided colometrically. It contains the whole of the
Pauline Epistles, save for casual mutilations through
which a few verses here and there are lost. Before the
Epistle to the Hebrews a list of the books of the New
Testament, with the number of oréyo.* contained in
each, has been written. This list must be derived from
an archetype of very early date, since it gives the
canonical books in a very unusual order, and adds to
them certain uncanonical books. It runs as follows :—
Matthew, John, Mark, Luke, Romans, 1 and 2
Corinthians, Galatians, Ephesians, 1 and 2 Timothy,
Titus, Colossians, Philemon, 1 and 2 Peter, James, 1,
2,and 3 John, Jude, the Epistle of Barnabas, Apocalypse
of John, Acts, the Shepherd (of Hermas), Acts ‘of
Paul, Apocalypse of Peter. It will be observed that
Thessalonians, Hebrews, and Philippians are omitted,
though with regard to the latter there may be some
confusion between it and Philemon. Two leaves are

1 A στίχος, in the bibliographical sense, was the unit of literary
measure in ancient times. Originally meaning simply a line, it came to
be used to indicate the length of an average hexameter line, 2.6. about
sixteen syllables, and this was the unit of measurement by which the
labours of the copyist were reckoned. Thus when a work is described as
containing 1000 στίχοι, it does not mean that it was actually written in
1000 lines, but that it contained about 16,000 syllables. According to the
tariff of Diocletian, a copyist received 25 or 20 denarii for 100 στίχοι
(according either to quality of writing, or, as Rendel Harris thinks, to the
στίχος being taken of the length of a hexameter or an iambic line). At
this rate Mr. Rendel Harris has calculated that the cost of production of
the Codex Sinaiticus would have been about £36. In the list given in the
Codex Claromontanus, the Gospel of St Luke (the longest of the canonical
books) is reckoned at 2900 στίχοι, the Gospel of St. John at 2000, the
Epistle to the Galatians at 350, and 2 and 3 John (the shortest books) at 20
each. The principal authorities on stichometry are Graux (Revue de
Philologie, 1878, pp. 97 ff.), Diels (Hermes xvii. 377 ff.), and Rendel Harris
(Stichometry, Cambridge, 1893). The unit of a sixteen-syllable hexameter
line is expressly stated by Galen, and confirmed by the author of a sticho-

metrical catalogue of the Biblical books and the works of Cyprian in a MS,
in the Phillipps Library at Cheltenham.

ΠῚ THE UNGIAL’ MANUSCRIPTS 99

palimpsest, the superimposed text being a part of the
otherwise lost Phaethon of Euripides.

The date of the MS. can be assigned on palaeo-
eraphical grounds with fair certainty to the sixth
century, a period for which we have a datable example
in the Vienna Dioscorides." According to Tischendorf,
the hands of nine correctors can be traced in it, the
most active being D** of the seventh century and D***
of the ninth or tenth. With regard to the text, the
controversies which beset Codex Bezae fortunately do
not arise, owing to the fact that the Pauline Epistles
afford much less ground for debate than the Gospels
and Acts. But the. text of Codex, Claromontanus
belongs (as is natural in a Graeco-Latin MS.) to the
type prevalent in the West, and is akin to that of
E,F,G;, which are likewise Graeco-Latin. All probably
go back to one common archetype, the origin of which
is attributed to Italy ;? and the excellence of the Greek
hand in which it is written is a point in favour of
this attribution. Professor Souter, however, has shown
(Journ. Theol. Stud. vi. 240, 1905) that the Latin text
of D, is practically identical with that of Lucifer, Bishop
of Cagliari in Sardinia in the fourth century, and
argues with considerable show of reason that not only
D,, but also E, and perhaps even D, were produced in
Sardinia after its annexation to the Eastern Empire
in 533. The palaeographical differences are, however,
considerable. The Latin text has not been adapted
to the Greek, as in Codex Bezae, but is practically
independent, representing an early type of the Old
Latin version which preceded the Vulgate. In the
longer Pauline Epistles the text has been modified by
the influence of the Vulgate, but subsequently it is

{ 1 See above, p. 67.
2 So Corssen, the chief elaborator of this theory, approved by Sanday
(Romans, Introd. p. Ixx).

τὸν «~CRITICISM (OF NEW TESTAMENT © cx.

pure Old Latin. The consideration of the position
which this manuscript holds in the textual criticism of
the New Testament belongs to a later chapter.

E. Codex Basiliensis [Sod. ε 55].—The remaining
manuscripts of the New Testament are of less import-
ance than those which have hitherto been described,
some on account of their fragmentariness, and others
on account of their comparatively late date. The
manuscript now to be described contains the four
Gospels, except for some mutilations in St. Luke,
which have been, for the most part, supplied in a
cursive hand. Some fragments of the Apocalypse are
attached to the end of it, also in a cursive hand. It
was probably brought from Constantinople by Cardinal
John de Ragusio about 1431, and given by him to the
Dominicans in Basle, whence it passed in 1559 to the
University library of that town. It was collated for
Mill, and afterwards by Wetstein, Tischendorf, and
Tregelles, but has never been published in full. It
appears to be of the eighth century, and is written
with only one column to the page, in a rather square
and thick hand of medium size. In character its text
belongs to the a-family.

E,. Codex Laudianus [Sod. a 1001].—This is a
much more interesting manuscript than its colleague
of the Gospels. It is a Graeco-Latin copy of the Acts
in the Bodleian Library at Oxford, differing, however,
from D and D, in having the Latin text in the place
of honour on the left. This in itself points to a later
date, when the natural primacy of the Greek, as the
original language of the New Testament, had been
overcome by the greater prevalence of Latin in the
West ; and the character of the writing confirms this
supposition. It is written in a large coarse hand,
which is sometimes assigned to the latter part of the

III THE: UNCIAL MANUSCRIPTS IOI

sixth century ; but the first half of the seventh century
is perhaps a more probable date. About the end of
that century it was in England, since it is practically
certain that it was used by Bede in his commentaries
on the Acts;* and it is very probable that it was
brought to this country, like the original of the
Lindisfarne Gospels and (probably) that of the Codex
Amiatinus,’ by Theodore of Tarsus, when he came to
be Archbishop of Canterbury in 669. At an earlier
date it was in Sardinia, since the first words of an edict
of Flavius Pancratius, “Dux” of Sardinia, are written
on a fly-leaf at the end.2 Officials with this title
administered Sardinia from 534 to 749, but the precise
date of Pancratius is not known. The actual place of
origin of the MS. cannot be demonstrated; but the
margin of time between its production and its probable
arrival in England is not great enough to allow for
much travelling, in the ordinary course of things, and
Sardinia, as an island in which Greek and Latin elements
met, would be a very natural birthplace for it. When
Bede used it, it must have been in the north of
England, but its precise history is unknown until
1636, when Archbishop Laud presented it to the
Bodleian Library at Oxford. It was published in
full by Hearne in 1715, and again (more accurately)
by Hansell in 1864 and Tischendorf in 1870.4 It
contains the Acts, complete except from xxvi. 29 to
XXVill. 26, the text being arranged colometrically, as
in D and D,, but with much smaller κῶλα, often con-
sisting only of a single word. A _ peculiar chapter-

1 Professor H. J. White states that Bede certainly quotes from this MS.
in both the Zxfosz¢zo and the Liber Retractation?s.

2 See below, p. 225.

* This, and other scribblings on the same leaf, are in Greek, in hands
that might be of the late sixth or seventh century.

4 Monumenta Sacra Inedita, Nov. Coll., vol. ix. ; specimen facsimile in
Pal, Soc. i. 80.

τῸ 5 CRITICISM OF NEW TESTAMENT ‘ca.

‘division, containing fifty-eight chapters up to ch. xxvi.

24, has; been added, by a=corrector/of ‘the seventh
century...) Phe? Gatin ‘text, \ino’spite Οἱ “ἀπ: holding
what is supposed to be the place of honour, has
been accommodated to the Greek, so that it is of
little value as evidence of the Latin version of the
Acts. The Greek text, on the other hand, is of con-
siderable value, having affinities with Codex Bezae, but
not going so far as that MS. in the way of peculiar
readings. It is the earliest MS. (D being imperfect
here) containing Acts viii. 37 (the confession of faith
demanded by Philip of the eunuch before baptism),
though there is evidence that that verse was known to
Irenaeus.

E,. Codex Sangermanensis [Sod. a 1027].—In the
Pauline Epistles the letter E is assigned to this MS.,
which is again a Graeco-Latin book, but arranged with
two columns on each page, the Greek being on the left
hand. In the eighteenth century it belonged to the
abbey of St. Germain des Prés, near Paris, which was
burnt during the Revolutionary period, when many
of its books disappeared. This MS. found its way to
St. Petersburg, where it was discovered by Matthaei in
1805, and where it now is. It is written in a large,
coarse hand, said to be not unlike that of E,, but of
later type, and is assigned to the ninth or tenth
century. It is of no independent value, being a tran-
script of the Codex Claromontanus (D,), made at a
date later than that of the fifth corrector of that
MS. (92.

Ἐπὶ Codex Boreeli | Sod.¢86].—-Formerly the property
of John Boreel, Dutch Ambassador at the court of
James 1., but since 1830 in the University Library at
Utrecht. Examined by Tregelles and Tischendorf; a
full collation by Heringa, published after his death by

ΠῚ THE UNCIAL ΑΝ υ5Ο ΕΙΡΥΒ 103

Vinke in 1843, with a facsimile. The MS. originally
contained the Gospels, but it is now seriously mutilated,
beginning at Matthew ix. 1 and ending at John xiii. 34,
besides having many gaps in other places. It is written
with two columns to the page, in tall, thick uncials of
the latest type, apparently of the ninth century, though
it has also been assigned to the eighth and tenth. The
text is of the usual a-type, and the late date of the MS.
gives it comparatively little authority.

F,. Codex Augiensis [Sod. a 1029].—This is yet
another Graeco-Latin manuscript, containing the Pauline
Epistles, with a few mutilations, the principal loss being
Romans i, I-iii. 19. It belonged to the monastery of
Reichenau (Augia Dives, whence its name), on an island
in Lake Constance; and Scrivener suggests that it
may have been actually written there. At the begin-
ning of the eighteenth century it was in private hands,
and was bought by Bentley, whereby it passed into the
possession of Trinity College, Cambridge, where it still
remains. The text is written with two columns to the
page, the Greek being on the inside of each page, the
Latin on the outside. It was collated by Bentley
and Wetstein, examined by Tischendorf and Tregelles,
and published in full by Scrivener in 1859.’ It is
neatly written, the Latin better than the Greek, as was
natural in the West at the date to which it probably
belongs, which is the ninth century. The Epistle to
the Hebrews is given in Latin only. The text of this
MS. is closely connected with that of G, (see below),
and both form one group with the other Graeco-Latin
MSS., D,, E,, representing the 6-type of text for the
Pauline Epistles.

G. Codex Wolfii A [Sod. ε 87]—This MS. was
brought from the East by A. E. Seidel, and acquired

1 Specimen facsimiles in Scrivener, of. cz¢., and in Pal. Soc. i. 80.

104: ‘CRITICISM’ OF (NEW) TESTAMENT = cu:

by J. C. Wolf, who published extracts from it in 1723.
Half a leaf of it was sent by him to Bentley, and is
now among his papers in the library of Trinity College,
Cambridge ; but the bulk of it was bought by Edward
Harley, second earl of Oxford,’ and came with the rest
of the Harleian library into the British Museum, where
it). is | now. Harl...MS,. 56824. Tt «was eollated «by
Tischendorf and Tregelles. It contains the Gospels in
a mutilated state, 372 verses in all being lost, including
the beginning of Matthew as far as chap. vi. 6 (the
Cambridge fragment contains part of v. 29-43). It is
written with two columns to the page, in medium-sized
uncials of the Slavonic type, apparently of the ninth
century.

G,. Codex Boernerianus [Sod. a 1028].—This is
the companion to F,, spoken of above, and derives its
name from C. F. Boerner of Leipzig, who bought it in
1705, and lent it to Bentley, who showed his apprecia-
tion of it by refusing to part with it for five years.
Having failed to buy it, he at last returned it, and it is
now in the Royal Library at Dresden. It contains the
Pauline Epistles, with the exception of the Epistle to
the Hebrews. It was published by Matthaei in 1791,
with facsimiles. It has since been shown that it was
originally part of the same MS. as A of the Gospels
' (see below), and it may have been written at St. Gall,
where A now is. St. Gall was much frequented by
Irish monks, and some curious Irish verses have been
written on one of the pages of Gz. It is written in
a peculiar hand, probably of the ninth century, with
a Latin version between the lines; and the mistakes in

1 Gregory (Prol. p. 376, Textkritik, p. 51) questions whether the MS.
was bought by the first or the second Lord Oxford. Wanley’s Diary
shows that the first lord tried to buy it in 1722, but Wolf would not sell
at any price; and there is no further mention of it in the Diary, which
extends till 1726, two years after the first lord’s death in 1724 (not 17209,
as Gregory states).

ΠῚ THE. UNCIAL MANUSCRIPTS 105

the Greek writing show that Latin was a more familiar
language to the scribe or scribes. Besides being thus
connected: with A,)G) is also closely takin ‘to Εν
Dr. Hort was inclined to believe that F, was actually
copied from G,; but the commoner view, held by
Scrivener and Corssen, and supported by an elaborate
examination by Professor W. B. Smith,’ is that both
were taken from a common ancestor, F, being perhaps
slightly the earlier: of the: two. Im any’ case the
connexion is so close that the two manuscripts only
have the force of a single authority.

H. Codex Wolfii B [Sod. ε 88]—This manuscript
was brought from the East with G, and passed with
1 init! the /possession, of \ J.C.) Wolf, ὙΠΟ sent)a
fragment of it (as of G) to Bentley at Cambridge,
where it still remains. Its subsequent history is un-
known until 1838, when it reappeared in the Public
Library at Hamburg. It was collated by both Tischen-
dorf and Tregelles, who assign it to the ninth century.
When perfect it contained the four Gospels, written
with a single column to the page, in rather small and
ill-formed uncials; but it is now seriously mutilated,
wanting (according to Scrivener) 679 verses out of the
3780 of which the Gospels consist, the most notable
loss being Matt. i.) τ κυνὶ 30; | The’ text) 15. οὗ ‘the
a-type.

H,. Codex Mutinensis [Sod. a 6].—A ninth-century
copy of the Acts (with some defects), with the Epistles
(Catholic and Pauline) added in a cursive hand. It is
in the Grand Ducal Library at Modena, and has been
collated by Tischendorf and Tregelles, but is not of
much importance.

H,. Codex Coislinianus 202 [Sod. a 1022].—This,
on the other hand, is a very interesting manuscript,

1 American Journal of Theology, 1903, pp. 452-485, 662-668.

106 CRITICISM. OF: WEW TESTAMENT αἰ

containing an important text of the Pauline Epistles.
Originally in the monastery of the Laura on Mt. Athos,
it was there used to supply materials for the binding of
several volumes, which have since been scattered in
various parts of the world. Forty-one leaves are at
present known to exist, of which eight are still on
Mt. Athos, twenty-two at Paris, three at St. Petersburg,
three at Moscow, three at Kieff, and two at Turin.
They are written in very large, square uncials, probably
of the sixth century, the text being arranged colo-
metrically. The letters have been retraced in a dark
ink of corrosive character, which has eaten through the
vellum in many places. The text has been edited by
Omont (Wotices et extratts, vol. xxxiii. pt. I, p. 141 ff),
with two facsimiles; and a photographic facsimile of
the Athos leaves has been published by Lake (Oxford,
1905). It contains scattered portions of most of the
Pauline Epistles (but not Romans, Philippians, Ephesians,
2 Thessalonians, or Philemon). A note appended to
the Epistle to Titus (with which the MS. concludes)
states that it was corrected from the copy in the library
at Caesarea, written by the hand of the holy Pamphilus
himself (cf. p. 65, above). Further, it has been shown
that it represents (perhaps in a modified form) the
edition of the Pauline Epistles, colometrically arranged,
which we know to have been prepared, about the
middle of the fourth century, by Euthalius of Sulca,
whose work, especially as the author of a division
of the 4 Acts, and. Catholic Mpistles: ἰηΐο sections,
hasbeen mentioned above (p.. 82). Traces of the
Euthalian text (for which we also have the evidence
of some minuscule MSS., mentioned below, pp. 135-
140) have been found in the Armenian version, but
it is still uncertain whether there is any intimate
connexion between them; and the whole subject

ΠῚ THE UNCIAL MANUSCRIPTS 107

of the edition of Euthalius is still involved in much
obscurity.’

I. This letter, which was formerly assigned to twenty-
eight palimpsest leaves, from seven different MSS.,
brought by Tischendorf from the convent of Mar Saba,
between Jerusalem and the Dead Sea, is now allotted to
a MS. obtained by Mr. C. L. Freer, of Detroit, in Egypt
in December 1906. It isa much-damaged manuscript
of the Pauline Epistles, apparently of the first half of
the sixth century. Portions of all the Epistles, except
Romans, are said to be recoverable, but nothing is yet
known as to their extent or quality.

K. Codex Cyprius [Sod. ε 71]—Brought from
Cyprus in 1673; now in the Bibliotheque Nationale.
Used by Mill and Scholz; thoroughly collated by that
indefatigable pair of scholars, Tischendorf and Tre-
gelles. It is one of the nine extant complete uncial
copies of the Gospels (see p. 57); but as it is as late as
the ninth century, and contains the normal a-text, it is
not of remarkable value. It is written in compressed
uncials of a late type, rather irregular, in one broad
column to the page.

K,. Codex Mosquensis [Sod. Ατρ' and I*].—
Brought from Mt. Athos to Moscow. It contains the
Catholic and Pauline Epistles, but not the Acts, and is

1 Euthalius’ work was first edited by Zacagni (Collectanea monumentorum
veterum ecclestae, 1698). The connexion of his text with the Armenian
version has been pointed out and partially examined by Corssen, Bousset,
Rendel Harris, and Conybeare; but by far the most important treatment
of the subject is that given by Dean Armitage Robinson (7Zexts and
Studies, iil. 3), in which, among other things, he fixes for the first time
the true date of Euthalius, who had previously been assigned to the fifth
century. He also gives the text of sixteen lost pages of H;, which he
recovered from the ‘‘sets-off” left on the pages opposite to them in the
Paris and Turin leaves. Similarly, Omont and Lake have been able
to recover the text of six additional pages from the ‘‘sets-off” in the
St. Petersburg and Athos leaves. Consequently we now possess the
text of twenty-two pages in addition to the forty-one leaves actually
preserved,

ros CRITICISM. OF NEW TESTAMENT "| cx:

assigned to the ninth century. It has been collated by
Matthaei alone.

L. Codex Regius [Sod. ε 56] (Plate VII.)—This is
a manuscript of the Gospels, nearly complete, in the
Bibliotheque Nationale at Paris, where it was used by
Stephanus in the sixteenth century, and subsequently
by Wetstein, Griesbach, and Tischendorf, the latter of
whom published it in full... This honour was due, not
to its age, but to its character. It is nearly as late as
the two MSS. last mentioned, being written in com-
pressed uncials which cannot be earlier than the eighth
century ; but its text differs very markedly from the
type which had long before that date established itself
as predominant in the Church. On the contrary, it
agrees in very many places with B, and has clearly
been copied from a manuscript of the same type. It is
badly written, containing many ignorant blunders. Its
most notable feature is in regard to the conclusion of
St. Mark, where it gives, after xvi. 8, first the shorter
conclusion and then the longer (the ordinary last twelve
verses), prefixing to each a note to say that these
passages are current in some quarters, but evidently
not recognising either as authoritative. The text is
generally regarded as having an Egyptian origin.

L,. Codex Bibliothecae Angelicae A. 2. 15
[Sod. a 5].—Belongs to the Augustinian monastery at
Rome. It contains the Acts from viii. 10, the Catholic
Epistles, and the Pauline Epistles as far as Hebrews
ΧΙ, 10. It 1s assigned to the ninth century, and has
been collated by Tischendorf and Tregelles, besides
earlier scholars. A facsimile is given by Montfaucon
(Palaeographia Graeca, p. 514).

M. Codex Campianus [Sod. ε 72].—Presented to
Louis XIV. in 1706, and now in the Bibliotheque

1 Monumenta Sacra Inedita, Nov. Coll. (1846), i. 57-399, with facsimiles,

PLATE VII.

AMES OATAK COAL RS armerray -
ra eS ae Pa
ΓΔΑ AW N~ τ ἢ
eae Ee Ὁ,

ΗΓΈΡ ΤΡ Ν ΞΕΟΡῚ' ΓΕ ΑΛΛΈΜΑΤ OLE

WA, ἔπ 1A, Ε δ Τ' Orr Oe ITEPITTONTIE TOON
ὃ πον ἔ rh HISANAN (VN TORAWCE ΖΗ
TON+A ANAAYITAT'E PViAAN™~ NAETA

MT EEMTIAIEET ONCRAA Afrayraraayles
RT ALLAN AT ΟΝ ICAL 01 ATICOANATTOAHE

TW TET SW OTE ICATAX BE A, ἐένν °
OAPEIVRAAC εὶς Piece cc send
splaminenkd och TW NT Phe =
ξιςεια σον oY ft. ss ICAT A) rv inti 5
τότε * I< arty wo anf pyrMa~ HEAT : a
VRAIN NUOYCWTH ΙΑἴ «4 ΕΖ
ἡ ALE EArt OV CALE THITS See ae
γπομάποπν BADIA, 6. ad ke

εἰ μη τα.ς
ΝΑ, EAN arate:
“RACCICALE erat itr -
ICATO'Y A, ENO" AEN 22
cio fds ΝΗ

TOS AS --
88 FI1PR Daa e223,

PME TAC ASH pvt a
. inane ee 2

CODEX REGIUS; Eighth Century.

(Scale 2: 3. Shows the alternative endings to St. Mark.)

To face p.

τοῦδ.

“4

ἐν οὐ πος

Ω

eo

Ἂν»

᾿

-ς ἡ, ἐδ,

ΠῚ THE: UNCIAL MANUSCRIPTS 109

Nationale; but that it formerly had a home, and
perhaps its origin, in the East is shown by some notes
in Arabic and Slavonic which its pages contain. It is
a complete copy of the Gospels, written with two
columns to the page, in small compressed uncials of
the ninth century. It has been collated by Wetstein,
Scholz, and Tregelles, and transcribed in full by Tisch-
endorf. Its text is of the normal a-type. Facsimiles’
in Montfaucon (0p. cet. p. 260) and Silvestre (Paléo-
graphie untverselle, Pl. 76).

N. Codex Purpureus Petropolitanus [Sod. ε 10].
—A few years ago this manuscript (with the omission
of the final adjective of its title, to which it had then
no claim) might have been briefly described as consist-
ing merely of forty-five leaves of purple vellum, divided
between four different libraries (thirty-three at Patmos,
six in the Vatican, four in the British Museum, and
two at Vienna), and containing portions of each of the
four Gospels, especially St. Mark. Within the last
fifteen years, however, the position has been wholly
altered. So far back as 1883, a purple manuscript of
the Gospels was reported to have been seen in Cap-
padocia, and various travellers in the East since that
date had heard of it, and even tried to purchase it;
but nothing came of these efforts until 1896, when it
was announced that the manuscript had been secured
by the Tsar of Russia. It was conjectured almost
immediately, from the fragmentary descriptions of it
that reached this country, that it might be a further
part, if not the whole remainder, of N ; and this con-
jecture was verified when the manuscript was examined
by competent hands. By the courtesy of the authori-
ties of the Imperial Library at (St \)Petersbure, the
first publication of the newly-acquired treasure was
committed to an English scholar, Mr. H. S. Cronin, of

τὸ CRITICISM OF NEW, TESTAMENT Ἐπ.

Trinity Hall, Cambridge; and it is from his careful
and thorough edition of it that the following description
is taken.’ |

The portion of the manuscript thus acquired does
not indeed complete the whole volume, but it adds 182
leaves to the forty-five previously known, making a
total of 227 leaves. In addition, one more leaf, con-
taining John iii, 14-21, has been identified at Genoa
(Cozza-Luzi, Nova Patrum Libliotheca x. iii. 21). It
can be calculated that the original volume, when com-
plete, contained 462 leaves, so that we have even now
slightly less than half of it; but this is quite sufficient
to show its general character. All four Gospels are
substantially represented, St. Matthew proportionately
the least (Matthew forty-seven leaves, Mark forty-four,
Luke seventy-three, John sixty-three). The leaves
measure about 13 inches by 10% inches, and contain
two columns to the page, with sixteen lines in each.
The writing is in silver, with gold for the abbreviations
of the sacred names, and the letters are unusually
large. The date appears to be in the sixth century.’
Of its provenance nothing is known, but Mr. Cronin
suggests Constantinople, alike from the magnificence
of its appearance and the character of its text. Mr.
Cronin has very carefully and ingeniously worked out
the history of the MS., from such indications as its
present state provides, and concludes that it was dis-
membered about the twelfth century, possibly by
Crusaders, some of its leaves being brought to Europe,
and reaching their present homes in the sixteenth and
seventeenth centuries. The rest was again subdivided,

1 Codex Purpureus Petropolitanus, by H.'S. Cronin, M.A. (Zexts and
Studtes, vol. v. No. 4, Cambridge, 1899).

2 The oval shape of some of the letters in the prefatory matter (which
is in a different but necessarily contemporaneous hand) does not appear
to militate against this date, since it is found in other hands of the sixth
century.

ΠῚ THE UNCIAL MANUSCRIPTS Fit

part finding its way to Patmos, and part to Ephesus,
where it appears to have been seen in the eighteenth
century by the writer of a note in the similar Codex
Beratinus (Ὁ). This latter part, possibly increased by
some other leaves from the same neighbourhood, was
carried between 1820 and 1847 to Sarumsahly, the
ancient Caesarea in Cappadocia, whence it was acquired
in 1896 by the Russian Government.’

The text (as had already been gathered from the
previously extant fragments) proves to be predominantly
of the a-type, though it also shares some readings with
authorities of the opposite class. It is, in fact, a mixed
text, marking a stage in the evolution of the Textus
Receptus. In this character it is associated with the
three other purple manuscripts of the Gospels, O, Σ,
and ®, described below; and Mr. Cronin shows that
its connexion with Σ is very close indeed. Both must
have proceeded from the same workshop, and probably
they were copied from the same original. Now > con-
tains only St. Matthew and St. Mark; and as N gives
us more than half of St. Luke and St. John, we now
have by far the greater part of the Gospels in this type
of text, derived from a common original not later than
the early part of the sixth century. @® is of similar
character, but its relationship is not so close as that
between N and Σ.

O. Codex Sinopensis [Sod. ε 21].—In the previous
edition, this MS. was described as >>. It consists of
forty-three leaves of purple vellum, written throughout
in letters of gold (in which respect it is unique among
all purple uncials, with the exception of the fragment

1 Fuller details of the various dismemberments and re-collections of the
manuscript will be found in Mr. Cronin’s book. A facsimile of one of the
London leaves is given in Facsemzles from Biblical MSS. in the British
Museum, Pl. IV. (1900); and the Vatican leaves were published in
coloured facsimile by Cozza-Luzi in Omaggio Giubilare (1888).

τς CRITICISM,OF NEW, TESFAMENT co

080), and contains portions of Matthew, mainly from
chapters xiii. to xxiv., with five illustrations in the
lower margins.’ A forty-fourth leaf, containing ch.
Xviii. 9-16, is stated to be at Marioupol, on the north
of the Sea of Azof (Omont, Monuments Piot, vii. pt. 2,
1901). The hand is a large one, resembling those of
N and &, and the illustrations are of the same general
character as those of Σ. Its date is presumably the
same, namely the sixth century. It was discovered at
Sinope in 1899 by a French naval officer, and is now
in the Bibliotheque Nationale. A preliminary descrip-
tion of it, with a facsimile, was published by M. Omont
(Journal des Savants, 1900, pp. 279-285), and a full
publication of its text, with facsimiles of four of the
illustrations, is given in JVofeces et extraits des manu-
scrits de la Bibl. Nat. xxxvi. (1900) by the same
scholar. M. Omont’s collation shows that the text is
closely akin to that of N and Σ.

P. Codex Guelpherbytanus A [Sod. ε 33]—A
palimpsest in the Ducal Library at Wolfenbiittel, first
published by the discoverer, F. A; Knittel,an 1762;
re-published more fully by Tischendorf.? It contains
portions of 518 verses, all the Evangelists being
represented ; and is assigned by Tischendorf to the
sixth century. It contains a considerable number of
early readings, but agrees with the a-type oftener than
with the §-type.

P,. Codex Porphyrianus [Sod. a 3].—A palimpsest
found by Tischendorf among the MSS. of Bishop
Porphyry of Kieff, and published by him in full.’ It is

* It may be convenient to state the contents of the MS. precisely :
Matthew vii. 7°22, xi, 5: 12; Xili. 7-47). xlil. 54—-Xiv. 4, xiv. 13-20, XV.
II—xvi. 18, xvil. 2-24, xvili. 4-9, 16- -30, XIX. 3-10, 17-25, Xx. Q-Xxi. 5, xxi,
I2-xxll. 7, XXll. 15-24, XXil. 32—xxill. 35, XXiv. 3-12.

2 Mon. Sac. Ined., Nov. Coll., vol. vi. 249-338; facsimile in vol. iii,

where Cod. Q is edited (ig) gt
3 [bid, vols. v. and vi., with facsimiles.

PLATE VIII.

ne leis Nea ἰ: > ὍΡΑ. :

coor), ἔν» οὐ]... tase

rea on fh. pede lee «.

who. “5.:Ὁ}» 9] λό «. ..,
ἰὼ οἷ 5 «ραν hd a> mw ©
-ς
loos ot oF «. Ἔ
Go \orahy! fasds lo. fais leno <

:

= ὅ soy. Scie rae

Aries cee te ;
ταὶ Hau ἘΠ ΠΕΣ ον

» dof fle: γ0
δὰ iene

*

τὰ INSa
= date
: ΩΣ

—

CODEX NITRIENSIS.

A palimpsest, having Syriac writin

Sixth Century.

The

g above the original Greek.
8-45.)

page shown contains Luke xxiii. 3

(Seales 3 5:

To face p. IX.

ΠῚ THE UNCIAL MANUSCRIPTS 112

now in the Imperial Library at St. Petersburg. It is
written in sloping uncials of the ninth century, and
contains (with some mutilations) not only the Acts and
Catholic and Pauline Epistles, but also the Apocalypse,
for which it is a useful authority, uncial manuscripts
of that book being so rare that only six others are
known (NAC 046, 051, 052). It shares with 046
the representation of the later stage in the textual
tradition of that book. The upper writing, which is
cursive, written in 1301, is stated by Tischendorf to
represent the Euthalian edition of the Acts and Pauline
Epistles (see above, pp. 82, 106), and consequently has
an independent value of its own; but it has not yet
been published.

Q. Codex Guelpherbytanus B [Sod. ε 4].—This
MS. shared the fortunes of P, having been used,
together with it and with a fragment of a copy of
Ulfilas’ Gothic Gospels, to receive a later text; and it
was discovered and published by the same persons, It
contains only 247 verses, from SS. Luke and John.
It is probably of the fifth century, written in double
columns in moderate-sized uncials, Its text is of a
similar character to that of P, but with a larger infusion
of readings of the §-type.

R. Codex Nitriensis [Sod. ε 22] (Plate VIII.).—
An imperfect palimpsest copy of St. Luke, now in the
British Museum, having been brought thither from the
convent of St. Maria Deipara, in the Nitrian desert.
The upper writing is a Syriac treatise, for the tran-
scription of which the Nitrian monk had taken, not only
a copy of the Gospels, but also an equally early MS.
of the Jad and a somewhat later copy of Euclid’s
Elements. The Gospel MS. is written in a large, plain
hand, probably of the sixth century, with two columns
to the page. About half the text of St. Luke is

I

ma CRITICISM, OF NEW TESTAMENT,” ὁ ica:

preserved, in detached portions. The decipherment of
it is due to Tregelles and Tischendorf, the latter of
whom published it in full’ The text is of an early
type, belonging to the @-family rather than to its more
prevalent rival.

S. Codex Vaticanus 354 [Sod. ε 1027].—A com-
plete copy of the Gospels, mainly noticeable as bearing
a precise date, having been written by a monk named
Michael in A.D. 949.2 It is written in the large com-
pressed Slavonic characters which were the latest phase
of uncial writing, It has been collated by Birch and
Tischendorf, but its text is not of a remarkable char-
acter, being mainly of the usual a-type.

T. Codex Borgianus [Sod. ε 5]—An important
fragment in the library of the Propaganda at Rome,
consisting of seventeen leaves from the Gospels of St.
Luke and St. John, the Greek text being on the right-
hand page, the Sahidic on the left, with two columns
to the page. Giorgi, who first edited part of it in
1789, assigned it to the fourth century ; Tischendorf,
with more probability, to the fifth. Its text is decidedly
of the B-type, being ranked by Hort next after B and
in this respect.

τ. Codex Nanianus [Sod. ε 90].—In the Biblioteca
Marciana at Venice; an entire copy of the Gospels,
written with two columns to the page, in large, well-
rounded uncials, showing an attempt to revive the
hands of the fifth and sixth centuries, but without
possessing their firmness and retaining certain details
of the “Slavonic” type. It probably belongs to the
latest age of uncial writing, in the tenth century. It
was collated by Tregelles and Tischendorf; its text is
of the a-type.

1 Mon. Sac. Ined., Nov. Coll., vol. ii. 1-92. A page is reproduced in
Facsimiles of Biblical MSS. in the British Museum, P). 1X1.
Facsimile in Wew Pal. Soc., Pl. 105.

ΠῚ THE ΝΟΙΑΙ MANUSCRIPTS 1

V. Codex Mosquensis [Sod. ε 75].—A copy of the
Gospels, formerly at Mt. Athos, now at Moscow, written
in uncials of the eighth (Scrivener) or ninth (Gregory)
century as far as John vii. 39, and thence continued in
a cursive hand, which, according to Bengel, bore a date
in the year 1000, but which (the end of St. John, with
the colophon which presumably contained this date,
having been since lost) is assigned by Matthaei and
Gregory to the thirteenth century. It was collated by
Matthaei.

W. Formerly this letter included a number of small
fragments, ranging from W* to W°. It has now been
assigned a much more important function, to represent
a manuscript of the Gospels, acquired in Egypt (with I)
by Mr. C. L. Freer. It is now in course of preparation
for publication by Professor H. A. Sanders of Michigan
University. The MS. is complete (though the first
quire of St. John is in a different hand from the rest),
and is written in a small, sloping hand to which there
are few parallels, but which appears to be not later
tian the fifth century: The ‘text’ is said!) to’ be ‘of
a high quality, but nothing specific is known of it,
except that it contains after Mark xvi. 14 a remarkable
apocryphal addition, hitherto only partially known from
a reference in Jerome (Contra Pelag.ii.15). The order
of the books is Matthew, John, Luke, Mark, and the
text is said to be best in John and Luke. The MS.
cannot fail to be an important addition to our early
evidence for the Gospel text.’

X. Codex Monacensis [Sod. A*].—A considerably
mutilated copy of the Gospels at Munich, written in
small thin uncials of the tenth century, approaching the
minuscule type, and accompanied by a marginal com-

1 See H. A. Sanders, American Journal of Archaeology, 2nd series,
vol, xil., 1908 ; Gregory, Das Freer-Logion (Leipzig, 1908). Facsimile of
two pages in Mew Pal. Soc., Pl. 201.

τὸ CRITICISM: OF NEW (TESTAMENT © cu.

mentary in a contemporary minuscule hand. Collated
chiefly by Tischendorf and Tregelles. It occasionally
contains readings of an early type, akin to the β-
family.

Y. Codex Macedonianus [Sod. ε 073].—A manu-
script of the Gospels, of the ninth century, acquired by
Mr. J. B. Braithwaite in Macedonia, and described by
him in the Eafosztory Times, December 1901. The
beginning is wanting, to Matthew ix. 11. Extracts
from a collation supplied by Mr. Braithwaite are printed
by Gregory, Zertkritzk, pp. 1028-1037. The text is of
the usual a-type.

Z. Codex Dublinensis [Sod. ε 26].—An interesting
palimpsest fragment, in the library of Trinity College,
Dublin, containing less than a third of St. Matthew
(295 verses) in detached portions, underlying a cursive
text of the tenth century. It was discovered by Dr.
Barrett in 1787, and published in full by him in 1801,
and again (with additional decipherments and facsimiles)
by Dr. T. K. Abbott in 1880. It is written in large
and broad uncials, with strongly marked Egyptian
characteristics, especially in the shapes of a and μι It
is probably of the sixth century, though Dr. Abbott is
inclined to put it as high as the fifth, The text is
of the 8-type, with considerable resemblances to pn.

I. Codex Tischendorfianus IV. [Sod. ε 70]—The
Latin alphabet being now exhausted, it is necessary to
have recourse to those letters of the Greek alphabet
which differ from the Latin in form. The first of these
is a considerable fragment in the Bodleian Library,
bought out of the spoil brought back by Tischendorf
from his second campaign in the East. It consists of
158 leaves, and contains the whole of SS. Mark and
Luke (except Mark iii. 35—vi. 20), with portions of
SS. Matthew and John. A further portion of the same

ΠῚ THE UNCIAL MANUSCRIPTS "ἰὼ ὧν

MS., completing it with the exception of the hiatus in
St. Mark and a few verses of St. Matthew, was secured
by Tischendorf on his third Eastern expedition, and is
now at St. Petersburg. This latter portion contains
a colophon, stating that the MS. was completed at
a date which can probably be identified with A.D. 844.
The hand is a very thick, heavy uncial of the Slavonic
type, sloping slightly backwards. The text has been
collated by Tischendorf and Tregelles; it is usually of
the standard a-type.

A. Codex Sangallensis [Sod. ε 76]|—This manu-
script, as has been stated above, originally formed one
volume with the MS. known as G,, the latter containing
the Pauline Epistles, while A contains the four Gospels,
practically complete. It is a Graeco-Latin manuscript,
the Latin version being written between the lines of
the Greek. The frequent misdivisions of the Greek
words seem to show that it was written by a scribe
imperfectly acquainted with Greek, who was copying
from a MS. in which the words were not separated.
It was evidently written in the West, probably at
St. Gall, where it now is. The writing is rough and
ugly, of the ninth century or later. It was fully edited
in 1836 by Rettig, who was the first to identify it with
G,;. In three out of the four Gospels, the text is of
the usual a-type; but in St. Mark it has evidently
been copied from a different original, belonging to the
B-family. The Latin version is of little value, being
an amalgamation of the Old Latin and the Vulgate,
with modifications to suit the Greek.’

A. Codex Tischendorfianus III. [Sod. ε 77].—This
was brought by Tischendorf from the East with Codex

1 Cf. J. Rendel Harris, The Codex Sangallensis (A), Cambridge, 1891,
in which the Latin text is discussed. Its main value is for the relics of the

Old Latin version which can be extracted from it. A facsimile is given in
Pal. Soc. i. 179.

118 CRITICISM OF NEW TESTAMENT . cu.

I, and like it is now in the Bodleian Library at Oxford.
It contains the Gospels of SS. Luke and John complete,
written with two columns to the page in sloping
Slavonic uncials of the ninth century. There is good
reason to suppose that the earlier portion of this MS.
is at St. Petersburg, Tischendorf having deposited there
a copy of SS. Matthew and Mark (Evan. 566 in
Gregory’s list) which corresponds in size and contains
marginal matter in the same hand. Moreover the St.
Petersburg MS. has not got the subscription to St.
Mark, which is preserved at the beginning of the
Oxford part, while the subscription to St. Matthew is
of the same rather unusual type as the three in the
Oxford MS.,. .It)is true. that the, St. Petersbure. part is
written in minuscules, while the Oxford part is in
uncials ; but this combination of hands can be paralleled
by Codex E of the Septuagint, which Tischendorf
divided between Oxford, London, and St. Petersburg,
while retaining in his own hands the tell-tale leaf
containing the transition from the uncial to the cursive
hand. In the case of A, however, the identification of
the two portions is due to Tischendorf himself, who
is very likely to have known that they were derived
from the same source. Its text is said to be rather
unlike the received type, and to form one group with
the cursive MSS. numbered 20 (as revised by its
comectOr), 157, 164, 215, 2025200, 270 uazo. FOS,
686, 718, 1071, all of which have (like A) subscriptions
stating that their text was derived “from the ancient
copies at Jerusalem.”

Ξ. Codex Zacynthius [Sod. A']—A palimpsest in
the possession of the British and Foreign Bible Society
in London, containing the greater part of Luke i. 1—xi.
33, with marginal commentary. Such marginal com-
mentaries, or catenae, are common enough in later

ΠῚ THE UNCIAL MANUSCRIPTS [19

MSS., especially of the Latin Bible; but this is, with
the exception of X, the only example of one attached
to an’ uncial, text: of, the New Testament.) It) was
brought from Zante in 1821, and published by
Tregelles (with a facsimile) in 1861. It is assigned to
the eighth century. Its text is of the @-type, and it
is notable as containing a system of chapter-division,
elsewhere known only in Codex B. The upper writing
is an Evangeliarium of the thirteenth century.

II. Codex Petropolitanus [Sod. ε 73].—A nearly
complete copy of the Gospels, formerly at Smyrna,
now (through the intervention of Tischendorf) at St.
Petersburg. It is of the ninth century, and its text is
of the a-type, agreeing especially with K. Tischendorf
appears to be the only scholar who has made use of it.

>. Codex Rossanensis [Sod. ε 18]—This copy of
the Gospels is remarkable for its external appearance.
It is written in silver letters on purple-stained vellum,
and is adorned with miniatures in water-colour. The
writing is of the sixth century, and it therefore shares
with the manuscript O, mentioned above, and with the
Vienna Genesis and Dioscorides, the Cottonian Genesis
and the Vatican Virgil, the honour of being the earliest
illustrated MS. in existence (apart from Egyptian
papyri), and of representing more faithfully than later
MSS. the characteristics of classical painting. Its
present home is at Rossano in Calabria, where it was
observed in 1879 by Gebhardt and Harnack, who edited
it in full. A study of it in its artistic aspect, with
photographic reproductions of all the miniatures, was
published in 1898 by A. Haseloff,, who regards it as
exhibiting pictorial characteristics of an earlier type
than any of its competitors. It contains the Gospels

1 Codex Purpureus Rossanensis (Berlin and Leipzig, 1898). Another
reproduction, with 16 plates in colour, and including the paintings in O,
was edited by A. Muiioz in 1907.

120. CRITICISM) OF (NEW TESTAMENT © ca:

of SS. Matthew and Mark only, the others (with the
last seven verses of St. Mark) being lost. The hand
is large and square. As has been stated above, it is a
sister MS. of N, and its text is of the a-type, in a com-
paratively early stage of its development.’

@, Codex Beratinus [Sod. ε 17].—This, like N
and Σ, is a sumptuous purple and silver copy of the
Gospels ; and it is curiously like the latter of these, and
the former until quite recently, in having its home in
a comparatively obscure locality. It is at Berat, or
Belgrade, in Albania (where it has been since 1356),
and was first made known in 1868 by the archbishop
of the diocese, Anthymius Alexoudi, whose mention of
its existence led to a fuller examination of it being
made by the Abbé (now Mgr.) Batiffol in 1885,
followed by a complete edition, with facsimile, in 1886.
It consists of 190 leaves, about the same size as those
of the Codex Alexandrinus, and having, like that MS.,
two columns to the page; but the letters are much
larger. Its date is probably in the sixth century. It
contains only St. Matthew and St. Mark, and these are
slightly mutilated. A note in the MS. states that the
loss of the other two Gospels is due to “the Franks of
Champagne,” ze. probably some of the Crusaders, who
may have seen it at Patmos, where it is believed
formerly to have been. In text, like N, O, and Σ, it
belongs to the received a-type, though with some
variations ; as, for instance, the inclusion of the long
passage after Matthew xx. 28, which is also found
in D.

WV [Sod. ὃ 6].—A copy of the New Testament in
the monastery of the Laura on Mount Athos, beginning
at Mark ix. 5, and containing the remainder of the
Gospels, the Acts, the Catholic Epistles, and the Pauline

1 See Sanday, Studia Brblica, i. 103 ff.

111 THE UNCIAL MANUSCRIPTS 121

Epistles, with the exception of a leaf of Hebrews. It is
assigned to the eighth or ninth century. [{ is especially
interesting as agreeing with L in inserting the alternative
conclusion to St. Mark before the ordinary last twelve
verses. The MS. was seen by Gregory in 1886, but
the first full examination of it was made in 1899 by
Professor K. Lake, who states that the text of Mark is
far more valuable than that of Luke or John. In this
Gospel its text is emphatically of an early type, includ-
ing readings of the 8 and 6-types, without ranging itself
definitely with either family. Professor Lake is inclined
to look to Alexandria for its home, and to connect it with
the group of authorities ~xCLA and the quotations in
Clement of Alexandria, in which we find just this com-
prehensive form of ancient text.’

Q [Sod. ε 61]—A complete copy of the four
Gospels in the monastery of Dionysius on Mount Athos,
of the eighth or ninth century. It was _ cursorily
examined by Gregory in 1886, and has presumably
been collated for von Soden.

With this MS. the letters of the Greek alphabet are
exhausted. According to the new system described
above (p. 56), the numeration is now continued from
046 onwards. It is not necessary to describe the
remaining 123 uncials in detail; for the great majority
of them consists of fragments of very small size and
slight importance. They include nine fragments which
formerly were grouped under the letter I, twenty-four
which stood under Τὶ fourteen under W, eight under 0,
and ten to which Hebrew letters had been assigned.
In addition there are a number of leaves and portions
of leaves which have turned up from time to time, along
with papyri, in the rubbish-heaps of Egyptian towns,

1 See The Text of Codex Ψ in St. Mark, by the Rev. K. Lake (Journal
of Theological Studtes, i. 290, 1900).

122 CRITICISM OF NEW TESTAMENT © cx.

while von Soden has contributed brief descriptions of a
dozen MSS. (mostly late) seen by him at Damascus.
Only two deserve somewhat fuller mention here.

046. Codex Vaticanus 2066 [Sod. a 1070].—
Formerly known as B,. B being defective as regards
the Apocalypse, the letter was for that part of the New
Testament attached to another MS., also in the Vatican
Library, which contains this book alone. In the case,
however, of so important a MS. as B, it is desirable
to avoid the confusion that may arise from using its
symbol for any other MS. 046 is of the eighth
century, written in somewhat sloping uncials. It was
surreptitiously collated by Tischendorf in 1843, pub-
lished by Mai along with B in 1857, and republished
by Tischendorf in 1869. As there are only seven
uncial MSS. of the Apocalypse in existence (xACP,,
046, 051, 052), it is of some value, though less
important than xAC, and more akin to the cursive
MSS. than P,. 051 and 052 are fragmentary MSS. at
Mt. Athos, of later date and probably slight importance.

048. Codex Patiriensis [Sod. a 1]—A palimpsest
of the fifth century, formerly in the monastery of St.
Mary of Patirium, a suburb of Rossano in Calabria,
whence it was taken about the end of the seventeenth
century to the Vatican (now Cod. Vat. gr. 2061).
Here, having been lost to sight for a long time, it was
rediscovered by Mer. Batiffol in 1887.1 It contains
fragments of the Acts, Catholic Epistles, and Pauline
Epistles, written (like B) with three columns to the
page. Unfortunately its condition is so bad as to
make photographic reproduction apparently impossible.
In text it is said to be akin rather to A.

Of the remaining MSS., which are mostly very
fragmentary, as stated above, it may be sufficient to

1 Batiffol, LZ’ Addaye de Rossano, pp. 71-74 (1890).

ΠῚ THE UNCIAL, MANUSCRIPTS 123

add that two vellum leaves at Berlin and one at Vienna
and an ostrakon (or potsherd) at Athens are assigned
to the fourth century ; seventeen fragments are referred
to the fifth, and twenty-three to the seventh. Their
extreme fragmentariness deprives them of the import-
ance to which their age would entitle them; and it is "
not necessary to delay the student with further con-
sideration of them. Full details may be found in
Gregory’s Zexrtkritzk.

CHAPTER AY
THE MINUSCULE MANUSCRIPTS

[Authorities : Gregory, off. cit¢. ; Scrivener- Miller, of. cit. ; Westcott and
Hort, of. cit. ; Nestle, of. cz¢. ; von Soden, of. czt.]

THE uncial period of vellum manuscripts, as will have
been seen from the foregoing chapter, extends from
the fourth century to the tenth; but for the last two
centuries of its course it overlaps with another style of
writing, which was destined to supersede it. As the
demand for books increased, the uncial method, with its
large characters, each separately formed, became too
cumbrous. A style of writing was needed which should
occupy less space and consume less time in its pro-
duction. For everyday purposes such a style had
existed as far back as we have any extant remains of
Greek writing, and (as we have seen in Chapter II.) it
had not infrequently been employed in the transcrip-
tion of literary works ; but it never had become the
professional hand of literature, and books intended for
sale or for preservation in a library were always written
in the regular uncial hand. In the ninth century,
however, the demand for a smaller and more manage-
able literary hand was met by the introduction of a
modified form of the running hand of everyday use.
The evolution cannot yet be traced in all its details,
124

cx. iy THE MINUSCULE MANUSCRIPTS 125

but the extant specimens of non-literary hands (on
papyrus) now come down as late as the first quarter of
the eighth century ; and in these we can see all the
elements of the hand which was taken into literary use
in the ninth century, and which is commonly called
“minuscule,” as opposed to the majuscule (uncial or
capital) hands of the earlier period. In the true
minuscule hand, not only is the writing considerably
smaller than the average uncial, but the forms of the
létters) are’ different.) We have ὧν δι oy E, τσ ow,
instead of I, A, Z, M, Ξ, C, YT, with lesser variations
in the case of the other letters. Also the new forms of
the letters lend themselves to combination by means
of ligatures into a running hand, in which several
letters are written continuously, without lifting the pen ;
from which fact minuscule hands are commonly de-
scribed as “cursive,” and it is usual to classify Greek
MSS?) as°“uneials” or “ cursives.”” The, :deseription,
however, is not accurate, since not every minuscule
hand is cursive, and (on papyrus at any rate) there is
nothing to prevent uncial letters being, to some extent,
linked together by ligatures. Broadly speaking, how-
ever, and with reference to vellum MSS. alone, the
distinction holds good. Most minuscule MSS. are also
cursive, and the tendency increases as time goes on,
until we find books written in hands almost as cursive
as those of non-literary papyri.

Another change in the external characteristics of
manuscripts falls within the minuscule period, namely,
the introduction of paper by the side of vellum as the
material for the reception of writing. Though its
manufacture in Asia can be traced back to the eighth
century (in China even to the second century), and
its importation into Europe to the tenth, it was not
manufactured in this continent until the twelfth century,

126 CRITICISM OF NEW ‘TESTANENT σἂ

and does not appear in common use until the
thirteenth... Even then it does not supersede vellum.
The '\two'' materials: continue’\im | use side’ by | side
through the rest of the period during which books
were normally written by hand; and it was only
the invention of printing which finally secured the
victory of paper. During the manuscript period
the best work could only be produced on vellum ;
consequently, though copies of the Scriptures on
paper are common enough, the finer, and generally
the better, volumes are still those which are written
on the older material.

The notation of minuscule MSS. for purposes of
reference is simply numerical; and until recently
separate series were arranged, as in the case of the
uncials, for each of the four principal divisions of
the New Testament—the Gospels, Acts and Catholic
Epistles, Pauline Epistles, and Apocalypse. Thus the
first MSS. in these four groups would be known
respectively as’ Evan: 1, θα, Paul.) ty Apo!
This arrangement keeps the numbers within a more
manageable compass, since most MSS. contain only
one of these groups, and it also shows what bulk of
testimony is available for each part of the New
Testament; but it has this disadvantage, that where
a volume contains more than one of the groups,
it will appear under different numbers in each of
the groups concerned. Thus, to take an example
at random, a certain MS. in the British Museum,
which contains three groups, was known as Evan.
82. εἴ 227, Paul. σὸς | whilevanother;) which
possesses all four groups, was Evan. 584, Act. 228,
Paul.'260, Apoc. 97.

Catalogues of the known minuscule MSS. of the

1 See Thompson, Greek and Latin Palaeography, p. 43.

IV THE MINDSCULE MANUSCRIPTS: 127

New Testament are given in Scrivener’s and Gregory’s
books, which on these lines were until recently the
recognised works of reference on this subject. So far
as Evan 440; Act. 8: Paul)! 220) and) Apoe, τ
the two lists are identical, both being derived from
earlier catalogues, ultimately that of Scholz. But from
this point they diverge, Gregory having elected not to
follow the order adopted by Scrivener‘ in dealing with
the MSS. of which the existence had been made
known since the work of Scholz; and as far as Evan.
774, Act. 264, Paul. 341, Apoc. 122, the numerations
are wholly different. After this point Gregory had the
first word, the subsequent MSS. having been first cata-
logued by him; and Scrivener’s latest editor, Mr. E.
Miller, was wisely content to follow his order and
adopt his numbers as far as possible. The inconveni-
ence, however, of the quadruple system was generally
felt, and the necessity of reforming the uncial numera-
tion provided an occasion for reforming that of the
minuscules also,

In order to meet this difficulty, it has now been
agreed by nearly all the New Testament students in
Europe and America that in future each MS. shall have
one number and only one. For this purpose, Gregory’s
existing list of Gospels MSS. is retained unaltered, and
every MS. which contains the Gospels is known by the
number which it has in this list. Manuscripts which
do not include the Gospels are added at the end of the
list, or fill gaps which have accidentally occurred in it.
This is the system arranged by Gregory after con-
sultation with other scholars (see above, p. 56); it is
adopted in Souter’s Oxford Greek Testament, and will

1 Scrivener’s numbers were taken from those assigned by Dean Burgon
in a series of letters to the Guardian in July 1882, in which he described a

considerable number of MSS. which had not previously been included in
the extant lists.

128. CRITICISM: OF NEW TESTAMENT °° cu.

be used in the following pages. Side by side with it
is the system of von Soden, which has been described
above (p. 52), and in which no distinction is made
between uncials and minuscules. The earlier lists
of Gregory and Scrivener may be regarded now as
superseded."

The total number of minuscule manuscripts of the
New Testament appears in Gregory’s latest catalogue
(Die gr. Handschriften, 1908,supplemented by Tertkriizk,
Pp. 1200-10, 1373-5, 19000) as 23158. , (Of:these, 507
contain the Acts and Catholic Epistles, 595 the Pauline
Epistles, and 223 the Apocalypse.

In addition, there is a whole class of authorities
not yet mentioned, of less value than those already
described, but serving to swell the total. These are the
Lectionaries, or volumes containing the Gospels and
Epistles appointed to be read throughout the year.
They are grouped in two classes, known respectively as
Evangeliaria (Evl.)? when they contain lessons from
the Gospels, and Apostoli or Praxapostoli (Apost.)
when hey contain the Acts or. Epistles. ΤῊ ΗΠ
catalogues of these MSS., uncials and minuscules are
reckoned together ; but the uncials are never of very
early date, being almost invariably of the ninth century
or later. The number of extant lectionaries is given
by Gregory as 1565, Of these, 1G6\scontain the
Apostolos as well as the Gospels, and 215 contain the
Apostolos alone.

Adding up all the figures which have now been
presented, we get the following list of Greek authorities
for the text of the New Testament :—

1 Tables giving cross-references to these and other systems of numeration
will be found in Gregory’s Dre griechischen Hanadschriften des N. T., Leipzig,
1908.

2 Scrivener and others use the term Zvangelistarium, but this properly
means only a table of lections, not the lections themselves ; cf. Brightman,

Journal of Theological Studies, i. 448.

Iv THE MINUSCULE MANUSCRIPTS 129

Papyri ‘ : : ἱ 14
Uneials: +... : ; ; 168
Minuscules : ὲ , 2318
Lectionaries : : Ι 1565

Total 4065

These figures may not be absolutely correct in all
details, but they serve at least to give an approximate
idea of the amount of evidence available for the text
of the New Testament. Not all of it has been fully
examined hitherto, though von Soden has carried the
examination of the minuscule evidence much further
than any previous scholar. Many of the minuscule
MSS. (especially those which are still preserved in the
monasteries of the East) have been only superficially
studied ; and the lectionaries have been even more
neglected than the minuscules. Also there are, no
doubt, many MSS. in existence which have not yet
found, 2 Place in the recognised’ lists, 10 “is "nor,
however, to be expected that any considerable acces-
sion to our knowledge is to be derived from further
collection or examination of minuscule MSS. or
lectionaries, beyond what is already available. The
bulk of evidence is ample; what is now needed is
its proper organisation and digestion.

It is obviously impossible to describe all these 4000
MSS. individually ; but it may be useful to point out
those which, so far as our present knowledge goes, are
the most notable, and especially those which have
been shown to be connected with one another, and
thereby to point back to some common ancestor.
Notability, in a minuscule MS., is chiefly to be achieved
by departure from the normal type; and since the
normal type of text is that which in the last chapter
has been called the a-text, the MSS. now to be
described are those which, in a greater or less degree,

K

130 ,|CRITICISM ‘OF NEW TESTAMENT © cz.

represent the 8 or 6-texts. To select these for special
mention is not necessarily to assume the superiority of
their text; it is merely to indicate their divergence,
whether for better or for worse, from the common
norm. In a future chapter we shall have to consider
whether the truth is likely to rest with the majority or
the minority.

In the following list each number normally belongs
to one MS. and one alone; but an exception is made
in the case of the first four MSS., which were used by
Erasmus in preparing the edztzo princeps of the Greek
New Testament. These retain their earlier desig-
nations.’

I [Sod. 6 254]: an eleventh century MS. at Basle,
with illuminations. Used by Erasmus, and collated
by Wetstein and Tregelles. Has Euthalian apparatus
in Acts and Paul. Its text frequently agrees with
that of seBL, while among minuscule MSS. it is
closely connected with 209, and somewhat less closely
with 118 and 131. A study of this group has been
published by Prof. K. Lake (Zezts and Studies, Cam-
bridge, vii. 3, 1902).

1’ [Sod. Av]: twelfth century, at Mayhingen in
Bavaria. The only MS. used by Erasmus for the
Apocalypse in his edztzo princeps; and since it is
defective at the end (xxii. 16-21) this part of his
Greek text was supplied by re-translation from the
Latin, whence it has happened that some words still
hold their ground in our Textus Receptus, for which
there is no authority in any Greek MS. whatever.
Hort describes this MS. as above the average in
quality, containing a large and good ancient element.

2 (Sodje 1214]: a late MS. (isteenth century) at

1 For a conspectus of the old and new numerations, see Appendix at end
of this chapter.

IV THEI MINUSCULE MANUSCRIPTS 135

Basle, only noticeable as having formed the main basis
of the first printed edition of the New Testament, pub-
lished by Erasmus in 1516.

2°) Sed; a 25 3.1m the case of ‘this. MS.,: which
likewise belongs to the group used by Erasmus, being
his principal authority in the Acts and Epistles, the
old number is retained. It is at Basle, and has been
variously assigned to the twelfth, thirteenth, and four-
teenth centuries.

13 [Sod. e 368]: twelfth century, at Paris. This
MS. was brought into prominence by Professor W. H.
Ferrar, of Trinity College, Dublin, who showed that it
is closely connected with 69, 124, and 346.

By a comparison of these four MSS. (which from
him are generally known as “the Ferrar group”),
Professor Ferrar showed that they are descended from
one not very distant ancestor, which he held must have
been an uncial MS. of good character; and he and his
coadjutor sought to recover the text of this archetype
by a collation of its descendants. The Abbé Martin
added the observation that three at least of the four
(13, 124, 346) were written in Calabria, which must
therefore have been the home of the archetype in
the twelfth century; also that 348 (this, however, is
doubtful) and 543 are related to the same group.
Other scholars have pointed out traces of relation in
713, 788, 826, 828, and it is probable that further
investigations would lead to the identification of other
members of the family. On the other hand, Mr. E. A.
Hutton (Adas of Textual Critzcism, Cambridge, 1911)
has shown with regard to several of these that all that
is valuable in them is contained in either 69 or 124.
With regard to the text represented by this group, it

1 A Collation of Four Important Manuscripts of the Gospels, by W. H.
Ferrar and T. K. Abbott (Dublin, 1877). Professor Abbott carried on and
completed Ferrar’s work after the death of the latter in 1871.

132 CRITICISM’ OF NEW TESTAMENT cx

is clear that it is predominantly of the a-type, but it
also contains many readings of the 8 or 6-type. Mr.
Rendel Harris, on the ground of certain affinities with
the Old Syriac version (see next chapter), sought to
establish a Syriac origin for the most characteristic
readings of the group,’ while in a subsequent study ” he
argued for an Arabic medium of transmission for this
Syriac influence; but this is part of a very large
question which cannot here be discussed.

18 [Sod. 6 411]: a complete New Testament,
written in 1364, now at Paris. Only forty-six com-
plete minuscule copies of the New Testament are
known, Viz. 48; 35) O21; δοΣ ; τὴ 149/175, (280; 201,
205, 209, 218, 241, 242, 296, 339, 307, 386, 498,
FOG 517,522) 5.85. ΘΑ, G80) 600,18 24) 656;(082,
935, 986, 1072, 1075, 1094, 1352, 1384, 1503,
[507 τοῦ ἢ; 4 6206, 165 2, R008; 1678, 1704, 1.88.
2136) ‘Of these; nine . (61 0, 201,408) 5.22; Ἔ 812)
664, 680, 699) are in England or Ireland.

28 [Sod. e 168]: eleventh century, at Paris. Care-
lessly written, but containing many noticeable readings,
chiefly of the 6-type of text.

33 [Sod. 6 48]: ninth century, at Paris. Examined
by many scholars, and fully collated by Tregelles.
Its text is more of the @-type than that of any other
minuscule MS. of the Gospels. It was called by
Bichhorn:“ the queen’ of the cursives” and Hort

1 On the Origin of the Ferrar Group (Cambridge, 1893). One of the
most notable examples of Syriac affinity occurs in Matthew i. 16, where
the Ferrar archetype evidently had the same reading as the Curetonian
Syriac. See below, p. 154. Cod. 346 preserves this reading now.
Another noticeable feature of the Ferrar group is that they place the
section John vii. 53-viii. 11 after Luke xxi. 38, while Luke xxii. 43, 44 is
transferred to Matthew xxvi. 39, as is also the case in some lectionaries.

2 Further Researches into the History of the Ferrar Group (Cambridge,
1900). In this work Rendel Harris works out the Calabro-Sicilian origin
of the group in great detail, concluding that part, at least, of it is
descended from a Graeco- Arabic archetype in Sicily in the twelfth
century.

IV THE: MINUSCULE, MANUSCRIPTS 133

considered it to rank in antiquity of text next to 81
(formerly Act. 61) alone among the cursives,

54 (Bodleian Library, Oxford), 56 (at Lincoln
College, Oxford), 58 (at New College, Oxford), and 61
(at Trinity College, Dublin), the first named written in
1338, the rest in the fifteenth century at earliest, are
very closely connected, but not of special importance
except for their association with the last named. 47,
109, and 171 also belong to this group.

59 [Sod. ε 272]: twelfth century, at Caius College,
Cambridge. Contains some notable readings of the
-type.

Gi {[Sed. 6 6031: filteenth or sixteenth’ century,
at Trinity College, Dublin. This MS. is historically
important, because it was the first Greek MS. discovered
which contained the passage relating to the Three
Heavenly Witnesses (1 John v. 7, 8), and thereby was
the cause of that passage being inserted in Erasmus’
third edition (1522), in fulfilment of a pledge given to
those who criticised his omission of it in his earlier
editions. It is now known to occur also in 629,
The Apocalypse is a later addition, probably copied
from 69.

69 [Sod. ὃ 505]: fifteenth century, at Leicester,
written partly on vellum and partly on paper, the
vellum forming the inner and outer sheet in each
gathering. Collated by Tregelles and Scrivener. It
belongs to the Ferrar group (see 13, above).’ Hort
quotes it as containing many ancient readings in Acts,
in spite of its late date.

71 [Sod.¢ 253]: a well-written copy, in the Lambeth

1 See special studies of this MS. in Scrivener’s Codex Augiensts,
appendix, and Rendel Harris’ Origin of the Leicester Codex (1887). Dr.
M. R. James has shown that the scribe of this MS., and of several others
in the same hand, was a Greek named Emmanuel, from Constantinople,
who worked for George Neville, archbishop of York, about 1468 (/ozrz. of
Theol. Studies, v. 445, x1. 291, xii. 465).

134 CRITICISM OF NEW TESTAMENT cu.

Library, transcribed in A.D. 1160, and containing a
good text.

81 [Sod. a 162]: one of the most important of the
cursives, formerly known as Act. 61. Written in A.D.
1044, in a rather rough hand; now Add. MS. 20003
in the British Museum (Plate X.), Contains the Acts,
somewhat mutilated. Agrees in many respects with
the B-type of text. Hort regarded it as by far the best
of the cursives in respect of freedom from late readings,
and as containing a very ancient text. Collated by
Tischendorf (who brought it from Egypt in 1853),
Tregelles, and Scrivener.’

$8. (Sed; a) | 200] | tenth) century? \aty) Naples.
Contains Acts, Cath. Paul.,):|Apoc "Like 307) has
colophon referring to the MS. of Pamphilus at Caesarea.

104[Sod. a 103]: A.D. 1087, in the British Museum
(from Asia Minor). Contains a good text of Acts,
Cath., Paul.) Apoc,

113, 114[Sod. e 134, 1018]: two early cursives, of
the eleventh and tenth centuries respectively, now in
the British Museum, well written and with good texts.

118 .[Soed;, « 346]: thirteenth century/in ‘the
Bodleian Library at Oxford. A palimpsest, with the
Gospels uppermost. Akin to I (¢.v.).

124 [Sod. e¢ 1211]: twelfth century, at Vienna.
One of the Ferrar group (see 13).

131 [Sod. 6 467]: fourteenth or fifteenth century,
in the Vatican at Rome. Akin to 1 (g.v.). Probably
employed for the Aldine edition of the New Testament
in i518;

157 [Sod. e 207]: twelfth century, in the Vatican
Written for John 11. Comnenus (1118-43). A hand-
some copy, with a remarkable text. Hort describes it

' Facsimile in Mew. Pal. Soc., a 1793,
* Facsimile, Ζόζα. pl. 179°.

(Scale 6 :
had ᾿Ιησοῦ after πνεῦμα, as in NABC?DE, etc.

by a corrector. )

7.

LENG Gm τοῦτο n Liobanobst oO
Tage σιν rer Oy re of n- Ke

rn ket aaron: Sew
ou Ἰ ον διαΐουσιτοισ ὁ πα 5, bore
πότοσοισ EKG OLE Neve ΓῪ ὦ
τάν" ° τοί ἐλ ὁ hp HE TON
ὥστ λῦν᾽ pee eret aera
oe AST “πάρε δγδοσαρ aso ie Pu
orn che yerrarre Ki κε ents

Sra τοῦ σε arr OAM me ταῦ dee doe)

"τῶ κυνὶ εροσολύμοισ. ee
phn οἴω αἱ BK KAN of Ht esp rou
“πηι “τά Ziad kph iy
Ρ' pie Ok ME ap δὶ iA onds
hse Hpudep Kat ear lenPxety”

rn τάγιασο τοῦ Af OU THN ς΄
μιν vu roby eoTH arf e Ux

ohn sro W)\0 vo
es eave K rors payor aps: ES!

epaord artrures θυ αν ES

ov Kf cost “αὐ ow 3 σιτία ἃ

CODEX 81 (Brit. Mus. Add. MS. 20003). A.D. 1044.

The page shown contains Acts xvi. 3-9.

In verse 7 the MS. originally
The word has been partially erased

To face p. 134.

ΕΞ ΛΜ et
᾿ vy. 7 “a " = Mens
7 pers ies is - -

᾿

o
Δ

vv THE MINUSCULE MANUSCRIPTS 135

as “the best example of the few cursives which more
nearly resemble 33” in the ancient elements of its text,
“though not connected with 33 by any near affinity.”
Zahn states that its text sometimes approaches that
of Marcion. Belongs to the same group as A (g.v.).

172 [Sod. a 404]: eleventh or twelfth century, at
Berlin, formerly in the Phillipps Library at Cheltenham.
Contains Acts, Cath., Paul, Apoc. The textis said to
be valuable in the Apocalypse.

181 [Sod. a 101]: eleventh century, in the Vatican.
Contains Acts, Cath. Paul. Apoc., with the apparatus
of Euthalius to the Acts and Epistles. Used by
Zacagni as the basis of his edition of this apparatus
(see description of Cod. H;). The Apocalypse is a
later addition, of the fifteenth century.

201 [Sod. 6 403]: A.D. 1357, in the British Museum.
A large and handsome copy of the whole New Testa-
ment ; collated by Scrivener.

203 [Sod. a 203]: A.D. 1111, in the British Museum.
A handsome copy of Acts, Cath. Paul., Apoc., with
Euthalian apparatus.

205. [Sod. 6 500]: fifteenth! century, at), Vemtice.
Written for Cardinal Bessarion. Closely akin to 209,
if not copied from it. Contains both Testaments, as do
206 and 218, while 664 did so when complete.

209 [Sod. 6 457 and a 1581]: variously assigned
to the eleventh and fourteenth centuries. At Venice,
formerly the property of Bessarion. Akin to I (¢.v.),
PiSe i 31;

1 [Sod. « 234]: twelfth, century, at Venice. A
liturgical Graeco-Arabic MS., probably executed in
Calabria or Sicily, and containing much of the additional
matter often found in MSS. of the Ferrar group, but
not having itself a Ferrar text (cf. Lake, Journ. of Theol.
studies, i. 117-120);

136.» CRITICISM OF NEW TESTAMENT δ

218 [Sod. 6 300]: thirteenth century, at Vienna.
Contains both Testaments, and has many notable
readings in the New Testament. It came to Vienna
from Constantinople, and was published in full by
F, K. Alter(1 756-7 ).

223 [Sod. a 186]: eleventh or twelfth century, in
the library formed by the late Baroness Burdett-Coutts.
Contains Acts, Cath., Paul., and is said to be one of the
handsomest New Testament manuscripts in existence.

235 (Sod: €' 456]: A.D) 1314; vat Vopenhagen.
Contains many readings of the 6-type.

262 {Sod |e) 1020]: ‘tenth ‘coatury, hat / Paris.
Probably written in Italy. Belongs to the same group
as A (qg.v.), and has many noticeable readings.

274 [Sod. ε 1024]: tenth century, at Paris. Only
noticeable as containing (in the margin) the shorter
ending to St. Mark, found also in L and WV.

307. [Sod. A™™]: eleventh century, at. Paris.
Contains Acts and Cath., with the commentary of
Andreas. Has colophon))‘stating “that ityiG@e. ap
ancestor) was collated with the MS. of Pamphilus at
Caesarea (cf. x and H,).

346 [Sod. « 226]: twelfth century, at Milan. One
of the Ferrar group (see 13). In Matthew i. 16 this
MS. (alone of Greek MSS.) has substantially the same
reading as the Curetonian Syriac (see below), Ἰωσὴφ ᾧ
μνηστευθῆσα (stc) παρθένος Μαριὰμ ἐγέννησεν ᾿Ιησοῦν τὸν
λεγόμενον Χριστόν.

99. 508. ἃ 353]: thirteenth century:)im, the
Bodleian Library. Contains Acts, Cath., and Paul,
with many readings of the 6-type in Acts xiii.-xxil.
(cf. A. Pott, Der abendlindische Text der Apostelgeschichte,
Leipzig, 1900).

398 [Sod. a 189]: eleventh century, at Cambridge,
containing Acts, Cath, Paul. Used by Stephanus,

IV THE MINUSCULE MANUSCRIPTS 137

and has several notable readings in the Catholic
Epistles.

424 [Sod. O”]: A.D. 1064-68, at Vienna. Valuable
on account of some marginal readings, which Hort says
must have been taken from a MS. akin to M,', though
it cannot have been M, itself. Contains Acts, Cath.,
Paul., Apoc.

431 [Sod. ὃ 268]: twelfth to thirteenth century, in
the Catholic Seminary at Strassburg. Long supposed
(and so formerly stated by Gregory and Scrivener)
to have perished in the bombardment of 1870; see
Valentine- Richards, /ourn. of Theol. Studies, i. 608.
Its text is partly of the 6-type and partly of the @6-type,
and it appears to be akin to 614.

436 [Sod.a 172]: eleventh century, in the Vatican.
Contains Acts, Cath. Paul. Αἰδπ ἐδ 66.

460 50 ὦ 3011: eleventh century, at~ Venice,
whither it was brought from Sicily. Has a good text
of Acts, Cath., Paul., accompanied by Latin and Arabic
translations.

a46t [Sod. ¢€ 92]: A.D: 835, at St.Petersburet
Notable as the earliest dated Greek MS. on vellum in
existence, and known as the Uspensky Gospels from its
former owner, Bishop Porphyry Uspensky of Kieff.

471,472 [Sod.e 254, 1386]: two MSS. at Lambeth,
of about the twelfth century, collated by Scrivener, who
states that they have valuable readings,

496 [Sod. 6 360]: thirteenth to fourteenth century,
in British Museum. Contains Acts, Cath., and Paul.,
as well as Gospels, with several readings of the §-type.

543 [Sod. ε 257]: twelfth century, formerly in the
possession of the Baroness Burdett-Coutts. Akin to the
Ferrar group ; see description of 13.

1 Hort says M, (Introd. p. 155) ; but the previous page (as well as the
context here) shows that he means MP#"!, which is described above as M..

138 CRITICISM OF NEW TESTAMENT ‘zu.

565 1 [Sod,. je 93]. ninth) ἴον templad century, | at
St. Petersburg. A beautiful MS., written in gold letters
on purple vellum, and containing a remarkable text.
Hort (who assigned it the number 81) regarded it as
the most valuable cursive for the preservation of read-
ings of the 6-type, especially in St. Mark, which was
separately edited by Belsheim (1885) and re-collated
by Cronin (Zexts and Studies, v. 4, Cambridge, 1899).
It has the same subscriptions as A (g.v.). ,

579 [Sod::e 376]:: thirteenth) century, at, Panis,
Has the double termination to; St Mark, asin L.
This MS. has been the subject of a special study by
A. Schmidtke (Dze Evangelten etnes alten Unzzalcodex,
Leipzig, 1903) who shows that it was copied from
an uncial MS., perhaps of the sixth century, with a
text of the xB family.

614 [Sod. @: 364): veleventh century,\at. Milan
(formerly at Corfu), containing Acts, Cath, Paul. The
text is of the 6-type, and especially useful for the end
of Acts, where D, the principal Greek authority of this
type, is mutilated.

629 [Sod. a 460]: fourteenth to fifteenth century,
in the Vatican. A bilingual text of Acts, Cath., Paul,
having the Vulgate text parallel with the Greek. This
is the only MS. besides 61 which contains the passage
relating to the Three Heavenly Witnesses ; and as 61
has been sometimes suspected of having been expressly
written for the purpose, after the controversy with
Erasmus, 629 is the only unsuspected Greek witness
to the passage. As, however, it is stated that in
other passages the Greek text has been accommodated
to the parallel Latin, its authority is not great.

699 [Sod. 6 104]: tenth or eleventh century, partly
in the British Museum, and partly at Sir R. Cholmeley’s
School, Highgate, the latter portion having been

a σ΄

_ ww.

PLATE x.

2 λφάε᾽
Θλλ 45 τὸ πρῶ ovue jiev oR nas Pe :
nortan! orm ἐμαδ΄ Un Oe χορ baw ia
Masry oe byeuiin καὶ ἐπ] Tie γα δὴν Ἃ >
oP πον ἀμαὴν τον br fed σον SY Nevaiuiy
πὸ Kata wep oy ° καὶ epee tiem
πἹασὴμωγ" Καὶ χοὐτοὶ αν γέγι ἐν πόψτῇ »:
08 fAcy π᾿ ἐμῖμ' "καὶ μὰ {σὲ ψκκεκμαξ,, eS
{co mt} paspoy' καὶ {-πόν pes avy
ἘΠ. Sete ἐξ ζδιλον κὶποραήτα πὸ,
POT αὕτον μιέσογν κ-τύον καὶ 4 πτν αὐ. ΠΗ
Φιλέ. χρλσομμοιτρφσδ που. πάλη. A
Qireeneme Gye vinenk 2 |

Sev κά ν π Sat el eco Soa καὶ κεῖνοσ Es

“e ᾿ τ΄. « 32 Ἕ.
ἐ ee Ἵ Δ: )έγλποχριβ εἰσ Conn: μαμοικόσγ᾽ = 7 r,t
on 4: Ἷ Tey “4 HAH BVP OUK EKA 4 πὶ Καὶ ταῦ 3 4 εἴ

gies Daeg βάῤ εν τε περ νοὶ παν 4015 : Ἃ
ou Payajin Vacs du pas (s/° Ageou 2
δὴν. Nd Le oy agus Alo

tee Γ.
ἀπε igi

CODEX 700 (Brit. Mus. Egerton MS. 2610). Twelfth Century.

(Scale nearly 1: 1. The page shown contains Luke xi. 2-8, with the remarkable variant
in the second petition of the Lord’s Prayer, ἐλθέτω τὸ πνεῦμά σου τὸ ἅγιον ἐφ᾽

ἡμᾶς καὶ καθαρισάτω ἡμᾶς.)
To face p. 139.

IV THE MINUSCULE MANUSCRIPTS 139

bought in Epirus by the Baroness Burdett-Coutts.
A very handsome manuscript.

700 jood: ε 133]: twelfth century, in the .British
Museum (Plate X.). Contains a large number of very
interesting readings, 270 being unique, while in several
others it represents the 6-type of text. Collated by
Mr. H. C. Hoskier (A Full Account and Collation of the
Greek Cursive Codex 604 [Scrivener’s number ], London,
1890).

713 [Sod. e« 351]: eleventh to twelfth century, in
the possession of Miss A. Peckover, of Wisbech. At
one time reckoned by Mr. Rendel Harris among the
Ferrar group; but this opinion does not appear to have
been maintained.

716 [Sod. ε 448]: twelfth century, in the British
Museum. A well-written copy, with many interesting
readings.

826, 828 [Sod. « 218, 219]: both eleventh to twelfth
century, at Grotta Ferrata. Belong to the Ferrar group.
Both written in Calabria. Described by Professor K.
Lake in Journ, Theol. Studies, i, 117-120.

876 [Sod. a 356]: twelfth century, in the Burdett-
Coutts Library, containing Acts, Cath, Paul. The
text is of the 6-type.

892 [Sod. « 1016]: tenth century, in the British
Museum. Contains many remarkable readings of an
early type. Collated by J. Rendel Harris in /ournal
of Biblical Literature (1890).

1071 [Sod. ¢ 1279]: twelfth century, in the monas-
tery of the Laura on Mount Athos, but written probably
in South Italy. It has the same subscriptions as A,
etc., and is remarkable for having a text of the section
relating to the woman taken in adultery practically
identical with that in D (cf. Lake, Journ. Theol. Studczes,

1, GAT);

i40 CRITICISM OF NEW TESTAMENT ὑ ca.

1518 [Sod. a 551]: fifteenth century, formerly at
Lambeth, but given in 1817 to the Patriarch of
Terusalem, Contains” Acts, Cath.,: Paul, ts text is
akin to that of 614 and 876. Hort denotes it by the
number 112.

1836 [Sod. a 65]: tenth century, at Grotta Ferrata,
containing Acts, Cath, Paul. Has the Euthalian
apparatus.

1845 [Sod. a 64]: tenth century, in the Vatican,
with the same contents as 1836. Another Euthalian
MS. Other Euthalian MSS. are 1846 (in the Vatican)
and 1875 (at Thessalonica), both of the eleventh century.

τοῦδ᾽ “| Sod,)!(O7"" |] 2"eleventh i century, Yun othe
Bodleian (formerly on the island of Chalcé). Has
a text of the Pauline Epistles akin to that of A and B
(which, it will be remembered, are nearer to one another
in the Epistles than in the Gospels).

1914 [Sod. ΟἿ": twelfth century, in the Vatican.
Contains the Pauline Epistles. Has the Euthalian
apparatus, as also have 1916, 1962, 1970.

19057. (Sod. -@ .1574]%) fitteenth centaur, in) the
Vatican. The late supplement to B, which is defective
here. Contains Pauline Epistles and Apocalypse.

zo15 [sod. ἃ 1580]: fifteenth ‘century, 'in ‘the
Bodleian. Contains the Apocalypse only. According
to’ Scrivener, it 15. akin to’) 1084. “and :2o414! while
Gregory, following Delitzsch, states that it appears to
be copied from 2036.

2020 150. a 1573]: ‘thitteénth πε μεν, in \ the
Vatican. Contains the Apocalypse. Akin to Apoc. I,
with good readings of an early type; according to
Scrivener it closely resembles A and C.

2032 [Sod. Αν 11]: eleventh century, in the Vatican.
A fragment of the Apocalypse, with text akin to A and
to 2016 and 172.

IV THE: MINUSCULE MANUSCRIPTS tat

2036 [Sod. ἀν : fourteenth century, in the Vatican.
Contains the Apocalypse. Cf. 2015 above.

2040 [Sod. Ap"]: twelfth century, in the Parham
Library (from Mount Athos), One of the best cursive
MSS. of the Apocalypse. It contains the commentary
of Andreas (not of Arethas, as stated by von Soden).

2041 [Sod.!a:1475) fourteenth century, in) the
Parham Library (from Mount Athos). Contains the
Apocalypse. Akin to 104 and 2015.

Many of the MSS. included in the latest lists are
almost wholly unknown, though the researchers em-
ployed to collect materials for von Soden’s great
edition may have examined test passages in them.
The catalogue is greatly enlarged by the inclusion
of many MSS. in Eastern monasteries. Thus numbers
757-811 are at Athens, 922-1140, 1390-1419, 1432-
1681, 1717-1756 (besides other smaller groups) at
Athos, 1185-1256 at Sinai, 1312-1359 at Jerusalem ;
and these by no means exhaust the Eastern libraries.
But little is known of them, and so far not much that
is of importance has been brought to light. Little, too,
is known of most of the lectionaries (uncial and minus-
cule), the only one that need be particularly mentioned
being 184 (A.D. 1319, in the British Museum), which
Hort (who cites it as 39) quotes as containing a con-
siderable early element. The texts of lectionaries may
be good; but as they are comparatively late in date,
and also labour under the suspicion that their tran-
scribers might feel themselves less bound to textual
accuracy than in the case of copies of the New
Testament itself, they have generally been left on
one side by textual students. Consequently in a
summary of results, such as the present, there is little
or nothing to be said of them.

So ends the roll of direct witnesses to the text of

142. CRITICISM ΟΝ NEW TESTAMENT cx

the New Testament—of witnesses, that is, which give
the text of the sacred books in the language in which
they were written. No doubt the roll is not complete.
There are many copies of the Greek New Testament
which have not yet been brought into the recognised
lists, because they have not yet found a home in any
of the great libraries or in accessible collections of
private owners. No doubt, also, the roll as it stands
is still largely nominal. Only a small proportion of
the whole number has been fully examined; many,
especially of those which are still in Eastern libraries,
have hardly even. been cursorily ‘inspected... it: is
possible, and even probable, that among these are some
of really notable character, fit to be ranked with the
best of those which we have described above, such
as 1,) 13.) 33,260,159; 200) or -7oo;| . δεν may) be
doubted whether they can contain anything which
will add substantially to our knowledge. The minus-
cules are in themselves so late, and we have so many
authorities reaching back to an age far nearer to that
of the original autographs, that they can have but
little independent value. Their value lies in the light
which they throw on the evidence of the earlier
witnesses. They show decisively what type of text
was prevalent in the mediaeval church, while in the
case of the less common types of text they may supply
gaps or explain difficulties which the scantiness of the
uncial evidence has left. But for this purpose the
minuscules which we have already are perhaps sufficient.
Something may be accomplished in the way of grouping
them, as has been done in the case of the Ferrar MSS.,
though even this is not of much value unless the
archetype can be thrown back to a very early date.
Something, too, may be obtained by the examination
of some of these groups, as in the case of the Euthalian

IV THE MINUSCULE MANUSCRIPTS. 1438

MSS. Von Soden, in particular, has done much in the
way of classifying the great mass of authorities which
contain various forms of the a-text. The most pro-
mising outlook, however, in the department of Greek
MSS. is the possibility of the discovery of early copies
on papyrus in Egypt. The fragments which have been
found show what may yet come to light in that un-
exhausted field, and hold out to us the hope that we
may yet possess copies of the Gospels written in the
third, or even in the second century. Save for such
chances as this, it does not seem as if there could be
much added to our existing knowledge in the way of
the evidence, of Greek MSS. to the text of the, New
Testament. Some points are settled by it, some are
left unsettled. The exact bearing of this mass of
testimony, and the principles upon which it should be
treated, will be the subject of consideration in a later
chapter. But first it is necessary to pass in review
another body of witnesses, whose evidence may be
called for when that of the Greek manuscripts proves
doubtful or inadequate.

APPENDIX’ TO CHAR LER: hv
CONSPECTUS OF OLD AND NEW NUMERATION

(a2) Ln MSS. of the Gospels.

The new numeration for the Gospels is the same as the
old numeration in Gregory, except that it is not necessary to
prefix ‘‘Evan.” Scrivener’s numeration (which was followed
in the first edition) differs from that of Gregory in the
following instances :—

Scr. (GREG. Scr. GREG.
473 = 565 582 = 496
481 = 461 603 = 699

144 CRITICISM OF NEW TESTAMENT ον

500, 11 μι, 472

Scr. GREG.

556=543
561=713
565=716

Scr. GREG.
604 = 700
624 = 826
626 = 828
743=579

(ὁ) In MSS. of the Acts, Epistles, and Apocalypse.

Act.
Act.
- 9= 398
..1.3.---. 35

a5 O07

25 ΞΈΘΗ
730-60
=a Θ 1
ΠΟ = ΤΟΙ

5c ses

ΠΟΙ ΞΟῚ
ΠΟ ΞΖ ΤῸ

᾿ς 70 ἢ
85 Ξ ὃὺ
«95 = 7290

. 96= 460
137,614

2 £62:=629
78 ΞΞῚ 72
ΠΟ Ξ131
205 [acm 232)

I=I
2=2°P

= 203

Act.

216 [Scr. 184]

Ξε τὸ

Act.

223 [Scr. 220]

= 223

Act.

224 (Scr. 221]

=876

Act.

256 {ocr..231]

= 603

Act. 317 (Ser. 243]
= 1836

Act. 328 [Scr. 334]
= 1845

Paul r=

Pauli2—2>

Paul 17 =33

Pal3.— ΤΟΙ

Paul 37 =69

Paul 40=— 61

Paul 47 = 1908

Paul 07 = 424

Paul 80 = 436

Paul ol ΠΟ [Ἢ

Paul 93 = 88

Paul 10s = 200

Paul238= 431

Paul 306 [Scr. 266 + 271]

= 699

Apoc. I=I'
Apoc. 7 = 104
Apoc. 14 = 69
Apoc. 28 = 2015
Apoc. 38 = 2020
Apoc. 68 = 2032
Apoc. 79 = 2036

Apoc. 87=172
Apoc. 91 = 1957
Apoc. 92 = 61

Apoc. 95 = 2040
Apoc. 96= 2041

ΟΠ Εν

THE ANCIENT VERSIONS

(Authorities: Gregory, Prolegomena and Textkritik; Scrivener-Miller,
op. cit., chapters by H. J. White, G. H. Gwilliam, A. C. Headlam,
F. C. Conybeare, etc. ; Westcott and Hort, of. cz¢. ; Nestle, of. czt.—
On Syriac versions: Studia Biblica, i. 39 ff, 151 ff., iii. 47 ff.,
jos ft. 7 ΤΠ Hall; 7 26 Zarliest Life’ of Christ... .'\. being” the
Diatessaron of Tatian (London, 1893); Rendel Harris, 776 Diates-
saron of Tatian, a preliminary study (London, 1890); A. S. Lewis,
The Four Gospels translated from the Stnaztic Palimpsest (London,
1894); F. C. Burkitt, Avangelion da-Mepharreshe (Cambridge, 1904) ;
G. H. Gwilliam, Zetraévangelium sanctum tuxta simplicem Syrorum
verstonem (Oxford, 1901); Gwynn, Zhe Apocalypse of St. John in a
Syriac version (Dublin, 1897); A. 5. Lewis and M. D. Gibson,
Studia Sinattica, pt. vi. (1897), and Zhe Palestinian Syriac Lectionary
(1899).—Armenian: F, C. Conybeare, art. in Hastings’ Dictionary
of the Bible; J. Armitage Robinson, Zuthaliana (Cambridge, 1895).
—Coptic : Hyvernat, Etude sur les versions coptes dela Bible, in
Revue Biblique, 1896-97 ; Forbes Robinson, art. in Hastings, of. czv. ;
[G. Horner], Zhe Coptic Version of the New Testament in the northern
dialect (Oxford, 1898), and The Coptic Version of the New Testament
in the southern dialect, otherwise called Sahtdic and Thebatc, vols. 1. - 11].
(Oxford, 1911); information from Mr. W. E. Crum.—Latin: Sabatier,
Bibliorum sacrorum Latinae versiones antiguae (Paris, 1751);
Bianchini, Avangeliarium quadruplex (Rome, 1749) ; Wordsworth,
Sanday, and White, Old Latin Biblical Texts, parts i.-vi. (Oxford,
1883-1911) ; Westcott, art. ‘‘ Vulgate”? in Smith’s Dictionary of the
Bible; F.C. Burkitt, Zhe Old Latin and the [tala (Cambridge, 1896) ;
5. Berger, Azstotre de la Vulgate (Paris, 1893); Wordsworth and
White, Movum Testamentum Domini nostri lesu Chrésti Latine (vol. i.
Oxford, 1889-98; vol. 11. fasc. i. 1905); H. A. A. Kennedy, art. in
Hastings, of. czt.; P. Corssen, Bericht uber die lateinischen Bibeliiberset-
zungen, in Jahresbericht tiber die Fortschritt αἰ, class. Altertumswissen-
schaft, bd. 101 (1899) ; and works on special MSS. mentioned below. ]

145 L

146 CRITICISM OF NEW 'TESTARIENT © cu

IN the case of most ancient books, the evidence for the
establishment of the true text is exhausted when we
have come to the end of the manuscripts which contain
it in the original language; but with the Bible the
situation is wholly different. For both Testaments
evidence of the greatest value is to be derived from the
early translations which were made of them into other
tongues. The cause of this difference has nothing to
do with the distinctive character of the sacred Scriptures;
it is simply due to the fact that in the case of the
Bible we possess copies of translations which were
made as early as, or earlier than, the most ancient
existing manuscripts in the original language, while of
the secular classics we have no such ancient translations.!
No doubt this evidence has to be used with caution.
In the first place, the true text of the version itself has
to be recovered, so far as may be possible, from the
various manuscripts which contain it ; next, the original
Hebrew or Greek text represented by the translation
has to be determined, due allowance being made for
possible liberties taken by the original translator (which,
in the case of the Old Testament especially, were often
very considerable) ; then the date of the original trans-
lation must be considered, to show at what point in the
stream of tradition this branch diverged from the main
current ; finally, its relation to the other witnesses must
be discovered and the value of its testimony estimated.
All this requires the exercise of considerable knowledge
and judgment; but in spite of all difficulties and
deductions the evidence of the versions is of the very
greatest importance, and (as will be seen later) questions
connected with them are now among the most interesting
of those which demand the attention of textual students.

1 There are Latin and Arabic translations of Aristotle which are of
some textual importance, and a Greek translation of Ovid’s Herozdes ; but
these are very trivial exceptions to the general rule.

V THE ANCIENT VERSIONS 147

In the present chapter the various ancient versions
of the New Testament—all those, namely, which can
be held to be of any textual value—will be considered
in turn, and an attempt made to summarise the exist-
ing state of knowledge with regard to them. The
most natural order would seem to be geographical,
taking first the versions made in the East, in the
countries bordering upon Palestine, and afterwards
those which were made further afield, in the more
distant, but not less important, West.

1. The Syriac Versions.—Syriac or Aramaic was
the language in common use in Palestine and the
surrounding country in the time of our Lord,’ and was
naturally the first language into which a translation of
the New Testament was required. It was the language
in which our Lord Himself spoke,? and in which,
according to some very ancient authorities, the Gospel
of St. Matthew was originally written. Phrases of it,
such as TZalitha cumz or Elot, Elot, lama sabachthanz,
remain imbedded in our Greek texts, As Christianity
spread through Syria, and as the canonical Gospels
and Epistles were more and more recognised as the
authoritative records of our Lord’s life and the
apostolic teaching, a demand would naturally arise
for translations of these books into the vernacular.
At first, perhaps, such translations would be merely
oral and provisional, but before long they would be
written down for greater convenience and security ;

1 It was in use in Northern Syria from very early times, and was
adopted as their vernacular by the Jews after their return from Babylon,
where another dialect of the same language (called ‘‘ Chaldee”’ in the Old
Testament) was spoken. Cf. Neubauer in Studia Biblica, i. 39 ff., on
‘The Dialects of Palestine in the Time of Christ.” The Aramaic of
Palestine is not identical with the dialect of the versions described below,
but it is closely akin to it.

2 This has been disputed, notably by Dr. A. Roberts, who maintains

that our Lord spoke Greek ; but he does not seem to have made out his
case.

148 CRITICISM OF NEW TESTAMENT © cu.

and from this first translation or translations the
versions now in existence may trace their descent.

Five such versions are at present known, and the
questions of their inter-relation with one another are
not yet finally settled. Seventy years ago, the priority
both in age and in importance would unquestionably
have been given to the Peshitto, which may be called
the authorised version of the Syrian Church ; and there
may be still some scholars who would claim this
position for it. Recent discoveries, however, have
thrown fresh light on the problem, and of these it will
be convenient to speak first.

a. The Diatessaron of Tatian.—Little more than
thirty years ago the name of this work was a battle-
cry for controversialists of opposing schools, and its
discovery is one of the most curious, and also one of
the most important, episodes in the history of modern
Biblical criticism. It belongs, however, rather to the
“higher criticism” of the New Testament, and it will
be sufficient here to indicate briefly its bearings on
textual problems. That Tatian, an Assyrian Christian,
compiled about the year 170 a Gospel narrative, by a
process of selection and harmonisation from the four
canonical Gospels, has long been maintained by
Christian apologists, on the strength of statements by
Eusebius and other ancient writers; but the inter-
pretation of these statements was disputed, and no
vestige of the original work seemed to have been left
to establish its character, much less its precise text.
It is true that in 1836 an Armenian version of a
commentary on the Dzatessaron by Ephraem of Syria
was published by the monks of the Mechitarist
monastery in Venice; but this was a sealed book to
Western scholars until a Latin translation of it was
produced in 1876 by Dr. G. Moesinger, to which

ν THE ANCIENT VERSIONS : 149

general attention was called by Dr. Ezra Abbot in
1880. The discovery of this commentary (in which
large parts of the WJDzvatessaron itself are quoted
verbatim) not only established finally the general
character of the work, proving that it was really a
compilation from the four canonical Gospels, but also
stimulated further research, as a result of which first
one and then another copy of an Arabic translation of
the entire Dzatessaron was brought to light (one in
Rome and one in Egypt, but both now in the Vatican
Library), the text of which appeared in print in 1888
under the editorship of Ciasca.’

The opinions of scholars have varied greatly as to
the original language of the Dzatessaron. ‘The earlier
view (based chiefly on its Greek name) was that it was
written in Greek. Then its close association with the
Syriac versions and its prevalent circulation in the
Syriac Church led many scholars (for instance Zahn)
to maintain that Syriac was its original language.
Recently, however, Burkitt has reverted to the older
view, holding that it was first composed by Tatian in
Greek, though it was translated into Syriac in his life-
time, and was the first version of the Gospel in Syriac.
This view is also maintained by von Soden. Certainly,
however, the Arabic version, which is all we now possess,
was made from the Syriac, and it is not unreasonable to
believe that this was the original language of Tatian’s
compilation, which is known to have been generally used
in the churches of Syria, and to have been annotated
by the Syrian father Ephraem ; while there are signs
that the Old ‘Testament quotations in it were in

1 The recovery of the Dzatessaron has shown that the Latin Harmony
of Victor of Capua, preserved in the Codex Fuldensis (see below, p. 228),
was in fact derived from Tatian’s work; but as Victor substituted the
Vulgate text for the Old Latin which he found in his exemplar, it does
not help us to reconstitute the text of the Dzatessaron.

150 CRITICISM OF NEW TESTAMENT σε!

accordance with the Syriac version of the Scriptures,
and (as we shall see later) its text has strong affini-
ties with that which is found in other Syriac versions,
In any case it is practically certain that in the third
quarter of the second century there was in existence
a Syriac harmony of the four Gospels. It is_per-
haps not admissible to argue that this implies the
existence of a complete Syriac version of the Gospels
at this date, since Tatian may have made his own
translation from the Greek; but the possibility must
be kept in mind. In -any case the Duzatessaron, so
far as we can recover its original form, provides us
with a text which must go back to Greek MSS. of at
least the middle of the second century, and possibly
much earlier.

The original form of the Dzatessaron can, of course,
only be approximately known to us, since what we
possess is but two copies of an Arabic version, made
early in the eleventh century from a Syriac MS.
written about the year 900, together with an Armenian
version of a Syriac commentary composed by a writer
who died in 378. No doubt the text has suffered
modification since it left the hand of Tatian, and (as is
invariably the case in such matters) has been partially
assimilated to the versions which were popular at a
later date. According to Burkitt (Av. da-Meph. ii. 4)
the Arabic text is conformed to the Peshitto ; but there
is also a large proportion of the text which has a
different character, and which may fairly be held to
represent the original form of the work. This belongs
to what has been called in the last chapter the 6-type
of the Gospel text. It has affinities with B and xn,
but even more with D and its associates, and therefore
tends to support the view that these come nearer than
the a-form to the primitive text of the books of the

PLAVE Xi.

Wah y

THE CURETONIAN SYRIAC MS. Fifth Century.

(Scale 3:8. The page shown includes part of the additional passage inserted after
Matt. xx. 28 in this MS., in D&, and in many MSS. of the Old Latin version. It
also contains Matt. xx. 29-xxi. 3.)

LO Jace p. Gein

ν THE ANCIENT WERSIONS 151

New Testament. Quite recently it has been main-
tained (by von Soden) that the Diatessaron was the
main disturbing factor in the textual history of the
New Testament, as Origen’s Hexapla was in that of
the Old; but the consideration of this theory belongs
to a later chapter.’

6. The Old Syriac. —Somewhat fuller, but still very
meagre, is the knowledge that we have of the next
Syriac version which falls to be described; and this
knowledge too is of recent date. Up to 1842 the
Peshitto held the field as the earliest Syriac version of
the Bible; but in that year a large number of Syriac
manuscripts were acquired by the British Museum,
through the instrumentality of Archdeacon Tattam and
others, from the monastery of St. Mary Deipara in the
Nitrian desert in Egypt. Among these Dr. W. Cure-
ton, then assistant-keeper in the department of manu-
scripts, identified one (now Add. MS. 14451) as con-
taining a hitherto unknown version of the Gospels.
The text of this MS., edited by Cureton, was printed
and privately circulated in 1848, though not formally
published until 1858; and from him the version has
been commonly known as the Curetonian Syriac,
though in view of the discovery to be mentioned in
the next paragraph it will now be more convenient to
assign the term Curetonian to this particular MS. and
not to the version of which it is but one of the repre-
sentatives. Three additional leaves of the same version,
and probably of the same MS., were brought from the
East by Dr. Sachau to Berlin, and were edited by
Roediger in 1872.

1 For the Déatessaron, see, in addition to the authorities quoted at the
head of this chapter, Zhe Dzatessaron of Tatian, edited by S. Hemphill
(1880). A good popular account of it is given in Recent Evidence for the
Authenticity of the Gospels: Tatian’s Diatessaron, by Michael Maher, S.].
(London, Catholic Truth Society, 1893).

152 CRITICISM OF NEW TESTAMENT οἱ

Cureton’s discovery stood by itself for just fifty
years, during which time the problem of its relation to
the Peshitto was a subject of active controversy ; but
in 1892 another discovery was made which enlarged
and at the same time complicated the problem. Two
Cambridge ladies, twin-sisters, Mrs. Lewis and Mrs.
Gibson, emboldened by the success of Mr. Rendel
Harris in finding the Syriac MS. of the Ajology of Aris-
tides in the monastery of St. Catherine on Mount
Sinai, undertook an expedition to the same monastery
in search of further treasures. Among the manuscripts
which they examined was a palimpsest containing
some remarkable lives of female saints, with a lower
writing which Mrs. Lewis so far identified as to see
that it was a copy of the Gospels; but it was not
until the photographs which she took of it had been
brought to Cambridge that it was recognised by Mr.
F. C. Burkitt and Professor Bensly as belonging to
the same family as the Curetonian MS. On this a
second expedition was organised, in 1893, by Messrs.
Burkitt, Bensly, and Rendel Harris, with the two ori-
ginal discoverers, in order to make a complete tran-
script of the MS.; and as a result of this expedition
the text was published in full in 1894. A revised
edition of several pages, with many supplementary
readings and an English translation, was subsequently
published by Mrs. Lewis, after a third visit to Sinai
and a re-examination of the original. The Gospel
text is assigned to the beginning of the fifth or
even to the fourth century, being thus somewhat

1 The Four Gospels in Syriac transcribed from the Sinaitic Palimpsest,
by the late R. L. Bensly, J. Rendel Harris, and F. C. Burkitt (Cambridge,
1894); Some Pages of the Four Gospels retranscribed, by A. 5. Lewis
(London, 1896). An English version, with brief introduction, was pub-
lished by Mrs. Lewis in 1894 (Zhe Four Gospels translated from the Sinaitec
Palimpsest). Morerecently Professor Burkitt has re-edited the Curetonian

ν THE ANCIENT VERSIONS 153

older than the Curetonian MS., which is placed later
in the fifth century. The upper text is dated in the
year 778.

The Sinaitic and Curetonian MSS. are far from con-
taining identical texts, but they agree so far as to make
it certain that they derive from a common original, and
may be treated as representatives of a single version in
different stages of development. Neither is complete,
the Sinaitic containing (with local mutilations) Matthew
i. I—vi. 10, viii. 3—xvi. 15, xvii. LI—xXx. 24, xxl. 20—
xxviii. 7; Mark i. 12-44, ii. 21-iv. 17, iv. 41—-v. 26, vi.
5-xvi. 8 (where the Gospel ends); Luke i. 1-16, i. 38—
v. 28, vi. 12—xxiv. 53; John i. 25-47, ii. 16—iv. 37, v.
6-25, v. 46—xviii. 31, xix. 40—xxi. 25 ; while the Cure-
tonian has only Matthew i. I—viil. 22, x. 32—xxilil. 25 ;
Mark xvi. 17-20; Luke il. 48--111. 16, vil. 33—xvi. 12,
Xvil. I—xxiv. 44; John i. I-42, iii. 5—vilil. 19, xiv. 10-
12, 15-19, 21-23, 26-29 (Luke being, however, placed
after John). Both manuscripts differ markedly in text
from the majority of Greek MSS. and from the Peshitto
Syriac ; but this divergence is greater in the Sinaitic
than in the Curetonian, the latter appearing to repre-
sent a later stage in the history of the version, and to
be the result of a revision in which many readings were
introduced from the texts in ordinary use. In both
forms the version belongs to the 6-type of text, often
ranging itself with Codex Bezae and the old Latin
version.

Some of the more notable readings may be
mentioned. That which has aroused most controversy

MS., with a collation of the Sinaitic (ZAvangelion da-Mepharreshe ; the
Curetonian Syriac Gospels, re-edited, together with the readings of the Sinaitic
palimpsest and the early Syriac patristic evidence ; with a translation into
English, Cambridge, 1904); and this is now the standard edition of the
version as a whole.

154 CRITICISM OF NEW TESTAMENT os cu.

is in Matthew i. 16, where the Sinaitic has the remark-
able reading, “ Joseph, to whom was betrothed Mary
the Virgin, begat Jesus, who is called the Christ,” while
the Curetonian has, “ Joseph, to whom was betrothed
Mary the Virgin, who bare Jesus Christ.” At first
sight the Sinaitic text appears to deny the divine
birth of our Lord, and it is not surprising that it was
on the one hand accused of heretical leanings and on
the other claimed as representing the true original
version of the passage, which had been corrupted in all
other known copies in all languages. That this text, 27 11
denies the divine birth, cannot be original, may easily
be shown, since the context of the passage proves the
writer's knowledge of the Christian story (“ Mary che
Virgin,’ “the Christ,’ “when Mary his mother was
espoused to Joseph, when they had not come near one
to the other, she was found with child of the Holy
Ghost,” and the reference to the fulfilment of Isaiah’s
prophecy of the virgin birth) ; and the difference of the
reading from that of all other authorities’ makes it
highly improbable that it is the true form of the
text. But, in addition, good reason has been shown
for the. belief that the) words; used, in) verse » 16
were never intended to deny the divine birth at all,
the use of the word “begat” being precisely analo-
gous to its use throughout the genealogy, in which,
as is well known, it does not always indicate literal
descent, but rather an official line of succession.” The
variant reading therefore, though interesting (and
possibly coming near to the text of the original
document from which St. Matthew’s genealogy was
derived, and in which our Lord would of course be

1 Partial agreements with it are found in the Ferrar group of Greek
MSS. (see p. 132, above) and in some copies of the Old Latin.

2 Cf. Mrs. Lewis’ English translation, pp. xxili-xxv, and a paper read
by Mr. F. C. Burkitt before the Church Congress of 1895.

Vv THE ANCIENT VERSIONS 155

entered as the son of Joseph), has no important
doctrinal bearings.

Other important readings are as follows." The
Curetonian MS. inserts the names Ahaziah, Joash, and
Amaziah in Matthew i. 8 with some support from D,
but the Sinaitic agrees with the mass of authorities in
omitting them; both MSS. agree with x and B in
omitting “first-born” in i. 25, and “ bless them that curse
you, do good to them that hate you” and “ despitefully
use; \you Tine vo ΜΡ} in /viv 13) the Curetonian: MS:
(against NBD) retains the doxology to the Lord’s
Prayer, except the words “and the power ” (the Sinaitic
is defective here); both omit xii. 47 with NBL; in
xiv. 24 Cur. has “was many furlongs distant from the
land” with B and the Ferrar group (Sin. is defective) ;
Xvi. 2, 3 and xvii. 21 are omitted by both, with xB;
Xviil. II is omitted by Sin. with xBL, but retained by
Gury: in/«xix.(9 7), both readi\\“ why ‘askest. thow! me
concerning the good” with xnBDL; in xx. 22 and 23
both omit “and to be baptized with the baptism that
I am baptized with” with sBDL; Cur. agrees with D
in inserting a long additional passage after xx. 28, but
Sin. is defective here; in xxiv. 36 Sin. omits the
words “neither the Son” against xBD (Cur. is defective
here and for the rest of this Gospel); in xxvii. 16, 17
Sin. has “Jesus Barabbas” with a few minuscules and
some MSS. mentioned by Origen. Mark ix. 44, 46
are omitted by Sin. with xBCL, and the latter half of
verse 49 with NBL; also xv. 28 with nABCD. Cur.
is defective in the whole of St. Mark, except one small
fragment containing xvi. 17-20, which is sufficient to
show that it contained the last twelve verses of the

1 Syriac students will find a careful collation of both MSS. with one
another (the readings of the Peshitto being also added where they differ,
for purposes of comparison) in Mr. A. Bonus’ Collatio Cod. Lewisianz
evangeliorum Syriacorum cum Cod. Curetoniano (Oxford, 1896).

156 CRITICISM OF NEW TESTAMENT . cu.

Gospel, which Sin., like » and B, omits. In Luke ii.
14 Sin. supports the common reading εὐδοκία, not
the εὐδοκίας found in SABD and the Latin versions ;
in iv. 18 Sin. omits “to heal the broken-hearted” with
xBDL; both are defective in vi. 5, where D has its
most remarkable addition (see p. 93); in ix. 55 Cur.
has the words “and said, Ye know not what manner
of spirit ye are of,” etc., with D (partially), the minor
uncials, minuscules, and Latin versions, while Sin.
omits them with nABCL and several other uncials ; in
x. 41 Sin. omits “thou art careful and troubled about
many things” with partial support from the Old Latin
version, which has “thou art troubled” only, while
Cur. retains the whole ; in the Lord’s Prayer in xi. 2-4
Sin. omits “Our,” “which art in heaven,” “thy will be
done, as in heaven, so in earth,” “but deliver us from
evil,” with BL and (except in the third passage) x,
while Cur. omits only the third passage; both insert
Xxll. 17, 18 in verse 20, agreeing with no Greek MS. ;
Sin. omits xxii. 43, 44 (the angel in the garden and
the Bloody Sweat) with x*ABRT and the Coptic
versions, while Cur. retains them with xD and the
great mass of authorities; similarly in xxiii. 34 Sin.
omits the word from the Cross, “ Father, forgive them,”
etc. with x*BD and the Coptic versions, while Cur.
retains it; in xxiii. 48 both add “saying, Woe to us,
what hath: befallen us! woe to us for our sins,’ with
one MS of the’ Old Latin. tn) xxiv706)(12:"36 | both
retain the words which D and the Old Latin version
omit, but both agree with these authorities in omitting
verse 40; in verse 42 Sin. omits “and of a honey-
comb” with sABDL, but Cur. retains it; in verse 47
Sin. has “in my name” for “in his name,” with one
cursive (33); in verse 51 Sin. has “he was lifted up
from them,” thus agreeing neither with xD (which do

V THE ANCIENT VERSIONS 157

not expressly mention the Ascension) nor with the
other authorities (Cur. is defective to the end of the
Gospel) ; and in verses 52, 53 Sin. omits προσκυνή-
σαντες αὐτόν with D and αἰνοῦντες with RBCL. In
St. John Cur. is very defective (see above); in iii. 13
Sin. retains “which is in heaven” against NBL, and in
iv. 9 “for the Jews have no dealings with the Samar-
itans, -against) 9D; both are defective: in:)v..3, 4.5 in
vi. 69 Sin. has “thou art the Christ, the Son of God”
against NBCDL (σὺ εἶ ὁ ἅγιος τοῦ θεοῦ) ; vii. 5 3—viii.
II is omitted by both, with all the best authorities ;
Sin. also omits the last words of viii. 59, with ΝΒ.
and “the Latin versions; in xi. .39 Sinw danserts:| in
Martha’s speech the words “ Why are they lifting away
the stone?” with no other authority; in xviii. the
sequence of the narrative is altered in Sin., verse 24
being inserted between verses 13 and 14, and verses
16-18 being placed after verse 23, thus representing
Caiaphas, not Annas, as the questioner of our Lord,
and bringing together the whole narrative of Peter’s
denial. There are also many smaller variants, some
peculiar to one or both of these MSS., which can only
be realised by a full examination of their texts, either
in the original or in the published translations.

The relation of the Sinaitic and Curetonian MSS.
to one another is still obscure, and different views
have been held by different scholars. The age of the
MSS. themselves, and their frequent agreement with
the oldest Greek authorities, show at once that their
text is an ancient one; while their divergences in-
dicate that the common original from which they are
unquestionably derived must almost certainly be placed
at a considerable distance of time from them. Some
scholars have contended that the Curetonian is the
earlier form of the text, and that the Sinaitic text has

i568 CRITICISM ὌΠ (NEW TESTANENT 9 cu

been produced from it by a process of excision ; but
this view is now generally abandoned. The common
tendency in literary history (especially in sacred
writings) is in the direction of expansion rather than
of reduction; and it is now generally agreed that the
Sinaitic MS. contains the earlier form of text, and that
the Curetonian shows signs of revision and the influence
of other authorities. It does not follow that in every
case where they differ the Sinaitic preserves the
original form, while the Curetonian is a modification.
Neither MS. accurately represents the common arche-
type; and if the Curetonian has diverged from it
oftener, it still may be, and probably is, the case that
sometimes it has remained uncorrupted when the
Sinaitic has been altered.

The wider question of the relation of the version as
a whole to the other Syriac versions is best reserved
until the Peshitto has been described.

c. The Peshitto.—This is the great stanaaval version
of the Scriptures in Syriac, which has occupied among
the Syrian Churches the place held by the Vulgate in
the Roman Church or the Authorised Version in the
English. The name by which it has generally been
known since the thirteenth century means “ the simple,”
but its meaning is not clear. It has been suggested
(by Tregelles and Field) that it was originally applied
only to the Syriac Old Testament, and was intended to
distinguish it from the Hexaplar version, of which a
translation into Syriac existed and still exists, and
was thence extended to the New Testament also; but
it is hardly likely that the Hexaplar version was in
sufficiently general use to make such a differentiation
necessary or popularly known. It might, perhaps, be
intended to distinguish it from Tatian’s composite
Gospel narrative; but evidence is wanting to carry

v THE ANCIENT VERSIONS [59

back the name to a date at which the “)Ζαϊε55α7Ὸ7,
was still in common use. Prof. Burkitt’s view is that
the epithet was intended to distinguish this version,
which consisted of a simple text, from the Hexaplar
version of the Old Testament and the Harkleian of
the New, which were accompanied by an apparatus
of critical signs. The point, however, is not of great
importance, and may remain unsettled.

The history of the Peshitto version offers a marked
contrast to that which has just been described. While
the Sinaitic-Curetonian version was unknown before
1842, and now is represented only by two imperfect
and divergent manuscripts, the Peshitto has been in
public and general use for some fifteen centuries, and
exists in many manuscripts and printed volumes.
Further, while the Sinaitic-Curetonian, so far as our
present knowledge goes, contains only the four Gospels,
no copy of the other books of the New Testament
being yet forthcoming,’ or of any part of the Old
Testament, the Peshitto contains the whole of the
Old Testament except the Apocrypha, and the whole
of the New except the Apocalypse and the four minor
Catholic Epistles (2 Peter, 2 and 3 John, and Jude).
These books were omitted from the original Peshitto as
not being then recognised as canonical by the Syrian
Church, but were subsequently supplied from the later
versions, the four Epistles from the Philoxenian and
the Apocalypse from the Harkleian (see below). The
Peshitto (without these additional books) was first
edited by A. Widmanstadt at Vienna in 1555, from
two MSS., and this was long the standard text,

1 Tt is, however, fairly certain that the rest of the New Testament once
existed in this form; for Armenian translations of commentaries by St.
Ephraem on the Acts and Pauline Epistles, which have been published in
Venice, show that Ephraem used a Syriac text differing considerably from
the Peshitto (cf. J. H. Bernard in the Guardian of May 9, 1894, and J. A.
Robinson, Zuthaliana, pp. 83, 91).

160 CRITICISM OF NEW TESTAMENT ou.

although later editions were issued by Tremellius in
1569, and by several other editors in the sixteenth
and seventeenth centuries, followed by that of Leusden
and Schaaf (1708-9), which contains a collation of its
predecessors. In the last century a reprint of Widman-
stadt was issued among the Bagster Biblical texts in
1828, and a fresh edition (intended primarily for
practical use, and therefore without critical apparatus)
was published by Prof. S. Lee in 1816 for the British
and Foreign Bible Society, based upon the collation
of three additional MSS. A new critical edition was
commenced by Mr. Philip Pusey, son of the great
Hebrew professor, and was continued after his death by
the Rev. G. H. Gwilliam. This edition, of which the
first part (containing the Gospels) appeared in 1902, is
based upon an examination of forty MSS., and has at
last (so far as it goes) provided scholars with a really
critical text of the Peshitto. Mr. Gwilliam, however,
states that his text, founded though it be on authorities
much more numerous and more ancient than that of
Widmanstadt, does not differ substantially from its
predecessors, owing to the accuracy with which the
Syrian scribes have preserved the sacred text from
corruption.

The number of MSS. in which the Peshitto is
preserved, though not approaching that of the Greek
or Latin authorities, is still very considerable. The
list given by Gregory consists of 178 MSS. of the
Gospels, 74 of the Acts and Catholic Epistles, and 81
of the Pauline Epistles (the Apocalypse, as stated
above, not being contained in this version) ; and after
making allowance for MSS. which appear in more than
one of these groups, the total number of separate copies
is 243. Of these not much less than half (102) are
preserved in the British Museum, which owes most of

ὧν

᾿ "πα ἃ ον, τὶ
Pome Keyes Sid
a

PEATE ΤῚΣ

ὃς ὁ SES τόσος σαν BASES

ihe — ara ehaias witha 54
πεδίον Ὁ Kes aida RASA Kae.

Ἢ

THE PESHITTO (Brit. Mus. Add. MS. 14459). A.D. 530-540.

(Scale 7:10. The first part of this volume, containing Mt. and Mk., is of the fifth
century ; the second part, here represented, contains Lk. and Jn., and is of the sixth
century. The page shown contains Lk. ix. 46-55, including the doubtful words in
verse 55, ‘‘and said, Ye know not what spirit ye are οἵ, ᾿᾿ which are found in Danda
few other MSS., in the Curetonian, Peshitto, and Harkleian Syriac, and the Latin

versions, but not in NABCL, etc. )
To face p. 161.

Vv THE ANCIENT VERSIONS 161

its wealth in Syriac literature to the Nitrian collection,
to which we are also indebted for the Curetonian MS.

Many of these MSS. are very ancient, and some
have the advantage over their Greek coevals in being
precisely dated. The oldest (Brit. Mus. Add. MS.
14459) is assigned to the fifth century, and another
(Brit. Mus. Add. MS. 17117) is nearly contemporary
with this. At least a dozen may be referred to the
sixth century, four of them being dated in the years
530-539, 534, 548, and 586. The critical materials
now available are consequently plentiful in number
and good in character; and it is a strong testimony
to the care with which the Syrian Christians copied
their sacred books that Widmanstadt’s text, based
upon two late MSS., is now found to be so correct
in all essentials.

With regard to the date of the original composition
of this important version, it is clear at once from the
age of the most ancient MSS. that it goes back to a
very early period. This is made even more certain by
the fact that the Peshitto is the version in use among
all the branches of the Syrian Church. The secession
of the Nestorian Church took place before the middle
of the fifth century, and since that date it is certain
that neither Nestorian nor Monophysite would have
adopted his Bible from the other. The Peshitto must
consequently have been not merely in existence before
431 (the date of the Council of Ephesus), but so well
established that its position could not be shaken by
any schism in the Church. How much further it can
be carried back is a matter of controversy, in the
determination of which the question of its relation with
the Sinaitic-Curetonian version plays an important part.
Mr. Gwilliam, whose opinion, as editor of the Peshitto,
is entitled to all respect, believes it to date back to the

M

162~ CRITICISM: QF NEW) TESTAMEN Tem

second century, and to be the original translation of
the Scriptures into Syriac ;' and his view was warmly
supported by Burgon and those who follow his lead.
On the other hand, Westcott and Hort and many other
modern scholars refused to place it (at any rate in its
present shape) earlier than the third or fourth century,
and held that the Sinaitic-Curetonian version represents
a more ancient stage in the history of the Syriac New
Testament.

The controversy is an old one, since so far back as
the days of Griesbach and Hug it had been suspected
that the Peshitto was not the original form of the
Syriac version, though so long as no rival was forth-
coming the belief of these scholars was no more than a
hypothesis ; and so long as no very early manuscripts
of it were known, the phenomena which suggested it
might have been accounted for as the result merely of
local corruption and gradual revision, not of a fresh
translation. The Nitrian MSS. abolished this latter
theory, by showing that the text of the Peshitto has
come down to us practically unchanged since the fifth
century, if not earlier; and Cureton’s discovery,
strengthened as it was by that of Mrs. Lewis, quite
altered the situation, suggesting that this, or something
like this, was the pre-Peshitto version which earlier
scholars desiderated. Still, a relatively early date
seemed to be postulated for the Peshitto by the fact
that it was held to have been used by St. Ephraem, who
died in 378. A detailed examination of the facts
by Prof. Burkitt (5. Ephraem’s Quotations from the
Gospels, in the Cambridge Texts and Studies, vii. 2, 1901)

1 Cf. his article on the Syriac versions in Scrivener (ed. 4), il. 6-24, and
Studia Biblica, i. 151-174, 111. 47-104. Mr. Gwilliam’s arguments io prove
that the Curetonian shows signs of revision, and therefore is probably later
than the Peshitto, are much weakened by the discovery of the Sinaitic MS.
The Curetonian may be the result of a revision ; not, however, from the
Peshitto, but from the Sinaitic or something similar,

v THE ANCIENT VERSIONS 163

showed, however, that all the supposed instances dis-
appeared on closer scrutiny. Some vanished on an
inspection of the MSS. of the treatises in question ;
others occur in works which there is no reason to
assion to St. Ephraem. This evidence for a fourth-
century origin of the version having thus been disposed
of, another conclusion becomes not only possible but
probable. It is known that Bishop Rabbula, who was
consecrated to the see of Edessa in 411, translated
the New Testament into Syriac, and ordered a copy of
it to be placed in every church. If this translation is
not the Peshitto, no trace of it has survived ; and since
it is now demonstrated that no quotations from the
Peshitto are found in writers earlier than Rabbula,
while after his date they occur regularly, the conclu-
sion is obvious, and is now generally accepted, that
Rabbula’s version was the Peshitto itself, which accord-
ingly belongs to the early years of the fifth century.
This conclusion is further fortified by the fact that the
Armenian version, which was made about 400 from
Syriac and Greek codices, proves to have a consider-
able element of kinship with the Sinaitic-Curetonian
fext.’

It may now, therefore, be considered as established
that the Sinaitic-Curetonian version can claim priority
over its better-known and more widely used rival, and
may rightly be called the Old Syriac. Its very rough-
ness is an additional proof of its antiquity; for it is
difficult to believe that it would ever have been pro-
duced if the Peshitto were already in existence. To
revise the Peshitto into anything like the Sinaitic-Cure-
tonian form would be a very remarkable literary per-
formance ; but the contrary process is quite conceiv-

1 Cf. Conybeare’s article in Scrivener, ii, 148-154, and Dean Armitage
Robinson’s Huthaliana ( Texts and Studies, iii. 3), pp. 72-98. ,

164 CRITICISM OF NEW TESTAMENT = @.

able. At the same time it would be going too far to
say that the Peshitto is merely a revision of the Old
Syriac, as our English Authorised Version is a revision
of the Bishops’ Bible, or the Revised Version of the
Authorised. A connexion between the two there
certainly is, so that the author (or authors) of the
Peshitto must have been acquainted with the Old
Syriac ;* but he must have used it in an independent
manner, rather as an assistance than as the basis of his
work. For one thing, he must have used Greek MSS.
of a different family from that which is represented by
the Old Syriac. This, as we have seen, belongs to the
6-type, agreeing mainly with D and the Old Latin,
and often also with xB; while the Peshitto ranges
itself rather with the authorities of the a-type. It does
not, however, contain the a-text pure and simple, as we
find it in the great mass of later Greek manuscripts,
but has a considerable intermixture of readings char-
acteristic of the 8 or 6-texts. The explanation of this
depends on the general view that we take of the textual
history of the New Testament, which will be the sub-
ject of a later chapter. If the a-text is the more
authentic, the Peshitto has been to some extent con-
taminated by erratic influences; but if the #-text
comes nearer to the original, while the a-text is the
result of a later revision, then the Peshitto represents
an intermediate stage in the process, though nearer to
the latter than to the former.

ad. The Philoxenian and Harkleian Syriac.—The
history of the remaining Syriac versions is much plainer
and more certain. In the year 508 (as we know from
the colophon appended to the Gospels in the MSS. of
this version) a fresh translation of the New Testament

1 Cf. Gwilliam in Scrivener, ii. 16, and Burkitt in Excyclopaedia Biblica,
iv. 5002, and Zvangelion da-Mepharreshe, Introduction.

v THE ANCIENT VERSIONS 165

into Syriac was prepared for Philoxenus, Jacobite
bishop of Mabug or Hierapolis in eastern Syria, by one
Polycarp, a χωρεπίσκοπος. Of this translation, how-
ever, as originally made, very little has come down to us;
a thorough-going revision of it having been made about
a century later by Thomas of Harkel, himself subse-
quently bishop of Mabug. This revision, which in-
volved a double process of collation of the original with
Greek MSS. at Alexandria, was completed (as the same
colophons inform us) in 616. Until recently the char-
acteristics of these two stages in the development of
the version, the Philoxenian and the Harkleian, were
little known, and the version as a whole had been
little studied ; but within the present generation some
of this obscurity has been removed. Before 1892 the
only part of the New Testament which could definitely
be assigned to the original Philoxenian version was the
four minor Catholic Epistles (2 Peter, 2 and 3 John,
and Jude), which had been taken from this version to
supply the omission in the Peshitto; and these hardly
provided sufficient material for a judgment on the
version as a whole. In that year, however, Dr. Gwynn,
of Trinity College, Dublin, published an account of a
twelfth-century manuscript belonging to Lord Crawford
(now inthe John Rylands Library at Manchester), con-
taining a hitherto unknown version of the Apocalypse,
which he showed conclusively to be Philoxenian.’
From this it is clear that the Philoxenian version was
written in free and idiomatic Syriac, being the most
literary in form of all the translations of the New
Testament into this language. The Greek text under-
lying it was that of the great mass of later MSS.,
which (as is abundantly clear from other evidence

1 The text of this MS. was published by Dr. Gwynn in 1897 (Zhe
Apocalypse of St. John in a Syriac Version, Dublin), with an introductory
dissertation on the various Syriac versions.

166 CRITICISM OF NEW TESTAMENT | cz.

also) was firmly established as the standard type of text
in the Greek-speaking Church at the time when Polycarp
prepared this version of the Scriptures for Philoxenus.
Possibly it was this very freedom of treatment that
led to the revision of the translation after little more
than a century. At any rate it is certain that when
Thomas of Harkel put his hand to the task, he did so
in a spirit wholly opposed to that of his predecessor.
The free rendering of Polycarp was converted into
extreme literalness, the Syriac being forced, even to the
extent of doing violence to the language, into precise
accordance with the Greek. The Harkleian version of
the New Testament may, in fact, be compared with that
made by Aquila of the Greek Old Testament. One
advantage there is in this method of treatment, from
the textual critic’s point of view, that it is usually easy to
determine what the exact form of Greek was that the
translator meant to represent. From this it appears
that Thomas modified the work of his predecessor, not
only in fidelity but in text, having evidently used
Greek MSS. of a type akin to Codex Bezae (D) and its
fellows. Additional readings of the same type are noted
in the margin of this version, which may perhaps be
the result of the second stage of Thomas’ revision.
With regard to the Gospels and Pauline Epistles
little attempt has yet been made to distinguish the
Philoxenian and Harkleian elements in the text which
has come down to us. Dr. Gwynn’s discovery should
stimulate further inquiry into the matter; but at pre-
sent it appears that all the known MSS. contain the
version in its final form after the Harkleian revision.
Fifty-one distinct MSS. are enumerated in Gregory’s
list, twenty-two of them being in England. The

. 1 Evan, 36, Act. 10, Paul. 5, Apoc. 13 ; but nearly allin the second and
third groups also contain the Gospels. In some MSS. one or more groups
of books contain the Peshitto text, while the rest are Harkleian.

ν THE ANCIENT VERSIONS 167

oldest, however, are at Rome, one copy (in the Vatican)
being assigned to the seventh century, and another to
the eighth, while one, which is at Florence, bears a
date in the year 757. Two important copies in the
British Museum belong to the ninth or tenth century.
The Italian copies have not been fully examined, but
at present the best text is believed to be contained in
a MS. at Cambridge, written in 1170, which, in spite
of its comparatively late date, appears to have been
very carefully and correctly written.

The Philoxenian-Harkleian version, though the four
minor Catholic Epistles were published from the
Philoxenian by Pococke in 1630, and the Apocalypse
from the Harkleian revision by De Dieu in 1627, was
practically unknown to scholars until the eighteenth
century. In 1730 two copies of it were sent from
the East to Dr. Ridley, of Oxford, who wrote a dis-
sertation on them; and after his death a text derived
from them was published by Professor J. White
(1778-1803). This remains the only printed edition
of the version to the present day. Preparations for
a new edition were made by the Rev. H. Deane,
of St. John’s College, Oxford, who examined fifteen
of the MSS. preserved in England; but the failure
of his eyesight unfortunately stopped his work. If
some scholar could be found to continue his labours,
he might do much to elucidate the history of what
Dr. Hort described as “one of the most confused texts
preserved.” ἢ |

e. The Palestinian Syriac.—Yet another version of
the New Testament in Syriac remains to be mentioned,
which holds a somewhat peculiar position. In 1789
the German scholar Adler, in a treatise on the Syriac
versions of the New Testament, gave a description

1 Introduction, p. 156.

168 CRITICISM’ OF NEW TESTAMENT “cx

and collation of a Gospel lectionary in the Vatican
(first observed by Assemani in 1758), containing a
text entirely different from any other that was then
known. The difference was not merely in text but in
dialect. Both in vocabulary and in grammatical forms
it resembled the “ Chaldee” found in certain books of
the Old Testament, or the dialect of the Palestinian
Targum, rather than the Syriac employed in the
Peshitto or Harkleian; and from its resemblance to
this Targum it has received the name of the Palestinian,
or Jerusalem, Syriac. The lectionary discovered by
Adler (an imperfect MS., written at Antioch in 1030)
was edited in full by Count Miniscalchi-Erizzo in
1861-64, and re-edited by Lagarde in a posthumous
volume published in 1892. It is not now, however,
the sole authority for this version. Some fragments
of it (also from Gospel lectionaries) came to the British
Museum among the Nitrian MSS.; the Imperial
Library at St. Petersburg possesses some fragments of
the Gospels and Acts;* the Bodleian has four small
fragments from the Pauline Epistles ;* and still more
recently the monastery of St. Catherine on Mount Sinai
has made some important contributions to our know-
ledge of this version. A single leaf from this monastery,
containing portions of the Epistle to the Galatians, was
published by Mr. Rendel Harris in 1890 ;" but this
was completely overshadowed by Mrs. Lewis’ discovery
in 1892 of a complete Gospel lectionary, while another,
equally complete, was identified by Mr. Rendel Harris
in the following year. Both lectionaries (the first of
which is dated 1104, and the second 1118) were
published by Mrs. Lewis and her sister, Mrs. Gibson, in

1 Edited by Dr. Land, of Leyden, in 1875.
2 Published by Mr. Gwilliam in Axecdota Oxontensia (1893).
3 Biblical Fragments from Mount Sinat (1890).

7 THE ANCIENT VERSIONS 169

1899.1 Meanwhile yet another MS., containing lessons
from the Acts and Epistles, besides parts of the Old
Testament, was acquired by Mrs Lewis in Cairo in
1895, and published two years later.” To complete
the list up to the present time, it may be added that a
few leaves from another Palestinian lectionary had
been used in the bindings of the Sinai volumes, and
are included in the above-mentioned edition of them ;
and four fragments of this version, froma MS. recently
acquired by the British Museum, were edited by Mr.
G. Margoliouth in 1897.2. Some palimpsest leaves
from the Cairo Genizah (to which we owe portions of
the Hebrew text of Ecclesiasticus, fragments of the
Old Testament of Aquila, etc.) are now at Cambridge,
containing portions of 2 Cor. and 1 Thess. In 1905
some fragments from Damascus were edited by F.
Schulthess ; and more recently Mrs. Lewis has added
portions of the Gospels, Acts, and Pauline Epistles,
from a manuscript which she assigns to the sixth
century... Thus, though only a few leaves which can
have belonged to copies of the New Testament
Scriptures in their ordinary consecutive form have
come to light, we now have considerable materials on
which to found an opinion as to the character of this
version. The earlier editors referred it to the neigh-
bourhood of Jerusalem, and to a relatively ancient date,
Miniscalchi-Erizzo placing it as high as the second
century ; but Mr. Burkitt® has demonstrated that it is
of Antiochian origin, that it probably originated in the

1 The Palestinian Syriac Lectionary, re-edited by A. S. Lewis and M.
D. Gibson (1899).

2 Studia Sinaitica, vol. vi. (1897).

3 The Palestinian Syriac Version of the Holy Scriptures, four recently
discovered portions, edited with a translation by the Rev. G. Margoliouth
(1897).

4 Horae Semiticae, No, viii., Cambridge, 1909.

5 Journal of Theological Studies, ii. 174-183 (1901).

170 CRITICISM OF NEW TESTAMENT cu.

sixth century as a part of the efforts of Justinian and
Heraclius to abolish Judaism from Judaea and Samaria,
and that it revived with a renaissance of Palestinian
Christianity about the eleventh century. Its text is of
good quality, though somewhat mixed in character,
sometimes associating itself with the 8-type, and some-
times with the 6-type. Hort regarded it as “not
altogether unaffected by the Syrian Vulgate [the
Peshitto], but more closely related to the Old Syriac” :?
while on this Old Syriac (or 6-type) base some §-type
readings have been engrafted in the course of tradition.
The later date now assigned to it, however, lessens the
importance of its evidence.

One other version sometimes figures in the list of
Syriac authorities, under the name of the Karkaphensian
version, to which allusion was supposed to be made by
an ancient writer, Gregory Bar-Hebraeus, and which
Cardinal Wiseman believed himself to have discovered
ina Vatican MS. in 1828. Other MSS. of the same
type have since been brought to light, and it is now
universally recognised that it is not a continuous version
at all, but a collection of texts accompanied by annota-
tions on their spelling or pronunciation. It corresponds,
in fact, to the Massorah of the Hebrew Old Testament,
and only those passages are quoted on which some
annotation is supposed to be required. It may, there-
fore, incidentally furnish us with evidence as to the text
of the Syriac Scriptures used by the commentators, but
it is not a version itself; and as no copy of it appears
to be earlier than the later part of the ninth century, its
value for textual purposes is not great. Seven MSS.
of this Syriac Massorah are known, six of which
emanate from the Jacobite branch of the Church, and
only one from the Nestorian. The name Karkaphenszan

1 [ntroduction, p. 157.

ν THE ANCIENT VERSIONS 171

is believed to be derived from the monastery in which
the Jacobite commentary was compiled, meaning [the
Convent of] the Skull.

2. The Armenian Version.—Adjoining Syria to
the north and north-west lay the territory of Armenia,
the debatable land between the Roman and Parthian
Empires; and as the circle of Christianity widened
outwards from Jerusalem, it was naturally through Syria
that Armenia received the Christian teaching and the
Christian Scriptures. Until recently little was known
of the Armenian version, and little interest taken in
it ; but, thanks especially to the labours of Mr. F. C.
Conybeare, its general character is now known, and
proves to be unexpectedly interesting. With respect
to its origin, the evidence is unusually explicit. Three
Armenian writers of the fifth century (whose statements,
though not identical, are reasonably reconcilable)
record that the Scriptures were translated into Armenian
partly by St. Mesrop, with the assistance of a Greek
scribe named Hrofanos (presumably=Rufinus), and
partly by St. Sahak (Isaac). The date of these first
translations would appear to be about A.D. 395-400.
Subsequently, however, after the Council of Ephesus
(A.D; '431), Sahak ‘and Meésrop received “correct ”
copies of the Greek Bible from Constantinople, where-
upon they revised their previous work in accordance
with this new evidence. With regard to the language
from which the original translations were made, the
evidence is conflicting, both Greek and Syriac being
mentioned; and since both languages were current
in Armenia it is very probable that authorities of both
kinds were employed. That Syriac formed the basis
of it is not only a przorz probable, but is made almost
certain by Dr. Armitage Robinson’s examination of the
Armenian New Testament, in which he establishes a

172 CRITICISM) OF NEW TESTAMENT °° ‘cx.

clear connection between its text and that of the Old
Syriac, not only in the Gospels, but also in the Pauline
Epistles. In the latter the Armenian version appears
to have been made from a Syriac text substantially
identical with that used by St. Ephraem. These
conclusions are confirmed and extended by the infor-
mation derivable from an Armenian writer of the
seventh century, named Theodore, who in the course of
a treatise against heretics refers to an ancient Armenian
version of the New Testament as containing Luke xxii.
43, 44, and also the apocryphal third Epistle to the
Corinthians, which he says was quoted by St. Gregory
the Illuminator, the Apostle of the Armenians at the
beginning of the fourth century. From this it appears
that Mesrop and Sahak were not the first translators
of the Bible into Armenian, but that there was an
Armenian version as early as the beginning of the
fourth century, including the Pauline Epistles as well
as the Gospels (since 3 Corinthians was contained in
it), and taken from the Syriac (since 3 Corinthians
occurred in the Syriac canon and in no other),’

The earliest known MS. of the Armenian Gospels
(now at Moscow) is dated in the year 887. Copies
written in the years 966, 986, and 98g are still in
Armenia (where the most important library is that of
Edschmiadzin) ; one of g60 is at Constantinople, and
two of 902 and 1006 at Venice. Two more of those
described by Mr. Conybeare® probably fall into the
ninth century. The other books of the New Testament
rarely appear in MSS. before the thirteenth century,
and never apart from the Gospels. These late MSS.
are much less trustworthy than the earlier ones, their
texts having been affected by the introduction of the

1 See F. C. Conybeare in the Academy, Feb. 1, 1896 (in a review of

Robinson’s Euthalzana).
2 In his article on the Armenian version in Scrivener, ii. 148-154.

V THE ANCIENT VERSIONS 173

Vulgate into the East during the Crusades. The
fullest collation hitherto of the Armenian version was
published by Tregelles and repeated by Tischendorf ;
but this is based on a few MSS. and those not the
oldest. The text was first printed at Amsterdam in
1666, while the first critical edition appeared in 1805
from the Mechitarist convent in Venice, to which we
are also indebted for the discovery of the commentary
of Ephraem on the Diatessaron, and so indirectly of
the Diatessaron itself.

The Armenian version contains some _ interesting
features, apart from its bearing on the question of the
priority of the Old Syriac over the Peshitto, which has
been mentioned already. The last twelve verses of St.
Mark are omitted from three out of the four very early
MSS. collated by Mr. Conybeare, while the fourth (the
earliest MS. at Edschmiadzin), which has them, adds
a note stating that they are “of the elder Ariston.”
This has naturally been taken to refer to Aristion,’
whom Papias mentions as one of the disciples of the
Lord to whom he had recourse for information with
regard to our Lord’s life and teaching. The statement
of the Armenian codex lacks confirmation, but in itself
it supplies a very plausible explanation of the difficulty
connected with these verses. The original ending of
the Gospel having been lost (or never written), a brief
summary was added, to round off the narrative, by
Aristion, one of the disciples of Jesus, and therefore in
a position to know the facts; but this conclusion, not
being by St. Mark, was sometimes omitted, and conse-
quently does not appear in some of our earliest extant

1 The slight discrepancy in the name is unimportant. By a curious
coincidence, exactly the same discrepancy occurs between Aristotle (’A@.
πολ.) and Plutarch with regard to the name of the adherent of Pisistratus,

who proposed that he should be provided with a body-guard ; the former
calling him Aristion, the latter Ariston,

174 CRITICISM OF NEW TESTAMENT |) jou

authorities. On this hypothesis we can accept the
passage as true and authentic narrative, though not an
original portion of St. Mark’s Gospel.

Two other notable passages may be mentioned.
Luke )xxtil\43,/44 is omitted in’ three out of Mr
Conybeare’s four collated MSS., but appears in the
oldest/;of, them’ (the Venice MS. of. 902); \' and this
agrees with the statement referred to above, that in the
“first translation” or “ancient edition” these verses
were found, but were omitted in the “newly issued
translations.” The same statement is made with refer-
ence to the apocryphal third epistle to the Corinthians.
On the other hand the testimony of the earliest
Armenian codices is entirely against the authenticity
of the episode of the woman taken in adultery. The
first which has it is the Edschmiadzin MS. of 989,
and this has it in a very different form from that in
which it generally appears, adding the curious detail
that what our Lord wrote on the ground was the
sins of the several self-constituted judges, so that
each slunk away as he saw his own fault written
down before him.

Further examination of the Armenian version may
yield additional fruit, especially in respect of its connec-
tion with the Old Syriac; but for this we must await
the pleasure of the few scholars capable of informing
us. Meanwhile we have good reason to be grateful for
the work which has been done and for the results
which have already been obtained.

3. The Georgian Version.—The Iberian Church in
the Caucasus, on account of its geographical position,
would naturally receive Christianity and the Christian
Scriptures by way of Armenia, and the Armenian
tradition is to the effect that St. Mesrop was the author
of the Georgian version as well as of that in his own

V THE ANCIENT VERSIONS 175

language. This, however, is denied by the Georgians,
who looked to Constantinople as the source from which
Christianity had been brought to them, and whose
associations were rather with the Greeks than with the
Armenians. The version itself, according to Mr.
Conybeare,’ shows signs of both influences. On the
one hand it has many parallels with the Old Syriac,
which seems to indicate that this is the text on which
it was originally based; while on the other, it has
evidently been revised from the Greek. This revision,
according to tradition, was made as late as the tenth
century, to which date the earliest MSS. appear to
belong. If, however, a somewhat earlier MS. should
come to light, as may easily happen, it may be found
to contain the original version, based on the Old
Syriac. Inthe Pauline Epistles the Georgian version,
like the Armenian, seems to have been made from a
Syriac text of the same type as that used by Ephraem,
often agreeing with Ephraem against the Peshitto when
the two differ. At present, however, so little is known
of either the history or the text of this version that it
does not need more detailed description here.

4. The Persian Versions.—Another kinsman of the
Syriac New Testament is to be found in the earliest of
the two Persian versions of the Gospels at present
known, which was evidently taken from the Peshitto,
though at what precise date is doubtful. It was printed
in Walton’s Polyglot from a single MS., which appears
fo be dated 1341. A later version from the Greek
was edited by A. Wheelocke, professor of Arabic at
Cambridge, and published posthumously in 1657.
Neither would seem to be of much value for critical

1 Academy, Feb. 1, 18y6. In his article in Scrivener (ii. 156) Mr.
Conybeare speaks of the Georgian version as certainly made from the

Greek ; but the article in the Academy is his latest utterance, and pre-
sumably represents the results of maturer study,

176 CRITICISM OF NEW TESTAMENT cu.

purposes, on account of their late date. Portions of
earlier versions are said to exist in Persia, but they are
still unpublished.

5. The Coptic Versions.—Turning southwards from
Jerusalem we reach a new group of versions, of which
the first and the most important are those which were
made in the adjoining country of Egypt. Here, and
especially at Alexandria, flourishing colonies of Jews
had been established even before the days of the earliest
Ptolemies ; and here the great Greek translation of the
Hebrew Scriptures, known as the Septuagint, had been
prepared. Here, therefore, there was ground on which
Christian preaching might work at once, and there is
evidence that it did so work. The story of Apollos
(Acts xviii. 24-28), the Jew of Alexandria, suggests
that some imperfect exposition of Christianity had
reached Egypt within a few years after our Lord’s
ministry, though it is possible that his knowledge on
the subject had only been acquired since his arrival in
Asia Minor ;* but, whether St. Mark (as tradition tells)
preached there or not, it may be taken as certain that
the generation of the Apostles did not pass away
without the Gospel having been carried into Egypt.
At first, however, this would not imply a translation of
the Scriptures into the Egyptian tongue, since the com-
munity first addressed would be the Greek-speaking
Jews of Alexandria, next to whom would come the
considerable Greek colonies in that town and in Egypt
generally, the existence of which is amply established,
not only by the statements of historians, but by the
Greek papyri of the Ptolemaic and Roman _ periods
which have come to light in such great numbers of late
years. These documents, indeed, show that not only

1 The reading of the 6-text would exclude this possibility, since it has ὃς
ἣν κατηχημένος ἐν τῇ πατρίδι τὸν λόγον τοῦ κυρίου.

¢ THE ANCIENT VERSIONS 177

Greeks and Romans but even native Egyptians not
infrequently used the Greek language for business
purposes, although in many instances the parties were
too illiterate to write with their own hands. For some
time, therefore, the Scriptures in the Greek tongue
would have been sufficient for the purpose of the
missionaries ; the more so as they would hardly have
been able to use any other. The hieroglyphic script
of ancient Egypt was by this time obsolete for practical
purposes, while it is not likely that the Christian
missionaries would have been able to use the demotic
characters, as they are called, which were then the
form of writing in everyday use among the native
Egyptians.

In the course of the second century, however, a new
kind of writing came into existence, which formed a
sort of bridge between Greek and Egyptian. It was,
in fact, the Egyptian language (somewhat modified by
its contact with Greek) written in Greek characters,
with the addition of six other letters borrowed with
modifications from the demotic alphabet for the repre-
sentation of special Egyptian sounds. To this the
name Coptic was subsequently given; and it is in the
Coptic dialects that the native Egyptian versions of
the Scriptures were written. The date of the adoption
of the Coptic script is somewhat uncertain, and it is
a matter of some importance with regard to the
probable date of the original Coptic versions. The
earliest known specimen of it occurs in a horoscope,
written on papyrus and now in the British Museum.
Astronomical calculations show that this horoscope (the
bulk of which is in Greek) was calculated for a nativity
in either 95 or 155.A.D., the former being the more
probable of the two; and the palaeographical indica-
tions also suit the earlier date best. The Coptic in

N

£78 CRITICISMMDF NEW TESTAMENT \ | ‘cu:

this document is of a very primitive type;' but
it is sufficient to make it practically certain that
this form of Egyptian writing established itself in
general use in the second century, and was therefore
available for the translation of the Scriptures before
A.D, ZOO;

Whether advantage was immediately taken of it is
uncertain. It has usually been supposed that it was,
and that the principal Coptic versions date back to the
second century.” In favour of this view a passage is
cited from a Life of St. Anthony, the authorship of
which is attributed to St. Athanasius, and in which St.
Anthony, when about twenty years of age (ze. about
A.D. 270), is said to have been greatly affected by
hearing the Gospel read in church. Since he is
known to have been unacquainted with Greek, this
passage is taken to imply that the Bible which he
heard read was in Coptic, which would prove that a
Coptic version was in existence soon after the middle
of the third century, and would make it very possible
that it was made at the beginning of that century, or
perhaps somewhat earlier. Mr. Forbes Robinson, how-
ever, argues that what St. Anthony heard may have
been only an oral paraphrase from a Greek Bible;
just as, in the early days of Christianity in England,
the Vulgate was paraphrased into English for the
benefit of the uninstructed converts. But even if this
be admitted as a possible (though by no means a
certain) explanation of the passage, clear evidence of
the existence of a Coptic Bible not much later is
provided by the life of St. Pachomius (the great
organiser of Egyptian monachism), whose monks

1 Mr. Ὁ. W. Goodwin calls it ‘‘the first effort of the system from which
Coptic was shortly afterwards developed.”

2 See Lightfoot (af. Scrivener), Headlam (zézd. ed. 4), Hort, Hyvernat,
etc,

a THE ANCIENT VERSIONS 179

(common Egyptians, without knowledge of Greek)
were required to be zealous in the study of the
Scriptures. This was at the beginning of the fourth
century, and affords a fair presumption that the
original Coptic version was made not later than the
middle of the third century, while it is of course
consistent with an even earlier date. There is
therefore not much disagreement between the best
authorities, whose estimates only range between the
latter part of the second century and the middle of the
third. In favour of the earlier view it may be noted
that (as will be shown below) the types of text con-
tained in the Coptic versions are unmistakably early,
the Sahidic New Testament in particular being of a
type which we know to have been prevalent in the
second century, while it can hardly have flourished
much later than the middle of the third. Similarly
the original Sahidic Old Testament was evidently pre-
Origenian in character, not containing those insertions
from the Hebrew which Origen made in his Hexapla,
and which thenceforth appear in all editions of the
Septuagint ; from which it may fairly be inferred that
this version was not made substantially later than the
death of Origen, while it may be decidedly earlier.
If therefore we put the origin of the Coptic versions
about A.D. 200, we shall be consistent with all the
extant evidence, and probably shall not be very far
wrong.

Different dialects of Coptic were spoken in different
parts of the country, but their number and their diver-
gences have only lately begun to be made known to us.
Two of them stand out in importance above the rest,
and until recently were the only two of which scholars
had any knowledge. They belonged to Lower and
Upper Egypt respectively, and the former used to be

180 CRITICISM OF NEW TESTAMENT CH.

entitled the J/emphitic version, and the latter the
Thebatc, from the most important towns in the two
districts. Fuller information, however, has shown that
the district of Memphis had its own dialect, which is
not that of the principal Lower Egyptian version ;
hence this is now generally termed Sohazric, from
Bohairah, the Arabic name of the coast district of
Lower Egypt, while its rival is called Sakzdic, from
Es-sa‘id, the Arabic name of Upper Egypt. These
are the names assigned to them by Athanasius, Bishop
of Kos in the Thebaid in the eleventh century, and
they are now generally adopted by scholars. The
Bohairic dialect was that of the sea-coast, including
Alexandria, the literary capital of Egypt; it was the
most literary of all the dialects, and ultimately it
superseded them all and became the accepted language
of the Coptic Church, as it remains to this day, when
the language is otherwise dead. The Sahidic dialect
had its home in the district about Thebes. Athanasius
of Kos mentions the former existence of a third dialect,
which he calls Bashmuric, from the district of Bashmur,
which appears to have lain in the marshes of the
Delta; but of this no remains now exist. On the
other hand, at least three additional dialects have been
found among the papyri which have come to light of
recent years. The first of these, when only a few frag-
ments of it had been discovered (about a century ago),
was provisionally named Bashmuric, but is now shown
to have belonged to the province of the Fayyum,
which, lying by itself away from the Nile, not un-
naturally had a dialect of its own, and in which most
of the discoveries of papyri in the last thirty years
have been made. This dialect is consequently now
known as fayyumic. Another dialect, found in
documents from the neighbourhood of Memphis, is

ν THE ANCIENT VERSIONS 181

generally termed Memphitic,, or Middle Egyptian;
while in Upper Egypt, apart from the Sahidic, a
distinct dialect has been found in papyri from Akhmim
(Panopolis), which is provisionally entitled Akhmimzc.
This last is marked by the possession of a new letter,
which is not found in the other dialects.

It is not be supposed, however, that these
dialects cover the whole field, or that they all occupy
clearly defined provinces. On the contrary, the more
papyri come to light, the more is it clear that the
greatest amount of intermixture of neighbouring
dialects prevailed, especially with regard to Middle
Egypt. Future discoveries will no doubt enable
Coptic scholars to reduce their subject-matter to
greater order, and especially to show how far the
intermixture of dialects, which is natural in colloquial
correspondence and the business documents of daily
life, extended into works of literature. For the
present it is probably most convenient to group all the
dialects of Central Egypt together as Middle Egyptian,
while provisionally classifying them into sub-species as
Fayyumic, Memphitic, Akhmimic, and other local
names, as may be found necessary.

a. The Bohairie Version.—The Bohairic dialect, as
stated above, ultimately superseded all the others, and
consequently the remains of it now extant are the
most plentiful. In it alone are there many MSS.
containing complete books of the New Testament ;
and although no single MS. contains the New Testa-
ment in its entirety, yet there is fairly plentiful
evidence for each book of it. The first scholar to

1 The only objection to this name is its former use to denote the version
which we now call Bohairic ; but this is an objection which becomes daily
of less importance, as the term Bohairic establishes itself in all text-books.
Middle Egyptian is wanted for a wider use, covering all the (as yet) ill-
defined dialects which range between Bohairic and Sahidic.

182 CRITICISM OF NEW TESTAMENT cu.

make it known was T. Marshall, Rector of Lincoln
College, Oxford, from whose papers many readings
were drawn for the New Testaments of Fell in 1675
and Mill in 1707. In 1716 the text of the whole
version was published at Oxford by the Prussian,
David Wilkins; but the execution of the task left
much to be desired. The Gospels were re-edited by
Schwartze in 1846-47, but with a very inadequate
critical apparatus; and the Acts and Epistles by
Lagarde (then known as Boetticher) in 1852. All
these editions, however, have been superseded by that
issued from the Oxford University Press by the Rev.
G. Horner.’ Mr. Horner prints the text of the Gospels
from the best MS. (Huntington MS, 17 in the Bod-
leian Library), and gives a very full critical apparatus.
Thirty-four MSS. were collated for St. Matthew, and
six more examined ; for the other Gospels it was found
sufficient to collate about twenty MSS. and to examine
about ten. In all, the character of forty-six MSS.
(sought out from all the principal collections in Europe
and Egypt) was ascertained with adequate certainty,
and a thoroughly broad and solid foundation laid for our
knowledge of the Bohairic version. For the remainder
of the New Testament thirty-four MSS. were used, of
which nineteen contained the Pauline Epistles, thirteen
the Catholic Epistles, thirteen the Acts, and eleven
the Apocalypse. The text of the Acts and Epistles is
printed from Brit. Mus. MS. Or. 424 (A.D. 1307), and
the Apocalypse from Curzon MS. 128 (A.D. 1320).
The MSS. of the Bohairic version, though fairly
numerous (Mr. Horner’s list,” which does not claim to

1 The Coptic Version of the New Testament in the northern dialect,
otherwise called Memphitic and Bohairic, vols. i. and ii. (1898), vols. iii.
and iv. (1905).

2 In Scrivener, ii. 110-123; Gregory gives a somewhat longer list, but
his additional MSS. are almost all of the eighteenth and nineteenth
centuries.

PLALE Xi

ASR PRIMI
PRTERLEE REGS OA PRTG Rau Oo ANRC G
Pics κα εγενενείτειν τιν}. map et Bas Seer
"GC s aera ae ater i ea Me ψεγεες,
xXee nancy = ACI ΚΝ ἐν IATA τ ΧΟ Σ

By aren εις τ. Gon AGG Pane 2A.
race baxcant oe
TPLAMNIG— « TATE να LANG EF γὰρ. . CETITOC
~2. ORG ξεν REPL κα ἀτεὐετεσλν thar terApx ts
MIME TALIS Atatis eonGe:} amano
LAN“ TGmAGraoG ©) GONGNGAGIGTO
RM NGAI A CCN NEL us ΞΕ tage ee
ΠΑΡΌΝ LEGGY TAU GAO | —
ae Japlocac AYEP UT GGA OEY ‘ednene|
“ἢ CURE εἰετεεξενενλεῖῖνγεινε udpel pcan
OYGTHAGLIIA Gay GCLELS Fiat pate
“rapxrm MME VAG ASEM
_ormyupatcp τ :
OG WAGLIGC HrAp> meuLle CGOE GAvep »
= A VST GIG NGN NG G_SINLIGIGOO TAG ray
berm} napecmas © SU AUDCEDIS EPEC |
τς δοῦλοι 3GOWAL = δον gen ea
aa APNG: 5 ee
Ἰ ΕΤ pros AGsaiGat Τ᾿ » AVGP Dues
2 ἐπα εατολτε une gy HWMLNGIGHONLE recut
ie υ Saban ena owe oo NGgus
2 “Truqanaweirlac oe ee en

τον

G)zor AC YAP MO HG mat Gue pemey
HOW NGLIMEG Cn NGIGKOAGeraf-
Spee UGGousAD wap. kh

a

DAG

THE BOHAIRIC VERSION (Curzon Catena.) A.D. 889.
(Scale 1: 3. The page shown is the first of St. Mark’s Gospel. )

To face p. 183.

ν THE ANCIENT VERSIONS 183

be complete, gives thirty-six copies of the Gospels,
eighteen of the Acts and Epistles, and ten of the
Apocalypse, besides Lectionaries), are all late in date.
The earliest is that known as the Curzon Catena, in
the Parham Library (at present located in the British
Museum), which is dated in the year 889; but in
this MS. text and commentary (derived from various
sources) are intermixed, often indistinguishably, so that
its value as an authority is impaired. Of the MSS. of
the Gospels, the oldest and best is the Huntington MS.
17, already mentioned, of the year 1174 (Horner’s A) ;
and closely akin to this are a MS. in the Bibliotheque
Nationale at Paris of 1196 (Horner’s C) and one in
the Institut Catholique at Paris of 1250 (Horner’s ΗΔ
These three embody the purest text of the Bohairic
version; but there are also MSS. of 1179 (Paris),
1184 (Cairo), 1205 (Rome), 1208 (British Museum),
PSL 950. ana? 1250 (Paris), 1257, 1272, and ‘T2091
(Cairo), besides others which, though not precisely dated,
may be as early as these. Late though all these MSS.
are, there is good reason to suppose that they contain
a substantially pure text. There are no such wide
variations as we find among Greek MSS.; rather the
Copts seem to have resembled the Jews, who have not
preserved the early copies of their Scriptures, but have
copied them with the greatest fidelity, so that their
MSS. of the tenth century and later contain a text
which has come down substantially unaltered from at
least the second century. Indeed many of the Bohairic
MSS. which contain corrections have notes affirming
that the variants are Greek, not Coptic, thus implying,
as Mr. Horner points out, that the Copts jealously
preserved their own textual tradition. The later MSS.

1 A few scattered leaves of earlier date have recently been found, but
not many.

184 CRITICISM OF NEW TESTAMENT ocx.

have, it is true, suffered some corruption by the inser-
tion of words and passages which bring them into
closer conformity with the Textus Receptus; but
these insertions can for the most part be recognised
by reference to the earlier copies. The Huntington
MS. 17, indeed, supplies an instructive example of this
process of corruption ; for while the main text is pure,
most of the passages which we find inserted in the later
copies are here written in the margin. It is easy to
understand that these would, before long, find their way
into the body of the text.

Taking, then, the earlier MSS. as substantially
representing the Bohairic version in its integrity, it
will be found that no such difficulties present them-
selves as in the case of the Old Syriac or (as will
be shown below) the Old Latin. It is an example
of the §-text of the New Testament in its purer form,
of which the leading representative is the Codex Vati-
canus, It is not marked by the numerous erratic
variants which are characteristic of the Codex Bezae
and its friends. Further, the translation is careful and
obviously adheres closely to the Greek, so that it can
be used with confidence as evidence for the Greek text.
Its general agreement with the text of NB is shown by
the fact that out of twenty-one passages enumerated
on pp. 57, 58, where those two MSS. agree as against
the Textus Receptus, the earlier MSS. of the Bohairic
support them in fifteen and oppose them in six; while
in two more cases where δὶ joins the Textus Receptus,
the Bohairic adheres to B. The last twelve verses of
St. Mark are contained in all Bohairic MSS. ; but two
copies (Hunt. 17 and Brit. Mus. Or. 1315) give in
their margins a short alternative ending which is
practically identical with that found in L. The
passage Luke xxii. 43, 44 is omitted in nearly all

ὰ THE ANCIENT VERSIONS 185

the better Bohairic MSS.; John v. 3, 4 (the angel
at the pool of Bethesda) is omitted by most of the
better MSS., but appears in the best of all (Hunt. 17);
John vii. 5 3—viii. 11 is omitted by all the better MSS.
The Apocalypse apparently did not form part of the
original version. In nearly all cases it is found in a
separate MS.; when otherwise, it is markedly dis-
tinguished from the other contents of the MS.; and
it is not noticed in the Copto-Arabic Bible vocabularies.
It is therefore probable that the version was made at a
time when the Apocalypse was not universally recog-
nised as a canonical book. From the end of the third
century it seems to have been accepted; but in the
middle of that century doubts were expressed about it.
To this period, therefore, the origin of the version is
ascribed by Lightfoot ;* though the possibility remains
that the version may have come into existence earlier,
and have dropped the Apocalypse at that date. On
the other hand, the fact that the Bohairic version
(unlike the Sahidic) contains Origen’s insertions in
the text of the:Old Testament; (see above; pi 1.79)
points rather ‘to the end of the third ,century. or, te
some part of the fourth. The point must remain
at present somewhat uncertain; but the type of
text contained in this version is distinctly in favour
of as early a date as is compatible with the other
evidence.

ὦ. The Sahidic Version.—Very little notice was
taken of this version until about a century and a
quarter ago, when (in 1778) Tuki published a grammar
of the dialect, with quotations from both Testaments,
and Woide simultaneously announced a forthcoming
edition of the fragments of the New Testament, which
did not appear until 1799, nine years after his death.

1 Scrivener, ii. 123,

186 CRITICISM OF NEW TESPAMERT ‘cx

Since that date the number of extant fragments has
increased enormously, and it is now possible (as Mr.
Horner is in process of showing) to piece together an
almost entire New Testament; though it would be a
very patchwork production, being compiled from in-
numerable scraps of different MSS., the relative worth
of which we are hardly in a condition to test. Until
recently, no complete copy of any book of the New
Testament was known to exist in the Sahidic dialect,}
though seven MSS. contained books which approached
completeness. Within the last few years, however, the
British Museum has acquired, first, a vellum MS. of the
Apocalypse, of about the eleventh century, complete
except for a few verses, and secondly a papyrus codex
of the first half of the fourth century which contains
the books of Deuteronomy, Jonah, and the Acts,
complete except for local mutilations.? Fragments
are very numerous, and increase almost yearly. In
1810 Zoega published a large number from the
Borgian Library, which have since found their way
into the library of the Propaganda at Rome. ‘These
have now been re-edited, the Old Testament by Ciasca
in 1885-89, the New Testament by Balestri in 1904.
A still larger collection was acquired by the Biblio-
{πόσα Nationale at Paris in 1883 from the great
White Monastery near Sohag in Upper Egypt; and
these have been catalogued by Amélineau.? The

1 The same might have been said of the Old Testament, with the
exception of the book of Job; but in 1898 Dr. E. A. W. Budge edited a
complete Sahidic Psalter from a papyrus codex (an almost unique example
of a complete papyrus MS. in book form, and of very large size) acquired
by him in Egypt for the British Museum ; and still more recently another
codex has been obtained which, in addition to the Acts (mentioned below)
contains the books of Deuteronomy and Jonah.

2 Both these MSS. have been published by Dr. Budge in the present
year (1912) in Coptic Biblical Texts in the British Museum.

3 For a list of these fragments see Scrivener, ii. 134-136. Some
other fragments from the White Monastery have been edited by Maspero
(Mémoires de la Mission francaise au Catre, vol. vi. 1892).

"Let -¢ anf of
(‘6-9 ‘1x ‘oody urejuoo umoys sased ay], ᾿59ΛΈΘ] OT ynoqe jo paysisuoo
aAvy plNoM 11 ‘ajafdwoo ueyM | auoye asdAyTeoody ay} Sululejuos ‘aumyjoa Au τ ‘AZIS [VUISIIQ)

(4) AmnyuaDy yyIT “(gIS€ 3O “SIN ‘sNIW UG) NOISNYAA DIGIHVWS AHL

ON PWELOINH™M
WIM OVODON . ἢ
Aow A019 τ 1.

τον SON

᾽
ᾷ

ΠΧ ALY Wd

v THE ANCIENT VERSIONS 187

Apocalypse was edited in 1895 by H. Goussen from
a very early but imperfect MS., partly in London
and partly in Berlin (see Plate XIV.), and again from
the same MS. by Delaporte in 1906. The British
Museum, besides the MSS. already mentioned, also
possesses many fragments, which have been catalogued
by Mr. Crum.’ Over sixty fragments (including six
bilingual leaves) in the Rainer collection at Vienna have
been edited by Wessely.” From all these materials
Mr. Horner has long been engaged in preparing an
edition of the Sahidic New Testament, similar to that
which he has completed for the Bohairic version. The
Gospels have already appeared in three volumes, and
the rest is in a forward state of preparation.

The manuscripts of the Sahidic version are much
earlier than those of the Bohairic; and the reason
of this is simple. In the earlier centuries of Coptic
Christianity, manuscripts of both versions no doubt
existed in considerable numbers; but since no special
effort was made to preserve them, they went the way
of all books, to the rubbish heap, and their places were
taken by later copies. But as the Bohairic dialect
ultimately became the literary and religious language
of the country, Sahidic manuscripts ceased to be
written, and the volumes which we possess, nearly all
of which (as we have seen) are later than the eleventh
century, are all Bohairic. On the other hand, the
fragments of earlier date which have been unearthed
by explorers and excavators of modern times are
almost all Sahidic, because the climate of Upper
Egypt is far more favourable for their preservation
than the moister air and soil of Lower Egypt. To
what precise age they should be referred is a matter of

1 Catalogue of the Coptic Papyrt in the British Museum (1905).
2 Studien zur Paliographie und Papyruskunde, xi. (Leipzig, 1911).

188 CRITICISM OF NEW TESTAMENT cu.

considerable doubt. Very few indeed are dated, or
even approximately datable, and consequently there is no
firm basis for early Coptic palaeography. In the case
of the papyrus MS., referred to above, containing the
books of Deuteronomy, Jonah, and the Acts, we are in an
exceptionally favourable position ; for some additional
matter has been appended at the end in a cursive hand
which can be assigned with certainty (from its resem-
blance to several dated examples among Greek papyri)
to about the middle of the fourth century. The Bible
text consequently cannot be later than this period, and
is therefore one of the earliest Biblical MSS. of any
considerable size in existence. Usually, however, trust-
worthy evidence of the date of Coptic MSS. is wanting.
The only clue is to be found in its resemblance to
Greek writing ; and this would lead us to suppose that
most of the oldest Sahidic fragments may be referred
to the fifth century, or perhaps slightly earlier. Since,
however, Coptic never attained the position of a literary
and fashionable language, it was not usually written by
scribes of the first class. The Coptic MSS. which
remain will seldom bear comparison with the Greek
uncial MSS., and the roughness of style which nearly
always characterises them makes any attempt to fix
their date extremely precarious.

It might be thought that light would be thrown on
this problem from two sources, namely, the bilingual
Greek and Coptic MSS., of which there are several
extant,” and the Coptic papyri which have been dis-
covered in large numbers of late years, and which

1 Mr. Horner gives facsimiles from ten MSS., of which he assigns one
to the fourth century, two to the fifth, three to the sixth, one to the seventh,
and three to the eighth.

2 See above, pp. 114, 187. A description of the Graeco-Sahidic frag-
ments in the Bibliotheque Nationale has been given by Amélineau, in
Notices et extraits des manuscrits de la Bibl. Nat., vol. xxxiv. pp. 363 ff.
(1895), with some facsimiles.

v THE ANCIENT VERSIONS 189

might be expected, like the similarly discovered Greek
papyri, to contain a considerable amount of dated
material. But in the case of the bilingual MSS., it is
generally the Coptic which regulates the style rather
than the Greek. The Greek is written with all the
peculiarities and roughnesses of the Coptic, and there-
fore takes no assured place in the development of
Greek palaeography. Some approximation to ordinary
Greek writing there is, and the clue is a valuable one ;
but it is only imperfectly and uncertainly applicable.
With regard to the papyri, it so happens that nearly
all the extant business documents (in which dates may
naturally be looked for) belong to that late Byzantine
period, from the seventh or eighth century onwards,
when dating was done by the fifteen-year indiction-
cycle, which is wholly useless after the lapse of a few
years, since there are no means of knowing which is
the indiction spoken of. One large group of Sahidic
papyri, from the neighbourhood of Thebes, can be
shown to belong to the eighth century ; others, from
the Fayyum, are said to belong to the sixth and
seventh centuries, but since these are unpublished it is
impossible to say what light they may throw on the
palaeographical question. At present the evidence
derivable from papyri is very slight.}

Taking, then, the fourth century as the starting
point furnished by the British Museum MS. above
mentioned, a considerable number of the extant Sahidic
Biblical fragments may be assigned to the period
between A.D. 300 and 800; but they are generally so
small in extent that no detailed description of them
can be given, and few can be singled out as pre-
eminently valuable above the rest. Among the largest

1 Professor Hyvernat has attempted to form a palaeographical classifi-
cation of Coptic hands ; but it is impossible not to feel that it rests on a very
precarious basis,

190’ ‘CRITICISM OF NEW ‘TESTAMENT οἱ

and most noticeable is the bilingual fragment at Rome,
which has already been described under the letter T
(p. 114), and which contains an almost unadulterated
sample of the 8-type of text, such as we find in the
best Bohairic MSS., though of considerably earlier
date than those. The British Museum MS. of Acts
confirms this conclusion, belonging generally to the
family of Bx. The Sahidic version, however, shows
a considerable admixture of those aberrant readings
which we associate with D and its allies, the Old
Syriac and Old Latin.’ It must have been made
independently of the Bohairic version, and from Greek
MSS. of a different type. Whether it was made
earlier or later is a point upon which experts formerly
differed. On the one hand there is the probability
that Lower Egypt, from its greater proximity to
Palestine and to the Jewish-Christian colony of Alex-
andria, would be first evangelised and would first
require a Bible in the vernacular ; to which it is added
(1) that some of the commonest abbreviations in
Coptic MSS. could only have been derived from the
Bohairic, which suggests that it was in this dialect that
the Coptic writing was first used, and (2) that the
greater purity of the Bohairic text is a sign of its
greater antiquity.” On the other hand it may be
argued that Greek would probably suffice for the
purposes of Christianity for a longer period in the
neighbourhood of Alexandria than in Upper Egypt ;
that the abbreviations of which mention has been
made cannot be shown to go back to a pre-Sahidic
period, since there are no Bohairic MSS. of earlier
date than the Sahidic, and at most they only prove
that these abbreviations were first adopted by Bohairic

1 Mr. Horner finds that this element is not so prominent as was

formerly supposed.
2 So Headlam in Scrivener, 126, 127, following Krall.

v THE ANCIENT VERSIONS IQ!

scribes; ' and that the type of text found in the
Sahidic version is shown by the evidence of the Old
Syriac and Old Latin versions and all the early
Fathers (including Clement of Alexandria) to go back
to quite primitive times, certainly to a period earlier
than the production of any Coptic version, so that its
non-appearance in the Bohairic MSS. (even admitting
their text to be intrinsically purer) is not due to its
non-existence when that translation was first made.
The Sahidic text: .may be: less. pure,. and , yet; more
ancient ; indeed, since that type of text is only found
in very early authorities, and evidently perished when
the growth of Christianity brought with it greater care
for accurate copies of the Scriptures, its occurrence in
the Sahidic version is rather to be taken as an
indication of a very early date for the origin of that
version.. To this it, may be added that: in the, Old
Testament the Sahidic version shows strong signs of
being older than the Bohairic. The original Sahidic
text was certainly pre-Origenian in some books (notably
in Job), and probably in all (though modified later) ;
while the Bohairic version appears to represent . the
Hesychian edition of the Septuagint, which was pro-
duced about the beginning of the fourth century, at
least half a century after the epoch-making labours of
Origen in his Hexapla. On the whole, therefore, the
balance of evidence seems certainly to be in favour of
the priority of the Sahidic version, and of its assignment
to a date not later than the middle of the third century,”

1 Mr. Forbes Robinson also denies the applicability of this argument of
Krall’s, affirming that the abbreviations are equally derivable from Middle
Egyptian.

2 If the original absence of the Apocalypse from the Bohairic version
may be taken as a fairly close indication of date (see above, p. 185), the
two versions must be nearly contemporary, since it seems equally to have
been absent from the Sahidic. But the omission cannot really be dated
with any certainty.

192 CRITICISM OF NEW TESTAMENT - ‘cx:

and probably appreciably earlier. In any case it
may be said with confidence that both versions are
ancient, in origin if not in extant representatives ;
both are valuable, and the more so because of their
independence ; and if the greater roughness and even
the more erratic readings of the Sahidic are indications
of an age when such licenses had not been pruned
away, the Bohairic probably comes nearer to the
original purity of the sacred text, as preserved in the
more literary atmosphere of Alexandria.

c. The Middle Egyptian Versions.—Doubtful as
many points are with regard to the Bohairic and
Sahidic versions, they are as clear as day compared
with what we know of the remaining dialects of Egypt.
Though the existence of a third dialect was known so
long ago as 1789, when Giorgi published a small
fragment from the Borgian Library, it is only of late
years that any considerable body of materials relating
to it has been brought to light. The excavations in
search of manuscripts, which have been made at
various points in Egypt within the present generation,
have led to the discovery of many fragments which
cannot be ranged as either Bohairic or Sahidic. As
has been indicated above, manuscripts have been found
in the neighbourhood of Memphis, the Fayyum, and
Akhmim, presenting dialects which, though akin to
one another, have distinctive differences; and these
have been tentatively classified as separate species.
The greatest obscurity, however, still rests over their
relationships. In the first: place, the) evidence’ of
locality is not always satisfactory. Many of the
fragments have been acquired from dealers’ shops ;
and the statements of native dealers as to the provenance
of their wares are ordinarily valueless. Nor is it
always safe to argue from the MSS. in company of

v THE ANCIENT VERSIONS 193

which they are found ; for the collocation may be only
that of the dealers’ box, not of actual discovery.
Further, the intermixture of dialects is great. All are
more associated with Sahidic than with Bohairic, but
the degree of affinity with Sahidic varies indefinitely.

Some points, however, seem fairly established. In
the first place, the Middle Egyptian versions are in-
dependent of both Bohairic and Sahidic in substance,
being derived from a different Greek text. Next,
traces have been found of more than one Middle
Egyptian text,’ so that more than one translation must
have been made. Thirdly, though the extant frag-
ments are not precisely datable, they plainly go back
to a very early date, to the sixth and fifth centuries at
least, if not the fourth. More than this it is impossible
to) Say vat. present; for want. of evidence. « ΟΥ̓ the
Memphitic version only a very few fragments are ex-
tant, published by Bouriant, Krall, Crum, and Chassinat.
The Fayyumic, which was the first to be published, the
fragment found by Giorgi being of this type, is more
fully represented, the Fayyum having hitherto been the
most prolific field for papyri of all kinds. Some were
published by Zoega in 1810, others more recently by
Maspero, Krall, Crum, Wessely, and David, while many
still remain unpublished. All, however, are small,
rarely reaching the extent of a chapter; the largest
being a bilingual palimpsest of the sixth century in the
British Museum, containing John iii. 5—iv. 18, iv. 23-35,
45-49, in a dialect which Mr. Crum hesitates to char-
acterise definitely as Fayyumic or Memphitic, and
with a very pure text.”

1 Mr. Crum has found two fragments of Romans (one in the British
Museum and one at Vienna), both in the Middle-Egyptian dialect, but
markedly different in text.

2 Published by Mr. Crum and myself in Journal of Theological Studies,
i, 415 ff.

O

194 CRITICISM OF NEW TESTAMENT ocx.

Finally, of the Akhmimic version the only extant
fragments were published by Mr. Crum (1893) and
Roesch (1910), but at present the remains are extremely
scanty, and the character of the text quite obscure.
Time will, no doubt, add much to our acquaintance
with all the Middle Egyptian dialects, and with the
versions of the Bible current in them; but meanwhile
it cannot be said that they yet contribute much of
value to the criticism of the sacred texts.

6. The Ethiopic Version.—Abyssinia might have
received Christianity either through its immediate neigh-
bour, Egypt, or more directly from Palestine, by
travellers voyaging down the Red Sea; but records
on the point are wanting.- About the end of the fifth
century Christianity became the national religion, and
to this date, or somewhat later, the Ethiopic version of
the Scriptures is assigned.’ Little, however, is known
of it; for, in the first place, the manuscripts, though
fairly plentiful, are very late in date, and secondly, the
text has never been critically edited. The New Testa-
ment was first printed at Rome in 1548—49, whence it
was included (with a Latin translation) in Walton’s
Polyglott ; and another text was issued by the Bible
Society in 1830; but neither of these editions was
based on a critical study of manuscripts, comparatively
few of which were then known. The Abyssinian war
led to the discovery of many more, and over a hundred
copies are now extant in the libraries of Europe. The
oldest is believed to be a manuscript in the Bibliotheque
Nationale (MS. aeth. 32) written in the thirteenth
century, while another is dated in the year 1378; but

1 So Guidi (quoted by Margoliouth in Scrivener, ii. 154), Gildemeister
(see Gregory, Zextkritik, p. 554), and Hackspill (Zedtschrift fiir Assyrt-
ologie, xi. 117, 1897). Dillmann, however, assigns the Ethiopic Old
Testament to the fourth century, and the New Testament would certainly
be translated as soon as, or sooner than, the Old Testament,

V THE ANCIENT VERSIONS 195

most are of the seventeenth century or later. Accord-
ing to the latest statement on the subject (by L. Hack-
spill, see note on previous page), the oldest MS. stands
alone for comparative purity of text, all the others repre-
senting a text revised in the fourteenth century from the
Arabic version then current in Egypt. It is not at all
impossible that this and other points will be cleared up
in the future by the discovery of earlier MSS.; but at
present our knowledge of the Ethiopic version is too
slight for much use to be made of it. Fortunately, an
edition of it is now in course of preparation by Mr. R.
H. Charles, the most competent scholar who could be
desired for the purpose.

7. The Arabic Versions.—Translations of the Bible
into Arabic are plentiful, but are useless for critical
purposes, on account of their late date and of the
extremely mixed character of their texts." Some are
translated from the Greek, others from Syriac, others
from Coptic; others have been translated from one
language and corrected from another. There is a tradi-
tion that a version was made from the Syriac in the
seventh century, and the earliest extant MS. is one in
the Vatican, which is assigned to the eighth century,
and appears to contain a translation from the Peshitto.
This comes from the monastery of Mar Saba, where
there are also MSS. of the ninth century containing
a version from the Greek, probably ultimately from the
bilingual MS. formerly known as ©* and now as 0136.
Other MSS. belong to the tenth and subsequent
centuries ; the predominant version being perhaps one
due to a revision undertaken in the Patriarchate of
Alexandria in the thirteenth century. This is the only
version which has found its way into print, several

1 See Burkitt’s article on the Arabic versions in Hastings’ Dictionary of
the Bible.

196 CRITICISM’ OF NEW TESTAMENT. cu,

editions of various parts of the Bible having appeared
since I591I. Recently some early Arabic MSS. have
been found at Mount Sinai, which may be derived from
an older translation than those hitherto known. Mr.
Rendel Harris first published a fragment of a Greek
and Arabic MS. of the ninth century from this source,
containing a few verses of St. Matthew.’ Since then
Mrs. Gibson has published a large part of the Pauline
Epistles (Romans, I and 2 Corinthians, Galatians, and
part of Ephesians) from a MS. of equal date,” and
more recently the Acts and Catholic Epistles from
a MS. of the eighth or ninth century.’ Both of these
MSS. contain texts different from any hitherto known,
and no doubt earlier. The former is said to have been
made from the Greek, the latter from the Syriac,—the
Acts and the three longer Epistles from the Peshitto,
the four shorter and more disputed Epistles (2 and 3
John, 2 Peter, and Jude) from the Philoxenian, In
connexion with the text of the Philoxenian version, the
Arabic has some value; but with this exception the
textual importance of all the Arabic versions hitherto
known is infinitesimal.

We pass now from the East to the West, where
the ground is almost wholly occupied by the Latin
versions.

8. The Latin Versions.—The history of the Latin
Bible divides itself into two well-marked portions,
namely, the history of the Vulgate, the great translation
which has been the Bible of Roman Christendom for
fifteen hundred years, and the history of the texts
which preceded it. Historically the former is infinitely
the more important, by reason of its commanding

1 Biblical Fragments from Mount Sinaz (1896), No. 9; reprinted with
additions in Studia Sinaztica, i. App. 1,

2 Studia Sinattica, ii. (1894),
8 Jbid. vii. (1899).

v THE ANCIENT VERSIONS 197

position in the Christianity of the West; but textually
its predecessor is the more interesting, by reason of its
early date and its remarkable characteristics. Of this
version it is now our duty to speak.

a. The Old Latin Version (or Versions).—Diff-
culties confront us at the outset. It is doubtful whether
one should speak of it in the singular or in the plural ;
for the extant MSS. differ so greatly from one another
that it is questionable whether they could have origin-
ated from a single archetype. Nor is it easy to
determine the locality in which it (or the earliest form
of it) had its birth.

A priori it would seem natural to look to Rome for
the origin of the Latin Bible. Christianity, as we know
from the Acts of the Apostles, was carried thither by
the middle of the first century, and it was a leading
Church by the end of that century. It is clear, how-
ever, that the early Roman Church was more Greek
than Latin. The earlier bishops of Rome, with few
exceptions, have Greek names ; and Clement, the most
notable exception, wrote his epistle in Greek. The
first clearly Latin bishop of Rome is Victor (189-199).
The early literature of the Roman Church, the Epistle
of Clement, the “Shepherd” of Hermas, the Apologies
of Justin Martyr, and the works of Hippolytus, are
in Greek. The first Roman liturgy was in Greek, as is
still indicated by the “ Kyrie” which survives from it.
Nor is this apparent anomaly difficult to explain. The
educated inhabitants of Italy used Greek as freely as
their own tongue; the private meditations of a Roman
emperor (Marcus Aurelius) were written in Greek. On
the other hand, the trading and slave population of the
towns, in which Christianity was probably preached
most in the early days, was largely recruited from the
Greek-speaking nations. Certain it is, in any case,

198 CRITICISM (Or NEW TESTAMENT | cz,

that the earliest Latin Bible shows every sign of having
been produced far from the literary influences of the
capital.

The roughness of the more primitive forms of the
Old Latin text, and the characteristic peculiarities of
its dialect and vocabulary, have commonly been held?
to point to Africa as its home, since they are also
found in the African Latin authors, such as Tertullian,
Apuleius, Cyprian, Arnobius, etc. Nor is this hypo-
thesis improbable on general grounds. Greek would
be little known in Northern Africa, which has always
been parted by a great gulf from Egypt, and easily
remains unaffected by its influence. On the other
hand, it was conspicuously flourishing in intellectual
activity during the second century, and the writings
of Tertullian (flor. 195-225) show that Christianity
must have been preached there at a very early date.
Here, then, a Latin version of the Scriptures would be
required earlier than at Rome, and as the linguistic
peculiarities of what appears to be the earliest Latin
Bible suit this hypothesis, it may be admitted that a
strong case has been made out, while contradictory
arguments are few. It must be noticed, however, that
the linguistic argument is not so strong as it looks;
for it so happens that there are hardly any extant
Latin authors of this period who were not Africans, so
that, while we can say that the characteristics of the
Old Latin vocabulary existed in Africa, we cannot
certainly say that they existed nowhere else in the
Roman Empire. Still, the roughness and vigour of
the language suit some such energetic province as this,
and the agreement in text between the quotations in
the African Fathers and what appears to be the earliest

1 Since Wiseman’s letters on the controversy concerning I John v. 7

(1832-33).

ν THE ANCIENT VERSIONS 199

form of the Latin Bible is a strong confirmation of
this hypothesis. Hence at present, at any rate, the
theory of an African origin of the Latin Bible must
be said to hold the field, and to hold it with increasing
strength.’

The date of it cannot be fixed with precision, nor
is it likely that all the books of the Bible were trans-
lated at one time or by one person. Later than the
end of the second century it cannot be, since Tertullian
clearly knew of a complete Latin Bible; and cor-
roborative evidence may be found in the fact that in
the Old Testament the primitive Latin translation
(which was made from the Septuagint) has none of
those insertions from the Hebrew which were made by
Origen in his Hexapla. On the other hand, we have
no evidence which will carry us back to the beginning
of that century. Early though the type of text found
in the Old Latin MSS. unquestionably is, some interval
of time must be allowed for the dissemination of its
peculiar vagaries. Probably if we assign it to the
middle of the second century we shall not be very far
wrong.

The Old Latin version having eventually been
superseded by the Vulgate, it is not surprising that
the extant manuscript authorities for it are neither
numerous nor perfect. On the other hand, they are
almost all of very early date. They differ so greatly
from one another that it is necessary to describe them
separately. The several MSS. (or fragments) are
indicated by the small letters of the Latin alphabet.
The authorities for the Gospels (which are almost

1 Sanday (Guardian, May 25, 1892; cf. Kennedy, in Hastings’ 2 Ζεί.
of the bzble) suggests Antioch as the original centre whence the Latin and
Syriac versions alike took their origin. In any case, as will be shown
below, the earliest form in which we now have the version appears to be
closely connected with Africa.

200 CRITICISM OF NEW TESTAMENT cu.

wholly distinct from those for the other books) are
given first.

a. Codex Vercellensis, at Vercelli: probably of
the fourth century. Traditionally written by Eusebius,
Bishop of Vercelli (ob. 371). Contains the four
Gospels (in the usual Latin order, Matthew, John,
Luke, Mark), somewhat mutilated, and in a much-
damaged condition.1 Collated by Bianchini in 1727,
published by Irico in 1748 and Bianchini in 1749
(with ὁ and f in his Evangelium Quadruplex, reprinted
in Migne, Patrologia, vol. xii.); finally by Belsheim (a
prolific editor of Old Latin texts) in 1894.

a”: see 2.

ὦ. Codex Veronensis, at Verona: fifth century.
Written in silver letters on purple vellum. Contains
the Gospels with mutilations. Edited by Bianchini
(and Migne) with @; some corrections and additions
by E. S. Buchanan in Journ. Theol. Studies, x. 120
(1908). Re-edited, with considerable additions and
improvements, by Buchanan in 1911 (Old Latin Biblical
Texts, no. vi.). Facsimile in the Turin M/onumenta
palaeographica sacra, Pl. 11.

c. Codex Colbertinus, at Paris: twelfth century.
An extraordinarily late copy of the Old Latin, due to
its having been written in Languedoc, where the use of
this version lingered late. Contains the four Gospels
complete, with the rest of the New Testament added
later from the Vulgate. Edited by Sabatier in his
great edition of the Old Latin Bible (1751), and
Belsheim (1888).

ad, Codex Bezae, the Latin text: see p. 88 ff. The
four Gospels and Acts, with 3 John 11-15.

e. Codex Palatinus, formerly at Trent, now at

1 White’s statement (in Scrivener) that this MS. is written in silver
letters on purple vellum appears to be erroneous.

v THE ANCIENT VERSIONS 201

Vienna, with one leaf at Dublin: fifth century.
Written in silver letters on purple vellum. Contains
the Gospels (Matthew, John, Luke, Mark), considerably
mutilated. Borrowed from Trent and copied by
Bianchini in 1762, but not published until 1847, when
Tischendorf edited it; and almost simultaneously the
Dublin leaf was acquired and edited by J. H. Todd.
In 1879 this was identified by T. G. Law as belonging
to e (Academy, 1879, March 1), subsequently being re-
edited by T. K. Abbott (1880). The text of ὁ is akin
to that of &, but represents a later stage of it.

yf. Codex Brixianus, at Brescia: sixth century.
Written in silver letters on purple vellum. Contains
the Gospels, nearly complete. Edited with a and ὁ by
Bianchini (and Migne); and thence printed by Words-
worth and White in their great edition of the Vulgate,
as representing, in their opinion, the type of Old Latin
text on which Jerome’s revision was based.

ff. Codex Corbeiensis I., originally at Corbey in
Picardy, then (circ. 1638) at Sti Germain des Prés in
Paris, which was plundered during the Revolution, now
in the Imperial Library at St. Petersburg. Variously
assigned to the eighth, ninth, or tenth century, the
latter being probably the true date. Edited by
Martianay in 1695, and (being the first Old Latin MS.
published) used by subsequent editors of the version.
Republished by Belsheim in 1881. Contains only St.
Matthew. Its text is a mixture of Old Latin and
Vulgate readings, and it is quoted by Hort (/utro-
duction, p. 82), among examples of the process by
which Old Latin readings found their way into texts
fundamentally Hieronymic.

ff. Codex Corbeiensis II., formerly at Corbey, now
at Paris: formerly assigned to the sixth or seventh
century, but Mr. E. S. Buchanan, who has minutely

202 CRITICISM OF NEW TESTAMENT occu.

examined and edited it (Old Latin Biblical Texts, v.
1907), regards it as almost as old as a, ze. of the
fourth century. This is perhaps over-sanguine, but it
may well be of the fifth century. It.contains the
Gospels, with considerable mutilations, especially in
St. Matthew, of which the first eleven and a half
chapters are lost. The text is a good and early
example of the European family, akin to a and ὁ. It
had been previously edited by Belsheim in 1887, and
was quoted by Calmet a century earlier.

g'. Codex Sangermanensis I., formerly at St.
Germain des Prés, now in the Bibliotheque Nationale :
eighth or ninth century. Quoted by Martianay and
others, and collated for Bentley; St. Matthew edited
by Wordsworth (Old Latin Biblical Texts, i. 1883).
The second volume of a complete Bible. Contains the
Gospels, but only Matthew is Old Latin, the rest
being Vulgate with some Old Latin readings. Hort
reckons the whole with 27} as a modified Vulgate, but
Wordsworth concludes with regard to Matthew that its
basis was not Hieronymian, but a mixture of different
types of Old Latin texts, occasionally corrected to the
Vulgate, and containing a large peculiar element.

ο΄. Codex Sangermanensis II., formerly at St.
Germain des Prés, now in the Bibliotheque Nationale:
tenth century. Used by Sabatier, but not published.
Contains the Gospels, written in an Irish hand. The
text is a mixture of Old Latin and Vulgate.

hk. Codex Claromontanus, formerly at Clermont,
now in the Vatican: assigned by Burkitt (/ourn.
Theol. Studies, iv. 587, 1903) to the sixth century.
Contains Matthew (mutilated) of the Old Latin, the
other Gospels (in a different hand) of the Vulgate

1 See Buchanan’s detailed description of the MS. in /owrnal of Theo-
vogical Studies, vii. 99, 236 (1905-6), and Burkitt’s review, 26. 1x. 304
(1907).

4
ἴ
-
:
.
«
>
+
᾿
> 4
—
-
.
“ i
_ ‘
= Pe Sait < ie
» Vie

<2
oN ae

_* ¥/
᾿ =
= 3
ae
7 χ
τος ee am
sag oe. we
8. Ἄν Ά
oo -
εν Ἢ
- ἣν ἄν
-
-
- oiled -
aa Ww -
2 a δ ,“"
sé a Nat
v =
es: ‘
2
Chad ὃ. -
~~ :
>:
"
ἂν oe Wy 4
>
7 Φ
‘
7 “
-
᾿ -» a
{ ieee
Ν ' τῷ
Tedd ᾿

or

te

\
dt ee, aes

evr.

νι wheel y

᾽
See

᾿
ἕ
ῬῸ σ« NS κυ δ secre we

\
“7
ane

y Pe
Εν»:

"
i
@
my
a
᾿
!
>
r
_
= eee

ἣἢ
a

a iL

=
hw
ὮΝ

PIRATE: XV.

k pate δέ opt σταμῆ oe ©
2 RespOndrTETD NI TAUG UIE

<TERINENSTERNT Resinet sue
Pe FETSRLUOLUNTATENISPATRIS INE)
‘ ee se SERENE TC RORSTMA
Doe SHiLDIsE ITE

CODEX BOBIENSIS (2). Fifth Century (?).

(Scale 4:5. The page shown contains Matt. xii. 45-xiii. 1, verse 47 being

omitted, as in NBLI, the Old Syriac, and 223.)

: FAT GUACEST MATSEIY) SNAUTERD ae
_ TREMEJCLENTENSENSOIANUNY,
“MEMSCIPUS SUS SSIXIT ECCENMIA,

To face p. 203.

ν THE ANCIENT VERSIONS 203

text. Quoted by Sabatier; Matthew edited by Mai
(1828) and Belsheim (1892).

ἡ, Codex Vindobonensis, formerly at Naples, now
at Vienna: variously assigned to the fifth, sixth, or
seventh century. Written in silver letters on purple
vellum. Contains fragments of Luke and Mark.
Quoted by Bianchini, edited (in periodicals) by Alter
(1791-95), re-edited by Belsheim (1885).

2. Codex Saretianus, at Sarezzano: fifth century.
Written in silver letters on purple vellum. Contains
only fragments of John. Discovered in 1872, and not
yet published. Its text is said by White to agree
much with adde,

k. Codex Bobiensis, formerly at Bobbio, now at
Turin: fifth or sixth century.’ Written in rather
rough uncials, with many blunders. Traditionally
said to have belonged to St. Columban (543-615),
who founded Bobbio in 613. Originally contained
the four Gospels, but now has Mark viii—xvi. (ending
at xvi. δ), Matthew i—xv., with some mutilations.
Edited by Fleck (1837), by Tischendorf (in several
numbers of a periodical, 1847-49), and finally by
Wordsworth (Old Latin Biblical Texts, ii. 1886), with
facsimile and elaborate introductions by himself and
Sanday. The results of independent recollations by
C. H. Turner and Burkitt are given in the /ournal of
Theological Studies, v. 88 (1903). One of the most
important Old Latin MSS., containing the version in
what is probably its oldest form, which Sanday shows
to be near akin to the Bible used by St. Cyprian.
The Greek text underlying it has a large element in
common with D, but one almost equally large in

* Prof. Burkitt has argued in favour of a fourth-century date for this MS.
(/. Z..S. v. 107). The hand is peculiar (see Plate XV.), but at present palaeo-

graphical evidence for so early a date is wanting. The internal character
of the MS. certainly favours an early period.

204 CRITICISM OF NEW TESTAMENT cu.

common with xB; while in having the shorter alter-
native ending to Mark it agrees with L, W, Cod.
274, Gte.

ἀ Codex Rehdigeranus, at Breslau: seventh cen-
tury. Contains the Gospels, nearly complete, except
the last five chapters of John. Edited in part (Matthew
and Mark) by Scheibel (1763), and in full by Haase
(1865-66).

m., This letter is assigned to the treatise entitled
Speculum, falsely assigned to St. Augustine, which
contains extracts from all books of the New Testament,
except Philemon, Hebrews, and 3 John, in an Old
Latin text. Attention was first called to it by Mai,
who edited it from a single MS. at Rome of the eighth
or ninth century (1852); re-edited from six MSS. by
Weihrich (in the Vienna Corpus script. eccles. Lat. xii.
1887). Its text is probably of Spanish origin. It is
notable as containing the famous passage on the
Three Heavenly Witnesses (1 John v. 7), for which
there is practically no Greek evidence, though the
Latin evidence goes back to Priscillian, in the fourth
century.

nm. Fragmenta Sangallensia, at St. Gall: fifth or
sixth century. Fragments of the four Gospels, mainly
Matthew and Mark. Mentioned by Lachmann (1842),
and independently edited by Batiffol (1885) and White
(Old Latin Biblical Texts, ii. 1886), the latter with the
help of a transcript made by Tischendorf. Two leaves
are separated from the rest, being in the town library
of St. Gall, while two more are at Coire, where they
were seen and edited by E. Ranke (1872), and identified
as belonging to x by Batiffol. These leaves were
separately numbered as αὐ by White (of. czz.), who
formerly disputed the identification with z, and Gregory,
who would transfer the whole MS. to that letter; but

v THE ANCIENT VERSIONS 205

ma has the ‘prior claim, and it is also undesirable to
multiply subtenants of a single letter (as a', a’, etc.),
if it can be avoided. The text of these fragments is
closely akin to that of a.

o. A single leaf at St. Gall, of the seventh or eighth
century, containing the last seven verses of Mark, and
perhaps written to supply a lost leaf in 2, which leaves
off exactly at this place. Edited by Batiffol and
White, 2 supra.

2. Two leaves at St. Gall: seventh or eighth century.
Written in an Irish hand, as are many other MSS. in
this monastery. Contains part of John xi., and is
believed to have formed part of a service-book, not a
copy of the whole Gospels. Edited by Forbes (1864),
Haddan and Stubbs (1869), and White (dz supra).

g. Codex Monacensis, formerly at Freising, now at
Munich: sixth or seventh century. Written by a
scribe named Valerianus, probably in Germany. Con-
tains the Gospels, somewhat mutilated. Transcribed
by Tischendorf, and edited by White (Old Latin
Biblical Texts, iti. 1888). In text it holds a middle
position between ὁ and αὶ

γί, Codex Usserianus I., at Dublin: sixth or seventh
century. Contains the Gospels (Matthew, John, Luke,
Mark), but mutilated, especially in Matthew. Edited
by T. K. Abbott (1884). Its text is akin to that of
2, which appears to indicate the existence of an Irish
type of Old Latin text.

*. Codex Usserianus II., also at Dublin: ninth
century. Contains the Gospels, but only Matthew is
Old Latin, the rest being Vulgate. Collated by Abbott
in his edition of 7. Besides 7 and 7’, Old Latin
passages of an Irish type have been found by H. J.
Lawlor in a Vulgate MS, at Dublin, to which Gregory
assigns the symbol μ.

206 CRITICISM OF NEW TESTAMENT cu.

s. Fragmenta Ambrosiana, formerly at Bobbio, now
at Milan: sixth century. Consists of four leaves from
Luke. Edited by Ceriani (1861) and Wordsworth
(Old Latin Biblical Texts, ii. 1886). Its text is very
much mixed.

¢. Fragmenta Bernensia, at Berne: fifth or sixth
century. Palimpsest fragments of Mark i.—iii, very
difficult to decipher. Edited by Hagen (1884) and
Wordsworth (wbz supra). The text is somewhat akin
to that of ὦ, and in a rather less degree to 6 and αὶ

v. Fragmentum Vindobonense, at Vienna: seventh
century. A single leaf of John. Edited by Wordsworth
and White (of. czz. iii. 1888).

ὃ. Codex Sangallensis: the interlinear Latin text
of A, g.v. Chiefly notable for giving many alternative
renderings of the Greek words.

For the Acts the authorities are as follows :—

d=d of the Gospels.

e. Codex Laudianus: the Latin text of E,. Edited
by Belsheim, 1893.

g. Codex Gigas, formerly in Prague, now at Stock-
holm: thirteenth century. This MS. receives its name
from its huge size. Contains the whole Bible, but only
the Acts and Apocalypse are Old Latin. These books
were edited by Belsheim (1879). The survival of an
Old Latin text in so late a MS. may be explained,
as in the case of c, by the secluded character of the
country (Bohemia) in which it was written.

“σ΄. Fragmentum Mediolanense, at Milan: tenth or
eleventh century. A portion of a lectionary, containing
some verses of Acts vi.—viii., in a text akin to that of g.
Edited by Ceriani (1866).

h. Palimpsestus Floriacensis, formerly at Fleury,
now at Paris: sixth or seventh century. Contains
fragments of the Apocalypse, Acts, 1 and 2 Peter,

v THE ANCIENT VERSIONS 207

and 1 John. Quoted by Sabatier, edited by Belsheim
(1887), Berger (1889, with facsimile), and Buchanan
(Old Latin Biblical Texts, v. 1907; cf. Journ. Theol.
Studies, vii. 454, viil. 96, ix. 98), Gregory assigns it the
name veg instead of %. Like £& its text is closely akin
to that used by St. Cyprian, except in the Catholic
Epistles, which seem to have come into the African
New Testament at a later stage.

m =m of the Gospels.

p. This letter is given by Nestle, and by Sanday
and Turner in their promised edition of the New
Testament of Irenaeus, to a manuscript discovered
in 1896 by M. Samuel Berger,’ the great French
authority on the Latin Bible. It is a MS. now in
the Bibliotheque Nationale at Paris, but is shown by
an inscription in it to have belonged, at or soon after
its origin, to a monastery at Perpignan. It is a
thirteenth-century copy of the New Testament, con-
taining a Vulgate text throughout, except in Acts
i, I—xtiii. 6, xxviii. 16-30, and the Catholic Epistles,
which are’ Old atin. It ‘thus «resembles ic as ‘an
example of a late survival of the ancient version in a
remote district. The text of the Catholic Epistles has
been printed by Buchanan in Journ, Theol. Studtes, xii.
497 (1911).

s. Codex Bobiensis, formerly at Bobbio, then at
Naples, now at Vienna: fifth or sixth century. Palimp-
sest fragments of the last six chapters of Acts, James,
and 1 Peter, in one volume, with portions of Lucan
and other works. First noticed by J. von Eichenfeld
(1824), partly deciphered and published by Tischendorf
(1847), more fully but less accurately by Belsheim
(1886); finally, with further additions, by White

1 Un ancien texte latin des Actes des Apotres découvert... par M.
Samuel Berger (JVotices et Extraits des Manuscrits de la Bibliotheque
Nationale, xxxv. 169, 1896),

208 CRITICISM OF NEW TESTAMENT δὲ

(Old Latin Biblical Texts, iv. 1897). Agrees mainly
with g.

x’. Codex Bodleianus, at Oxford: seventh or
eighth century. Contains Acts, nearly perfect. Berger,
however, reckons this among the Anglo-Saxon MSS.
of the Vulgate. Described, with specimen facsimile, by
Westcott (Smith’s Dzct. of Bible, art. “ Vulgate”), who
says that it is a most valuable MS.

For the Catholic Epistles :—

ff. Codex Corbeiensis, formerly at Corbey, now
(like 22) at St. Petersburg: probably of the tenth
century. Contains the Epistle of James, together with
the unique Latin text of the Epistle of Barnabas, and
two other treatises. Published by Martianay (1695)
with 23. ; subsequently by Sabatier and Belsheim, and
finally by Wordsworth (Studia Bzblica, i. 1885). The
text is predominantly Old Latin, but has many agree-
ments with the Vulgate, and therefore probably re-
presents a comparatively late stage in the development
of the ‘Old, Latin text; The Epistle.of James, does
not seem to have been recognised in the Latin Church
before about the end of the fourth century, and it
will be observed that in this MS. it is associated with
uncanonical writers.

kR=h of Acts.

m=m of the Gospels.

g. Fragmenta Monacensia, at Munich: sixth or
seventh century. Contains fragments of I John, I
and 2 Peter (including 1 John v. 7, which is placed
after verse 8). Extracted from the bindings of books
by Ziegler, and edited by him (1877).

s=s of Acts. Its text here agrees generally with
the Vulgate (which in the Epistles was not greatly
altered from the Old Latin), with perhaps a strain of
late African text.

ν THE ANCIENT VERSIONS 209

For the Pauline Epistles :—

dad. Codex Claromontanus, the Latin text of τ᾽

6. Codex Sangermanensis, the Latin text of E,,

Ff. Codex Augiensis, the Latin text of F, The
text is said to be largely Vulgate.

g. Codex Boernerianus, the Latin text of G,, but
with a purer form of Old Latin text than its kinsman /

gue. Codex Guelferbytanus, at Wolfenbiittel: sixth
century. Fragments of Romans found in the palimp-
sest MS. containing ΡΟ of the Greek Gospels. Edited
by Knittel (1772) and Tischendorf (1855).

m =m of the Gospels.

vy. Codex Frisingensis, at Munich: fifth or sixth
century. Contains twenty-six leaves from Romans, I
and 2 Corinthians, Galatians, Ephesians, Philippians,
1 Timothy, Hebrews. Noticed by Tischendorf, and
published by Ziegler in 1876. Two leaves subsequently
identified by C. Schnorr were edited by Wolfflin in
1893.

v*, A single leaf, also at Munich, of the seventh
century, containing a few verses from Philippians and
1 Thessalonians, Published by Ziegler with 7

Yr, Codex Gottvicensis, at Gdttweig on the Danube:
sixth or seventh century. Contains small fragments of
Romans and Galatians. Edited by Ronsch (1879).

x’, Codex Bodleianus, at Oxford: ninth century.
Contains the Pauline Epistles, nearly complete. The
text often agrees with that of ὦ Described by West-
cott, with x of the Acts.

Finally, for the Apocalypse :-—

wee of the Acts.

#=kiol the Acts.

m =m of the Gospels.

In addition to these MSS. the Old Latin text of
the Apocalypse is to be found almost complete in the

Ρ

2r0. CRITICISM OF, NEW TESTAMENT |. ‘cu.

commentary of Primasius, an African father of the
sixth century, and considerable portions of it in the
writings of Cyprian (third century) and Tyconius (late
fourth century). Indeed, for all parts of the Old Latin
version the evidence of patristic quotations is especially
important, on account of the fragmentary and divergent
character of the manuscript authorities. The consider-
ation of these, however, belongs rather to the next
chapter.

Such, then, being the roll of witnesses to the Old
Latin text, what do they tell us of the history of the
version? At first sight the differences among them
are so great as to appear to confirm the complaint of
Jerome, when he undertook the revision of the Latin
Bible: “si enim Latinis exemplaribus fides est adhi-
benda, respondeant quibus; tot sunt paene quam
codices.”’ Augustine also complains of the “infinita
varietas” of the Latin translators, and asserts that
in the early days of Christianity any one who
possessed a Greek MS., and some degree of familiarity
with both languages, would take upon himself the
office of a translator.

Closer examination, however, has shown the
possibility of grouping and classifying the authorities,
at: least to some extent. On the one: hand, the
evidence that the version originated in Africa suggested
the possibility that some of the MSS. might be asso-
ciated with that country, especially as Augustine speci-
fically mentions “codices Afros” (Retr. i. 21, 3). On
the other hand, a well-known passage in Augustine
appears expressly to name an Italian type of text as
superior to all the multitude of others: “In ipsis
autem interpretationibus Itala ceteris praeferatur ; nam
est verborum tenacior cum perspicuitate sententiae”

1 22. ad Damasum, prefixed to the Vulgate New Testament.

ν THE ANCIENT VERSIONS 51

(De Doct. Christ. ii, 22). Starting from these indica-
tions, Hort, followed by the great majority of modern
scholars, divided the Old Latin version into three
types—(1) African, the most primitive ; (2) European,
including the greater number of extant MS. authorities ;
(3) Italian, the type praised by Augustine, and repre-
senting the text which formed the basis of Jerome’s
revision. Whether the African and European families
spring froma single origin, or are the products of two
distinct translations, was then and still remains an
undetermined point. On the one hand the differences,
both in readings and in renderings, are so very marked
as almost to preclude the possibility of a common origin ;
on the other, there occur here and there resemblances
of a kind which can hardly be otherwise accounted for.
The most careful and competent investigators, such as
Hort and Sanday, express themselves as uncertain ;
and lesser students can but accept their verdict. In
any case there has been a considerable amount of
intermixture of texts ; and in any case it is possible to
distinguish broadly between characteristically African
and characteristically European readings. With regard
to the Italian family, no such question arises. Some
scholars, indeed, deny its existence in the sense in
which the passage of Augustine is usually interpreted,
as will be shown more fully below; but if it exists, it
is the result of a revision of the European text, carried
out in North Italy early in the fourth century.

When we come to apply this classification to our
extant authorities, the African family can be tested and
identified by reference to the Biblical quotations in the

1 Bentley, who did not believe in an Italian recension of the Old Latin,
proposed to read δ z//a for /tala; but this is condemned (1) by the fact that
it involves the further alteration of sam into guae before it will make any
sense at all, and (2) by the exceedingly commonplace character of the
remark which is the result of these emendations.

219. CRITICISM) OF NEW TESTAMENT. cx.

African Fathers; not so much Tertullian, who seems
habitually to have made his own translations direct
from the Greek, as Cyprian, who quotes copiously
and textually. The MS. & (Codex Bobiensis) comes
closest of all to this standard; next to it, though at
some little distance, stands ὁ (Codex Palatinus) ; and a
somewhat later stage is represented by m. In the
Acts we have, besides 22, only the fragments of 4; in
the Epistles, only #, and the quotations of Priscillian,
which are most frequent here,and mainly agree with m ;
in the Apocalypse, Z and Primasius, and the quotations
in Tyconius, whose commentary on this book is to a
considerable extent preserved in the works of other
writers." The European family is much more fully
represented, though it must not be understood that all
the MSS. here mentioned form a homogeneous group.
On the contrary, they are full of diversities among
themselves ; but in spite of these diversities, the MSS.
a (with its close ally 2), ὁ (perhaps the most typical
representative of the family, the other MSS., however
much they differ among themselves, all seeming to
show kinship with this), c (a more mixed text), 7, 4, 2,
the fragments x, 0, 5, ¢ (ἡ and ~ giving an Irish sub-
species of the same family), and 2, all appear to belong
to the European family in the Gospels. To these may
be added the Latin version of Irenaeus. In the Acts
the European text is represented by g, 2”, 2, 5, and the
quotations of Lucifer of Cagliari; in the Catholic
Epistles by 27 (?); in the Pauline Epistles by nothing
that can be certainly established ; in the Apocalypse
by g (though Gregory regards this as rather Italian).
Finally, the Italian family is especially to be looked for
in f,and somewhat less in g, of the Gospels; in g of the

1 The African Old Latin has been edited recently by Hans von Soden
(Das lateinische Neue Testament in Afrika zur Zeit Cyprians, 1909), from
ὦ, 6, h, and Cyprian.

‘ THE ANCIENT VERSIONS 213

Catholic Epistles; and in 7 7, 75. of the Pauline.
Of the MSS. not accounted for in this classification,
d of the Gospels and Acts, and οἱ ¢, f, g of the Pauline
Epistles are the Latin texts of bilingual MSS., and have
suffered by their collocation with the Greek ; while the
characters of the remainder are mixed or uncertain.
Such is the view of the Old Latin version which
holds the field to-day, giving us the picture of a trans-
lation originally free in text and rough in rendering in
its African home, and progressively toned down in both
respects in Europe, having its readings brought more
into conformity with the Greek copies and its Latinity
smoothed down into more conventional forms, until the
process reached its consummation in an Italian revision,
which itself paved the way for the further revision
by Jerome, to which the Vulgate owes its origin.
Recently, however, Bentley’s disbelief in the very
existence of the “Itala” has been revived by Prof.
F, C. Burkitt. Prof. Burkitt’s main position’ is that
by his “Itala interpretatio” Augustine meant nothing
more nor less than the Vulgate, the New Testament
portion of which had been published for some ten years
at the time when he wrote; in support of which view
he appeals to Augustine’s own works. In short quota-
tions, such as a writer would naturally make from
memory, he often quotes the Old Latin, even to the
end of his life; but in longer passages, requiring
reference to a manuscript, he appears in his later
works to have used the Vulgate. Prof. Burkitt
refers especially to the De Consensu Evangelistarum,
written about A.D. 400, in which precise textual quota-
tions are constantly required ; and to the Acta contra
Felicem, a report of a trial for heresy at Hippo in 404,

1 See Zhe Old Latin and the [tala (Cambridge Zexts and Studies, iv. 3,
1896).

214 CRITICISM OF NEW TESTAMENT cu.

in which Augustine is said to have read a long passage
from a copy of the Gospels and another from a copy of
the Acts. These passages are quoted in full, and while
that from the Acts is undoubtedly of the African Old
Latin type, that from the Gospels is pure Vulgate. If
this were due to subsequent corruption in the MS. it is
unlikely that the other passage would have escaped
similar alteration; hence Prof. Burkitt is fully entitled
to argue that the Church at Hippo in 404 read the
Gospels in the Vulgate text, though it had not cared to
adopt the revised version of the Acts.

These are the two most striking testimonies to
Augustine’s use of the Vulgate (in the New Testament ;
in the Old Testament we know that he approved less
of Jerome’s more extensive alterations); but in his
later works in general Prof. Burkitt holds that his
Gospel quotations exhibit a Vulgate base with occa-
sional Old Latin readings. Hence he may naturally
have been referring to Jerome’s work when he com-
mended the “ Itala interpretatio” as “ verborum tenacior
cum perspicuitate sententiae.” And if it be asked, by
‘way of answer, “ But have we not concrete evidence of
the existence of the Itala in such MSS., as f and g?”
Prof. Burkitt is apparently prepared to deny to /,
at any rate, the character of an Old Latin MS. at all,
regarding it as a post-Vulgate text which owes its
peculiarities to the corrupting influence of the Gothic ;?
while g would no doubt be classified as merely a variety
of the European family.

Prof. Burkitt’s theory has not as yet met with
either acceptance or rejection, and indeed neither is
possible without a detailed examination of the Biblical
quotations in Augustine’s later works; but the case in

1 See Journal of Theological Studies, i. 129-134 (in a review of Words-
worth and White’s Vulgate).

ν THE ANCIENT VERSIONS 215

its favour appears strong. It may be observed, how-
ever, that even if Augustine’s testimony to the existence
of a specially Italian pre-Vulgate text be thus disposed
of, the facts with regard to the Old Latin version
established by the manuscripts themselves remain un-
altered. We have on the one hand a group of texts
which is shown by the evidence of the African Fathers
to be closely connected with Africa, and on the other
we have a number of texts in a somewhat later stage
of development, and showing among themselves signs
of progressive revision, until at last we reach texts
which may have furnished a basis for Jerome’s work.
The classification of our MSS. consequently remains
the same, whether the final stage is entitled Italian or
not. Whether / represents the final stage is a further
question, on which the verdict of Bishop Wordsworth
and Professor White, the editors of the Vulgate, who
have selected it as the nearest extant representative of
the text upon which Jerome worked, will rightly have
great weight.’ Prof. Burkitt’s suggestion of corrup-
tion from the Gothic is not, przma facze, attractive. It
does not seem likely that the owner or transcriber of a
Latin MS. would introduce corrections from what he
would consider as the less authoritative Gothic text,
and it would seem most natural to explain any co-
incidences between / and the Gothic as due rather to
the influence of the Latin text on the Gothic during
‘the occupation of Italy by the Goths, than to the
reverse process.

Looking now finally at the Old Latin version as a
whole, its text is found to be of a very early character.
It belongs to, and is a principal member of, that class
of authorities (the 6-type, as we have called it) which is

1 Prof. Souter has recently (/. 7S. xii. 583) shown good grounds for
holding that, at any rate in Luke, Jerome used a text practically identical

with that of a.

2216. CRITICISM OF NEW TESTAMENT cx.

distinguished by the boldest and most striking departures
from the received text. It is found in company with
the Codex Bezae and its attendant group of minuscules,
and with the Old Syriac. It shares with these the
additions to, and (at the end of St. Luke) the omissions
from, the received text, of which some mention has
been made on p. 93. It shares also their constant
verbal divergences, suggestive of a time when strict
accuracy in the transmission of the sacred texts was
not much regarded. The history of this type of text,
and its position in modern textual criticism, will have
to be considered in the final chapter; at present we are
only concerned in pointing out the witnesses to its
existence. It is, naturally, in the earlier or African
family of the Old Latin that these characteristics are
most marked. Successive revisions gradually toned
them down or removed them, until in the last days of
the existence of this version it approximates to the
type of text which we know to-day as the Vulgate.

6. The Vulgate.—The chaotic character of the Old
Latin texts, still evident to us in the scanty remnants
which have survived to our time, led at the close of the
fourth century to the production of the great version
which supplied all Western Christendom with its Bible
for over a thousand years, and still is the Bible of that
great branch of the Church which owns allegiance to
Rome. It is to Damasus, Pope from 366 to 384, that
the credit of its origin is due. Impressed with the
uncertainty as to the true text which the variations in
the manuscripts of the day made so evident, he applied
to Eusebius Hieronymus (better known to us as Jerome)
to undertake an authoritative revision of the Latin Bible.
No living scholar was equally competent for the task.
Born about 345 at Stridon in Pannonia, in the region
of modern Trieste, he had devoted himself before

v THE ANCIENT VERSIONS 247

everything to Biblical studies. After many years at
Rome and in Gaul, in 372 he visited the East, spend-
ing ten years there, mainly in the study of Hebrew,
before returning to Italy. It was on his return, about
the year 382, that the request of Damasus laid upon
him the great work of his life, the production of an
authoritative Latin Bible. In its beginnings the under-
taking was not so great as it subsequently became.
The Pope’s invitation to him was that he should revise
the existing text by reference to the original Greek in
the New Testament, and to the Septuagint in the Old
Testament ; and in the part of the work first taken in
hand, the New Testament and the Psalter, this was all
he did. It was only later that, becoming dissatisfied
with the process of revision, he laid aside all that he
had done with regard to the Old Testament, and under-
took a new translation of it from the Hebrew. With
this we have nothing to do here; and in the New
Testament he was emphatically a reviser, not a new
translator.

The revised version of the Gospels, which (with the
Psalter) were Jerome’s first care, appeared in 383;
and his preface tells us something of his principles and
methods of work. Only in passages of some im-
portance did he think himself entitled to introduce
alterations; smaller blunders and inexactnesses re-
mained uncorrected, in order that the familiar language
of the Bible should be left untouched, so far as possible.
For the purpose of comparison and correction he used
several Greek MSS., the character of which will be
considered later. In the other books of the New
Testament, a revision of which followed very shortly
on that of the Gospels, his work was more perfunctory.
Some, indeed, have doubted whether he revised any
part of the New Testament except the Gospels, and

a8 CRITICISM OF NEW) TESTAMENT νη,

point to the fact that, contrary to his habit, he wrote
no preface to these books ; but his own statement that
he revised them all is express.

So far, then, as the New Testament is concerned,
the Vulgate is merely a revision of the Old Latin,
fairly thorough, though not pedantically exact, in the
Gospels, but only superficial in the remaining books,
One result of this conservative treatment was that the
new version met with general acceptance—far more so
than was the case with Jerome’s wholly new translation
ot) the’ Old Testament; Jt did not supersede ithe
older version at once, as is shown by the fact that all
the extant copies of the Old Latin, except possibly the
Codex Vercellensis, were written subsequently to the
publication of Jerome’s work; but it was adopted, no
doubt by the order of Pope Damasus, as the official
Bible of the Church of Rome, and gradually won its
way to universal acceptance among the Latin-speaking
peoples of the West. It would be going outside the
scope of this book to trace the fortunes of the Vulgate
throughout the Middle Ages." With success and
general adoption came, as was natural, extensive
corruption of the text through the carelessness or
rashness of scribes and editors. From time to time
attempts were made to purify it. Bede and his
contemporaries and successors in northern England
reached a high level of Biblical scholarship. The
great revival of France under Charlemagne led to two
revisions of the Vulgate text: one the private under-
taking of Theodulf, Bishop of Orleans, about 800; the
other the work of the English scholar Alcuin, whom
Charlemagne invited to France to superintend the
education of his people. But even Alcuin’s official

1 For this the reader cannot do better than consult M. Berger’s
Histoire de la Vulgate pendant les premiers siécles du moyen dge (Paris,
1893), an admirable and fascinating study of a most intricate subject.

ν THE ANCIENT VERSIONS 219

edition had but a temporary effect in staying the
progress of corruption, and it was not until four
hundred years later that any vigorous effort was made
for a reform. In the thirteenth century great activity
prevailed in the production of copies of the Bible, as
the numbers: of manuscripts of this date in our public
libraries to-day amply testify; and this activity was
largely stimulated and directed by the doctors of the
recently founded University of Paris. To one of
them, Stephen Langton, afterwards Archbishop of
Canterbury and leader of the barons in their contest
with King John, we owe our modern division of the
Bible into chapters ; to others, known or unknown, we
owe the redaction of the Vulgate text, not indeed into
a scientifically accurate form, but substantially into
that form in which it now circulates in the printed
copies officially sanctioned by the heads of the Roman
Church.

With the invention of printing came naturally the
demand for a printed Bible; or rather we may say
that the supply preceded the demand. The first book
committed to the press in Europe was the Latin Bible,
which appeared in 1456 in the splendid edition of
Gutenberg, now commonly known as the Mazarin
Bible. For the purposes of this edition, however, no
critical examination of texts had been undertaken. It
was merely an example of the current text of the
fifteenth century. The editors employed by Cardinal
Ximenes in the preparation of his great Complutensian
Polyglott (New Testament printed in 1514, published
in 1522) devoted considerable labour to the Latin
portion of their text, and Erasmus quotes readings from
various Vulgate MSS. examined by himself; but the
first really critical editions of the Latin Bible are those
published by Stephanus at Paris between 1528 and

220 CRITICISM OF NEW TESTAMENT os cu.

1546, of which the best is said to be the fourth, issued
in 1540. This, in which seventeen MSS. are quoted,
is taken by Wordsworth and White as the typical
representative of Stephanus’s editions. In 1546 the
Council of Trent passed a decree, prescribing the
Vulgate as the standard text of the Bible to be used
in the services of the Church, and desiring that an
accurate edition of it should be printed ; and it may
have been in consequence of this pronouncement that
John Hentenius in 1547 published a new edition at
Louvain, in which he made use of no less than thirty-
one MSS. This edition was very commonly used in
church services, but was never officially adopted to the
exclusion of others; and it was not until the accession
of Sixtus V. in 1585 that steps were taken for the
preparation of an authorised text. This Pope took up
the matter vigorously, both by appointing a committee
of scholars to undertake the work, and by devoting
himself strenuously to the task of revision. In 1590,
accordingly, the Sixtine Vulgate issued from the
Vatican press, accompanied by a papal bull declaring
1 τὸ ‘be the “true; legitimate; authentic,” and” im-
dubitable” text of the Holy Scriptures, which alone
was to be regarded as authoritative. |
Unfortunately for the success of this edition, Sixtus
died in the year of its issue; and Clement VIII., who
was elected Pope in 1592, promptly decreed its recall
and suppression. The pretext for this action was the
inaccuracy of its printing ; and in justification of this
plea it has been pointed out that many corrections
were made in it, after the sheets were printed off, by
means of hand-stamped type. Mr. H. J. White,
however, who has recently examined the Sixtine
Vulgate minutely in the Gospels, declares that in these
books, at any rate, the charge of inaccurate printing

v THE ANCIENT VERSIONS 221

cannot be sustained, and that the corrections are
almost wholly in the introductory prefaces, the text
itself being in fact printed with unusual care and
accuracy. It therefore appears that the other reason
which has been suggested is the real one, namely, that
the Jesuits had not forgiven Sixtus for placing one of
Bellarmin’s books on the /udex, and took this method
of revenging themselves. Certain it is that the alter-
native edition issued by Clement’s authority in 1592
was accompanied by a preface by Bellarmin himself,
in which the allegation of inaccurate printing is made,
and it is even asserted that Sixtus himself had intended
to recall his own edition, but was prevented by death.
The Clementine edition is estimated to differ from the
Sixtine in about 3000 places, approximating rather to
the text of Hentenius, while the Sixtine is nearer to
that of Stephanus ; nevertheless, in order to disguise the
conflict of popes, it was sometimes issued under the name
of Sixtus instead of, or as well as, that of Clement.
Under one name or the other, the edition of 1592
became the standard text of the Roman Church, and
has so continued from that day to this. Not only so,
but, in order the more to ensure its authority, the bull
with which Clement accompanied its issue forbade the
slightest alteration in it, or any insertion of various
readings in the margins. By this measure the textual
study of the Latin Bible was effectually killed in the
Church of its home,’ although increasing knowledge has
shown beyond the possibility of doubt that the text
issued by Clement is by no means an accurate re-
presentation of the version as it left the hands of

1 An exception must be made to this statement in the case of the
editors of Jerome, Martianay in 1693 and Vallarsi in 1734, who were able
to treat the Vulgate as part of the saint’s literary works, and give some-
thing like a critical edition of it. Vercellone in 1860 published a collection
of Variae Lectiones, but without a continuous text.

222) CRITICISM OF |NEW TESTAMENT “δε.

Jerome. For over three hundred years it was left
to scholars of other countries, and of other branches
of the Christian Church, and especially to those of
England and Germany, to undertake the task of re-
covering the true text of the Vulgate. Bentley, with
the assistance of his colleague John Walker, made
great preparations for a critical edition, making or
procuring collations of a large number of MSS.; but
he rashly committed himself beforehand to the assertion
that the Greek and Latin texts of the Bible could be
shown to be identical down to their smallest details,
and it is supposed that the discovery of the fallacy of
this assertion, forced upon him by increasing know-
ledge, was a main cause of his failure to bring his
work to any conclusion. Nothing was printed by him,
and since Walker had died shortly before his master
and colleague, the work came to nothing, and the
collations which they had accumulated have since
remained in the library of Trinity College, Cambridge,
almost unused from that day until, in this present
generation, the work has been resumed by another pair
of scholars from the sister university.

Before reaching, however, this stage in the history
of the Vulgate text, mention is due of the labours of
Lachmann, who, carrying out the principles of Bentley
to a happier result, issued in 1842-50 a critical edition
of the Latin and Greek New Testament which was far
in advance of anything that had previously been done.
The two texts are printed together, the upper part of
the page containing the Greek and the lower the
Latin, with the authorities between them; the Latin
text being based upon two excellent authorities, the
Codex Fuldensis and the Codex Amiatinus (see below),
though the latter, unfortunately, was only accessible to
him in an imperfect collation.

ν THE ANCIENT VERSIONS 223

Lachmann’s authorities, though good, were few, and
his text could not be considered a fully critical edition
of the Vulgate. Such an edition, however, was at last
undertaken by the Rev. John Wordsworth of Brasenose
College, Oxford, subsequently Bishop of Salisbury, with
whom has been associated during the greater part of
the work the Rev. H. J. White of Merton College,
Oxford (now of King’s College, London). The first
part of this, which will long remain for critical purposes
the definitive edition of the Vulgate New Testament,
appeared in 1880, after twelve years of preparation ;
while another nine years saw the completion of the
text of the four Gospels, together with elaborate 270-
legomena and epfilogus. The Acts followed in 1905,
and it is understood that the Epistles are in a forward
state of preparation; but the work suffered a serious
blow through the death of Bishop Wordsworth in 1911.
Meanwhile it is worth noticing, for the sake of students,
that two handy editions, embodying the results of the
Oxford Vulgate, have recently appeared. The first,
edited by Nestle in 1906, gives the ordinary Clementine
text, with a full collation of the Sixtine edition and the
readings of Wordsworth and White in footnotes as far
as Acts, and selected readings from Lachmann (1850),
Tischendorf (1854), and the codices A and F for the
rest of the New Testament. The second, by White
(Oxford, 1912), gives the text of the larger edition for
the Gospels and Acts, and a provisionally revised text
of the remaining books, with a select apparatus criticus}
This should be an exceedingly serviceable edition for
general use, and may be commended to the notice of
schools and colleges where the New Testament is not
read in Greek.

1 The apparatus gives the important readings of the manuscripts A C D
F GH V, and in the Gospels of M Z also (see descriptions below) ; it also
gives the readings of the Sixtine and Clementine editions.

224: CRITICISM OF NEW) TESTAMENT © ‘ca.

Wordsworth and White’s edition, so far as the
Gospels are concerned, is based upon a collation of
thirty selected manuscripts, accompanied by the full
text of the Codex Brixianus, as best representing (in
_ the opinion of the editors) the type of Old Latin text
upon which Jerome’s revision was based.’ A description
of these manuscripts, and of the general conclusions to
which the editors have come with regard to Jerome’s
work, will be given below; but first, for completeness’
sake, it is right to refer to the great enterprise
which, to the credit of Pope Pius X., has lately been
undertaken by the Vatican. This is a complete new
edition of the Vulgate, for which materials are being
collected on the largest scale, under the editorship of
Abbot Gasquet. In view of the work of Wordsworth
and White, the Vatican authorities have decided, wisely
as well as courteously, to direct their attention in the
first instance to the Old Testament; and this will in
itself be the work of a generation. Meanwhile, except
so far as the researches thus set on foot may, as a by-
product, throw light on the textual history of the New
Testament also, they do not directly concern us here.

The task of editing even the New Testament alone
is no light one. The number of extant MSS. of the
Vulgate is immense. No complete catalogue of them
exists, and the precise total is unknown; but it must
greatly exceed the 4000 (approximately) of the Greek
New Testament. Gregory gives (in addition to the
46 MSS. quoted by Tischendorf) a catalogue of 2472
copies, which he calls merely “ pauca ex innumera multi-
tudine codicum”; while White estimates that there are
at least 8000 scattered throughout the various libraries
of Europe, out of which he gives a select list of 181,
with brief descriptions. For the ordinary purposes of

1 On the accuracy of this opinion, see above, p. 214 f.

Ἢ THE ANCIENT VERSIONS 225

a student, however, it will be sufficient to give some
account of the forty copies comprised in the list of |
Wordsworth and White, using the notation which they
have adopted in their edition.

A. Codex Amiatinus, in the Laurentian Library at
Florence. The history of this magnificent MS. has
been made clear by the ingenuity of De Rossi and
Hort. The former observed (as others had done
before him without carrying the matter further) that in
some dedicatory lines at the beginning, which purported
to state that the volume was the gift of Peter the
Lombard to the monastery of Monte Amiata, these
names were written over erasures and spoilt the metre
of the verses; and by a brilliant conjecture he sub-
stituted for them the names of Ceolfrid of England
and the See of Rome respectively. These conjectures
commanded confidence in themselves, being suitable to
the context and to the known history of Ceolfrid, but
were converted (with slight alterations) into certainties
by Professor Hort, who showed that the verses them-
selves were already extant in an anonymous life of ©
Ceolfrid, used by Bede, where they are expressly
stated to have been prefixed to a copy of the Vulgate
written at Ceolfrid’s order for a gift to the Pope. The
history of the MS. was then clear. It was written in
the north of England, at either Wearmouth or Jarrow,
of both of which monasteries Ceolfrid was abbot, and
was copied from MSS. brought from Italy either by
Ceolfrid himself or his master, Benedict Biscop, or,
perhaps more probably (see description of Y below),
by Theodore of Tarsus when he came to England to
be Archbishop of Canterbury in 669. It must have
been written quite early in the eighth century, and was
taken by Ceolfrid as a present to Pope Gregory in 716.

1 For the fullest account see White in Studza Biblica, ii. 273 ff.

Q

226 CRITICISM OF NEW TESTAMENT οεη.

Ceolfrid himself died on the way, but the manuscript
was carried on to Rome by some of his companions,
was subsequently given to Monte Amiata, was recalled
temporarily to Rome to be consulted for the Sixtine
edition of the Vulgate, and finally found a home at
Florence, where the inquiring visitor may now see it.
It is a huge and splendid volume, measuring 1 ft.
74 in. in height by 1 ft. 14 in. in width, written in
double columns in a large and beautiful uncial hand.
The text is not written continuously, but in short
clauses corresponding to breaks in the sense (techni-
cally known as cola and commata, i.e. clauses and
sub-clauses). This system of division goes back at
least to the middle of the fifth century. The text
is also divided into paragraphs, the Gospels having
the Ammonian sections (see p. 68, note), while in the
Acts a section-numeration is found which occurs also
in the Codex Fuldensis (F, see below) and in the
Greek MSS. ws and B (see p. 81). The MS. contains
the whole Bible in an excellent form, and is generally
regarded as the best authority extant for the Vulgate
text. It was collated by Fleck in 1834 (published in
1840), by Tischendorf in 1843, by Tregelles in 1846,
and the New Testament published in full by Tischen-
dorf in 1850. Finally it was re-collated by Mr. H. J.
White in 1887 for the Oxford edition, in which it
heads the most important and trustworthy group of
MSS. Specimen facsimiles of it are given by the
Palaeographical Society (ii. 65, 66)."

Recently Dom Chapman appears to have succeeded
in demonstrating that the text of A, and therewith of

1 A single leaf of a sister MS. (containing 3 Kings xi. 29-xii. 18) was
discovered at Newcastle by Dr. W. Greenwell, identified by Mr. C. H.
Turner (Journ, Theol. Studies, x. §30, 1909) and published in facsimile by
the New Palaeographical Society (Part vii., plates 158-9, 1909). It is now
in the British Museum.

V THE ANCIENT VERSIONS 227

the Northumbrian family in general, is derived from
the edition of Cassiodorus, the well-known statesman
and scholar, who ended his life as abbot of a Calabrian
monastery, founded by himself, about 580; while he
has, rather more precariously, made out a case for con-
necting Cassiodorus’s text with a manuscript in the
library of Eugippius at Naples, which was said to have
belonged to Jerome himself (see description below of
“P, in which occurs the note mentioning this manu-
script).

B. Codex Bigotianus, formerly at Fécamp, now at
Paris: eighth or ninth century. Collated by Walker
and Wordsworth. Contains the four Gospels. Its
text shows a mixture of Irish and French influences,

a. Codex Beneventanus, formerly at Beneventum,
now in the British Museum; written for an abbot
Atto, who has generally been supposed to be identical
with an Atto who was abbot of St. Vincent, near Bene-
ventum, from 739 to 760. Berger, however, states that
the text is French rather than Italian, and believes that
it was written in France, probably in the ninth century.
Collated by Bentley, and in Luke and John by White.
It contains the four Gospels.

C. Codex Cavensis, at La Cava in South Italy:
ninth century. Written in Spain, in a small Visigothic
hand, and contains the whole Bible. It is a typical
representative of the Spanish type of text, in which
respect it is associated with T. A transcript of it,
made about the beginning of the nineteenth century, is
in the Vatican. Collated by Wordsworth.

D. Codex Dublinensis, at Trinity College, Dublin,
known as the Book of Armagh: eighth or ninth
century. Contains the New Testament (including

1 See Notes on the Early History of the Vulgate Gospels, by Dom 7.
Chapman (Oxford, 1908), supplemented by an article on “" Cassiodorus
and the Echternach Gospels,” in the Revue Bénédictine, 1911, p. 283.

228 ‘CRITICISM.'OF NEW TESTAMENT τ

the apocryphal Epistle to the Laodiceans), Tran-
scribed by G. M. Youngman for the Oxford edition.
It belongs to the Irish type of Vulgate text, which
is characterised by small additions and insertions, but
is based upon a good and ancient tradition. It shows
signs of having been corrected from Greek MSS. akin
to the Ferrar group (see p. 131).

A. Codex Dunelmensis, in the Cathedral Library
at Durham: seventh or eighth century. Traditionally
said to have been written by Bede. Contains the
Gospels. Its text is akin to that of A, and, like that
MS., it may very probably have been written at Wear-
mouth or Jarrow. Collated by Bentley. Used by
Wordsworth for John only.

FE. Codex Egertonensis, formerly at Tours, now
in the British Museum: ninth century. Contains the
Gospels, considerably mutilated, written in Caroline
minuscules, but with ornamentation in the Irish style;
its text also is of the Irish type, akin to that of D.
Collated by Youngman.

ΞῥΡ. Codex Epternacensis, formerly at Echternach,
now at Paris: ninth century, although it contains a note
(no doubt copied from its ancestor) affirming that it
was corrected in A.D. 558 from a MS. attributed to St.
Jerome’s own hand. It contains the Gospels, in a very
mixed text. Berger classes it with the Irish family,
but Wordsworth considers this as true mainly in such
matters as orthography, its readings approaching rather
to those of B and Z. The marginal readings are
oftener Irish than those of the main text. Collated
by White.

F. Codex Fuldensis, at Fulda in Germany: written
A.D. 541—546 at the order of Bishop Victor of Capua.
Contains the whole New Testament, the Gospels being
arranged in a continuous narrative, according to the

v THE ANCIENT VERSIONS 229

plan of Tatian’s Diatessaron, a copy of which, in a
Latin translation, had fallen into the Bishop’s hands.
The author’s name was wanting, but Victor rightly (as
we now know) guessed it to be Tatian’s, and took it as
his model, substituting, however, the Vulgate text for the
Old Latin of his original. Among the Epistles that to
the Laodiceans is inserted. Its text is very good, being
akin to that of A. It was the principal authority used
by Lachmann for his Latin text, and was published in
full by E. Ranke in 1868 (with facsimiles).

G. Codex Sangermanensis, formerly at St. Germain
des Prés: eighth or ninth century. Described above
as g' of the Old Latin version. In the Old Testament
the text is wholly of the Spanish type; but in the New
Testament (where it is not Old Latin) it presents a
mixture of French and Irish readings. Berger (who
gives a minute description of it, Hzst. de la Vulgate, pp.
65-72) thinks it was written in the region of Lyons ;
and it is noticeable that it contains a curious collection
of sortes for purposes of divination, written in the
margin of St. John’s Gospel, which also occur, in
Greek, in Codex Bezae. Collated by Walker, and
again by Wordsworth and Youngman. Wordsworth
and White consider it especially important in Acts,
where they prefer it even to A.

H. Codex Hubertianus, formerly at the monastery
of St. Hubert in the Ardennes, now in the British
Museum: ninth century. It contains the whole Bible,
written in a small Caroline minuscule, with three
columns to the page. Berger and Wordsworth differ
somewhat in their description of its text, the former
treating it as a representative of the edition of Theo-
dulf (see above, p. 218), though somewhat negligently
written, while the latter describes it as closely akin to
A and Y, though with occasional affinities to the Theo-

230 CRITICISM OF NEW TESTAMENT cx.

dulfian Bible, and in its corrections being thoroughly
Theodulfian. This divergence may be explained by
the fact that Theodulf’s text in the Gospels is of the
Anglo-Saxon type, though elsewhere it is predomin-
antly Spanish. Collated by Wordsworth. Facsimile
in Hacszmiles of Biblical MSS. in the British Museum,
PLOY.

©. Codex Theodulfianus, the property success-
ively of the Cathedral of Orleans (eleventh century),
the family of Mesmes, and the Bibliotheque Nationale
at Paris: ninth century. Contains the whole Bible,
written in a small hand much resembling that of H,
and, like it, is of the edition of Theodulf. Indeed
Berger believes it to have been actually prepared under
Theodulf’s directions, the many corrections which are
made in the margins and between the lines being due
to his own editorial work. It may be doubted, how-
ever, whether the writing is as early as the time of
Theodulf. Collated by C. Wordsworth and H. J.
White. Facsimile in Delisle’s Album Paléographique,
Pl. 18 (1887).

I. Codex Ingolstadiensis, formerly at Ingolstadt,
now at Munich: ninth century. Contains the Gospels,
St. Matthew being much mutilated. Tischendorf (who
assigned it to the seventh century) made a colla-
tion, which, having been purchased by the Oxford
University Press, was used by Wordsworth, with the
result that his representation of its readings is not
infrequently inaccurate (see Wordsworth and White,
1°67 3):

J. Codex Forojuliensis, mainly at Cividale in Friuli,
partly at Venice and Prague; sixth or seventh century.
Contains the Gospels, with a text of the class which is
headed by Z. The Friuli portion (Matthew, Luke, and
John) was edited by Bianchini (1749), the Prague

ν THE ANCIENT VERSIONS 231

portion by Dobrowski (1778), while the Venice frag-
ment is almost wholly illegible.

K. Codex Karolinus, in the British Museum; ninth
century. A huge MS., containing the whole Bible
according to the edition of Alcuin (see above, p. 218).
Consequently it agrees generally with V, which is also
Alcuinian, though of the two V is the better; and both
have affinities to A and Y, since Alcuin made use of
MSS. from Northumbria. Collated by Youngman and
White for the Oxford Vulgate. Facsimile in Facstmzles
of Biblical MSS. in the British Museum, Pl. XIV.

L. Codex Lichfeldensis, formerly at Llandaff, now
in the Chapter Library at Lichfield, known as the
Gospels of St. Chad; seventh or eighth century.
Contains Matthew and Mark and Luke 1.—iii. 9, with
illuminations in the Celtic-Saxon style. Its text also
belongs to the Celtic family, like that of D. Collated
by Scrivener (1887).

M. Codex Mediolanensis, in the Ambrosian Library
at Milan; sixth century. Contains the Gospels, in a
somewhat independent type of text, agreeing now with
one group, now with another. It appears to have been
corrected from the Greek. Transcribed by Padre F.
Villa for the Oxford Vulgate.

M. Codex Martini-Turonensis, formerly in the
monastery of St. Martin at Tours, now in the Public
Library ; eighth or ninth century. Contains the
Gospels, written in golden letters, in a text akin to that
of Alcuin, but with more of an Irish element in it.
Used by Sabatier, and from him by Tischendorf.
Collated by Walker and Youngman.

O. Codex Oxoniensis, formerly in the monastery of
St. Augustine, Canterbury, now in the Bodleian Library,
Oxford, and known as St. Augustine’s Gospels ; seventh
century. Contains the Gospels, in a mixed text, showing

232 CRITICISM OF MEW TESTAMENT . cx.

Irish influence, akin to that of X. Collated by Words-
worth, Madan, and Youngman.

P. Codex Perusinus, in the Chapter Library at
Perugia ; sixth or seventh century. A fragment of a
MS. on purple vellum, containing only Luke i. 26—xii. 7,
with many mutilations. In spite of its age, its text,
according to Wordsworth, is not very good. Edited
by Bianchini (1749).

Q. Codex Kenanensis, the famous Book of Kells,
at Trinity College, Dublin ; seventh or eighth century.
Contains the Gospels, in an Irish hand, and with the
most elaborate and beautiful Celtic decorations. The
text is, naturally, of the Irish type, with a peculiar
tendency to duplicate renderings. Collated by Dr. T.
K. Abbott (1884). Facsimiles in Pal. Soc. i. 55-58,
88, 89.

R. Codex Rushworthianus, in the Bodleian Library,
known also as the Gospels of Mac Regol, from the
name of the scribe, who died in A.D. 820. Contains
the Gospels, with an interlinear English gloss (as in Y) ;
Matthew in the Mercian dialect, the other Gospels in
Northumbrian. Written by an Irish scribe, and has an
Irish text, with corrections apparently from the Greek.
Collated by Stevenson and Waring (1854-65) and
Skeat (1887).

S. Codex Stonyhurstensis, formerly at Durham,
now at Stonyhurst College ; seventh century. Tradi-
tionally said to have belonged to St. Cuthbert and
to have been preserved with (or in) his coffin. A
beautiful little copy of the Gospel of St. John, with
an excellent text, akin to that of AAY, which like-
wise belong to the north of England. Collated by
Wordsworth.

T. Codex Toletanus, formerly at Seville, then
Toledo, now in the National Library at Madrid; eighth

7 THE ANCIENT VERSIONS 233

century (cf. Berger, Hest. de la Vulgate, pp. 13, 14).
Contains the whole Bible, in the Spanish type of text,
so that it is an ally of C. A collation was made by
C. Palomares for the Sixtine edition of the Vulgate, but
was received too late to be used. This collation was
printed by Bianchini in 1740. Re-collated for the
New Testament by Wordsworth.

U. Fragmenta Ultratraiectina, at Utrecht ; seventh
or eighth century. Fragments, containing only Mt. i. 1--
iii. 4, Jn. i. 1-21, bound up with the well-known Utrecht
Psalter ; written in a hand closely resembling that of
the Amiatinus, and evidently produced in the same
scriptorium. Published in facsimile with the Utrecht
Psalter (187 3).

V. Codex Vallicellianus, formerly in the Oratory of
S. Maria in Vallicella in Rome, now in the Biblioteca
Vittorio-Emanuele ; ninth century. Contains the whole
Bible written in three columns to the page. The text
is that of the edition of Alcuin ; see description of K,
above. Collated by Wordsworth.

W. Codex Willelmi de Hales, formerly at Salisbury
(having been written for T. de la Wile, master of the
schools there), now among the Royal MSS. in the
British Museum ; written in 1254. Contains the whole
Bible. Employed by Wordsworth and White in their
edition as an average representative of the thirteenth-
century Vulgate (see p. 219), which is closely related to
the printed Vulgates of the sixteenth century. Collated
by White and Youngman.

X. Codex Corporis Christi Cantabrigiensis, formerly
at St. Augustine’s, Canterbury, now among Archbishop
Parker’s MSS. at Corpus Christi College, Cambridge ;
seventh century. Contains the Gospels. Its original
text is closely akin to that of O, formerly its companion
at Canterbury ; but it has been corrected into close

234 CRITICISM OF NEW TESTAMENT ὦ

agreement with AY. Collated by A. W. Streane for
the Oxford Vulgate.

Y. Codex Lindisfarnensis, the celebrated Lindisfarne
Gospels, formerly at Lindisfarne and Durham, now in
the British Museum; late seventh or early eighth
century, being written in honour of St. Cuthbert, who
died in 687, and preserved with his body. Written in
a splendid half-uncial hand, and adorned with beautiful
decorations in the Celtic-Saxon style. A table of
festivals on which special lessons were read shows that
it must have been copied from a Bible used in a church
at Naples, no doubt one brought to England by the
Neapolitan abbot Hadrian, the companion of Archbishop
Theodore. Its text is closely akin to ‘that of ἍΙ
Edited by Stevenson and Waring (1854-65) and Skeat
(£887). ' Faesimiles in» Par, ) See Gt, 65,66) tana
Facsimiles of Biblical MSS. in the British Museum,
Pl XI,

Z. Codex Harleianus, formerly in the Bibliotheque
du Roi in Paris, now among the Harley MSS. in the
British Museum ; sixth or seventh century. A beauti-
fully written little copy of the Gospels, standing at the
head of the group of MSS. opposed to AY, and therefore
tracing its origin to a different archetype. Collated by
Griesbach and others, finally by White. Facsimiles in
Pal, Soc. (i. 16) and Facstmiles of Biblical MSS. tn the
British Museum, Pl, 1X.

In addition to these, mention should perhaps be
made of the Hamilton Gospels, a very handsome MS.
formerly in the Hamilton library, then in that of Mr.
T. Irwin of Oswego, and now in the collection of Mr.
J. Pierpont Morgan, in New York. It is written in
gold letters on purple vellum. The writing, which is
of the eighth century, is not distinctively English, but
Mr. H. C. Hoskier, who has made a most elaborate

PLATE. ΣΕΥ,

"ν; x Κ΄“
ἊΣ aS : “ok an ee ey be eRe bad
ae J δ τ χε «ἴῃ ἢ ἈΚ
Ϊ δεν Ὶ yo? nx τ ρησδύσνν dk ἣ ez
γί Cer Ό mcdne-
f ποκκὸ αὐ Boge Wen c i
|
πος ; porrouohic CoN Foruons Commi 4) 3
αγασαι ὀϊσεπαχοιπα, “Ἢ
τοὶ δε- “pee ε

_ «ἀρποΐρατιβο onabac

' A A x Ἴ 49 og dia 4
: he 4
Lew RR (ee contue Gpacuscbsadonaw a
be ee oy may oo ee nop pe 3
En , camemn Tinonten | fica qrommes cD |
᾿ τος > : R bless? a ee Saccwan i . i
| ΟἹ ae πο, ἀν ee ιν 0 tn Po sa ee 5 ;
feo firvoucem ua _ytenneam) ἡ ἀν |
Gxdisaputr se Goon ain fannexi πο 1)
Γ oo ne es ae ake estame Rape : Fue fem τοῦ 3 . a
oe SUL mn penucnssa- ᾿ ee a
: por ue + Re Sue oe το tos ct”
om alocum = el cage eas
Ci - ¢ boy 5:
ὌΤΙΧΙΟΥ SsuOoSs =. 4
Ὄ ἀρ Sot S31 ne in Ee f Serr Ce, τ“. τ :
οτιοααισοτθ - Inueurecosdonmeny,
¥ pc Si ls ht pe BARD GAT νὰν γα
THceMoMOnEM phaxe ontario, :
ars κ γνώ ἢ per τ deste asd lep Stee > |
BOY pseGuulsusett- | Gores scuidon
Suites fag oe One & ΩΣ Τῇ Se) ngefaahe ᾿
- Quoarcam 10LCKUS eta nes pitieciticete(e) scam
[ran ey “p72 ate gn een : ΑΡΥΝ ine Ep rece e. Es 4 |
Laprors Qoosius- Sn coe Cirhuceoto ποτὰ.
θη iced ἈΝ ᾿ ς hens ; mine Ὁ Reap Mare 59 Cy
τὸ: onabacdicer 2° eccetuRoa + pee
Feber SE bagel ὁ" 1 4 pede φοσδις σὲ
Βα is Cogs fer ἴασαι uocabaxun - -
ΤΩΣ Ἶ ᾧ ἐνὶ Wig Aas Ee mes : ay
ee Licem cem inftamn ame: sadas unusclepuo
γο REO γι Δ. τ we PFs κόρα
uenummunetommed lec qarce- ΠΣ
poll my “Pern dey ἢ Vue
uolmcas secure pus eos
᾿
ie ,
ὃ ᾿ ;

THE LINDISFARNE GOSPELS.

Circ, A.D; 700:

(Scale 2:5. The page shown contains Luke xxii. 37-47. It includes verses 43 and 44,
which are omitted in NABRT and several versions. The small interlinear writing
is an Anglo-Saxon gloss,—the earliest form of the Gospels in English. )

To face 2. 234.

σὰν

V THE ANCIENT VERSIONS 235

(and beautifully printed) study of it," adduces several
reasons for assigning to it an English origin. The text
is a mixture of Northumbrian and Irish strains. Mr.
Hoskier’s introduction, which is very full and detailed,
leads him to offer many suggestions with regard to the
textual history of the Latin Bible in general, which
cannot be discussed here.

For the Acts, Epistles, and Apocalypse eleven MSS.
out of the preceding list are available, viz. ACDFGH
@KTVW, and twelve more have been selected by
Bishop Wordsworth and Mr. White. The additional
MSS. (which may be increased in number when the
Epistles come to be dealt with) are the following :-—

B,. Codex Bambergensis, originally written in the
monastery of St. Martin at Tours, now in the Royal
Library at Bamberg; ninth century. A beautiful copy
of the entire Bible, wanting the Apocalypse, with
elaborate ornamentation of the Carolingian type. The
text is Alcuinian. Collated by White.

I. Codex Juueniani Vallicellianus, in the Biblioteca
Vittorio-Emanuele at Rome; eighth or ninth century.
Contains the Acts, Catholic Epistles, and Apocalypse,
with a text akin to that of A.

L,. Codex Lombardicus, in the Bibliotheque Nationale
at Paris; eighth century. Contains the Pauline Epistles,
written in Lombardic characters. The text is said to
be valuable, but apparently it has not yet been fully
collated.

L,. Codex Lemovicensis, also in the Bibliotheque
Nationale, formerly in the monastery of St. Martial at
Limoges ; eighth-ninth century. Contains the Catholic
Epistles, in a mixed text akin to those of the southern
MSS. generally.

1 The Golden Latin Gospels (P) in the Library of 7. Pierpont Morgan
. edited by H. C. Hoskier. New York. Privately printed, 1910.

236. CRITICISM OF NEW TESTAMENT οἱ,

M,. Codex Monacensis, formerly at Freising, now at
Munich ; ninth or tenth century. Contains the Acts,
Catholic Epistles, and Apocalypse, in a good but mixed
text. Collated by White.

O,. Codex Oxoniensis II., in the Bodleian Library
at Oxford, known as the Selden Acts; seventh or eighth
century. Contains the Acts, written in an Anglo-Saxon
hand, with an excellent text of the Irish type.
Collated by Wordsworth.

O,. Codex Oxoniensis III., also in the Bodleian ;
ninth century. Contains the Pauline Epistles, in a
text originally Old Latin, but corrected into conformity
with the Irish type of Vulgate text. White notes
that it often agrees with the Latin text of the Codex
Claromontanus.

R,. Codex de Rosas, now at Paris, but written in
eastern Spain in the tenth century. The MS. contains
the whole Bible, in four volumes ; but the New Testa-
ment text is not remarkable except in the Acts, where
it agrees generally with the family BK V. In chapters
ΧΙ. and xii. two versions are given, one in the text and
one in the margin; and Vulgate and Old Latin render-
ings are strangely intermixed in both of them.

R,. Codex Regius, in the Vatican Library at Rome:
seventh century. Contains the Pauline Epistles in a
fairly good text. Used by Corssen in his edition of
Galatians, and collated by Dr. Meyncke for the Oxford
Vulgate.

S,, Codex Sangallensis, written at St. Gall by the
monk Winithar, and still preserved there; eighth century.
Contains Acts and Apocalypse, with a quantity of non-
biblical matter, bound up with another MS., containing
the books of Numbers and Deuteronomy. ‘The text
of Acts is akin to that of F (Codex Fuldensis), but
with admixture of other elements. Collated by White.

V THE ANCIENT VERSIONS 227

Εἶν Codex Ulmensis; written for Hartmut, abbot
of St. Gall in 872—883, the Epistle to the Laodiceans
being added (after Hebrews) in his own hand; sub-
sequently at Ulm, now in the British Museum. Con-
tains Acts, Epistles, and Apocalypse. The text, as
usual in the St. Gall MSS., is of a mixed character,
but with the southern type predominating. Collated
by White. Facsimile in Facszmzles of Biblical MSS.
in the British Museum, P\. XVI.

Z,. Codex Harleianus II.; like Z of the Gospels,
formerly at Paris, now among the Harleian MSS. in
the British Museum; eighth century. Contains the
Epistles (without 3 John and Jude) and Apocalypse
(to xiv. 16), written probably in France, but with
decorations in the Irish style. The text is peculiar.
Westcott regards it as Old Latin, and it certainly has
an admixture of old readings; Berger notes that
Heb. x., xi, especially have a text quite different from
the Vulgate. Collated by White; an edition is now in
preparation by E. S. Buchanan.

From the above descriptions it will be seen that
Bishop Wordsworth and Mr. White have not only been
able to select and examine a large number of early
and important MSS. of the Vulgate (indeed we may
assume that no MS. of first-rate importance, at present
known, has been overlooked by them), but also to
classify them in groups. With regard to the later
books of the New Testament, this work is not yet
complete ; but for the Gospels and Acts their results
have been stated, and are not likely to be disputed.
In the Gospels the first place is given by them to
what may be called the Northumbrian group of MSS.,
whose origin can be traced to the great schools of
Wearmouth and Jarrow, founded about 674 by
Benedict Biscop, and promoted by his successor,

238 CRITICISM OF NEW TESTAMENT cx:

Ceolfrid. The text of this group (as appears from the
history of the Codex Amiatinus and the Lindisfarne
Gospels) goes back to MSS. brought from Italy in the
seventh century. The group includes the manuscripts
AASY, with the first hand of H; while F, though not
Northumbrian, is closely akin to this family, and M,
though its text 1s mixed, belongs more to it than to
any other. U also, which is Northumbrian, would no
doubt have to be reckoned here, if there were more of
it. The whole group, therefore, may be said to consist
of AAFHMSUY ; but A and S are only available for
St. John, and U is hardly available at all. The best
text of all, it can hardly be doubted, is that of A, with
Y in close attendance.

Over against this group may be set another, con-
sisting of manuscripts which are good, but less good
than those which have just been enumerated. It is
headed by Z, and composed primarily of BJPZ, with
not infrequent assistance from the mixed texts of
2PGM, and the Canterbury books OX. These last
have, however, an Irish element in them, and conse-
quently stand half-way between the B-Z group and the
thoroughly Irish MSS., DELOR. Another distinct
local family is found in CT, the two Spanish MSS. ;
while the influence of both the Anglo-Irish and the
Spanish families is to be seen in the two special
editions of Alcuin (KV, with affinities in M) and
Theodulf (® and the second hand of H). Finally W
stands apart from all (though nearer to the group of Z
than to that of A), as the representative of a later
stage in the history of the Latin Bible.

In Acts the classification is rather different. Here
Wordsworth and White give the first place to the
Codex Sangermanensis (G), which they have followed
even oftener than A, mainly, as it appears, on the

v THE ANCIENT VERSIONS 239

ground of its accordance with the best Greek authorities.
This, however, is a somewhat doubtful argument, since
it is admitted that Jerome had access to Greek MSS.
of a type different from those known to us, and he
may have estimated their evidence differently. Next
to G they place C, the Codex Cavensis, then A and F,
representing the two main families of our MSS., then
D, which represents the Irish type. The agreement of
GCAFD, as representing several different strains of
tradition, constitutes the highest possible authority.

The editors’ classification of the evidence for the
Epistles and Apocalypse has not yet been made
public.

The original text of the Vulgate being thus more
perfectly restored than ever before, it becomes possible
to estimate Jerome’s work more adequately. An ex-
amination of the corrections introduced by him into
the Old Latin version’ shows that while he must
sometimes have used Greek MSS. unlike any which
we now possess, his Greek authorities on the whole
were of the type represented by our xBL and their
associates ; δὴ standing out above all as the most con-
stant supporter of his readings. In spite of this, it must
be remembered that the Vulgate cannot as a whole
be reckoned as a witness on this side. Jerome’s
revision of the New Testament was very partial, and
the basis of his text remains Old Latin which belongs
to a somewhat different type and family. Still less
is it possible to use the current Clementine Vulgate
as a witness of this early class; for that is a Vulgate
corrupted by much use, and by long centuries of
alternate neglect and revision, which have approxi-
mated it to the late Greek MSS. from which our
Textus Receptus is derived. It is the great service

1 Wordsworth and White, pp. 655-672.

240° CRITICISM OF: NEW TESTAMENT. δ,

of Bishop Wordsworth and Mr. White to have given
us back the Vulgate, so far as the Gospels and Acts
are concerned, much as it left the hands of Jerome,
and to have enabled us to estimate alike the materials
with which he worked, and the deterioration which
his work underwent in the course of the Middle
Ages.

9. The Gothic Version.—One more version remains
to be briefly noticed, namely, that which was made in
the fourth century for the Goths who were then settled
in Moesia. Its author was Ulfilas, a Cappadocian by
descent, who became bishop of the Goths in 348, and
died about 380. The Gothic language being wholly
unliterary up to that time, he had to devise an
alphabet for its expression. The translation was made
from the Greek, both in the New Testament and in
the Old; and the Greek is followed with great fidelity.
The type ‘of text represented in’ it is; for the ‘mest
part that which is found in the majority of Greek
MSS. (the a-family), but it also contains readings of
the β and 6-types. It has perhaps also been modified
by the influence of the Latin versions, as was not
unlikely to happen when the Goths occupied Italy.
For textual purposes, therefore, its evidence must be
used with care. iM

The version only exists in fragments. The most
important MS. of it is a beautiful volume preserved in
the University Library at Upsala, known as the Codex
Argenteus. It contains rather more than half the
Gospels (in the Western order, Matthew, John, Luke,
Mark), written in silver letters upon purple vellum.
The writing is uncial, of the fifth or sixth century.
It is supposed to have been written in Northern Italy,
and in the sixteenth century it was in Germany.
In 1648 it was secured by the Swedes at the capture

ν THE, ANCIENT; VERSIONS 241

of Prague, and after having been presented to Isaac
Voss, was bought back in 1662. It was edited in 1665
by Patrick Young, and in 1854 by A. Uppstr6m.

In addition to this beautiful manuscript, the Gothic
version is also represented by a small Latino-Gothic
fragment from the Epistle to the Romans (sixth
century) in the palimpsest MS. at Wolfenbiittel, which
also contains the Greek MSS. P and Q; and in five
palimpsest fragments (from two MSS. of the sixth
century) at Milan, containing considerable portions of
the Pauline Epistles, and a few verses of the Gospels,
All these fragments appear to have come originally
from Bobbio. The Wolfenbiittel fragment was pub-
lished by F. A. Knittel in 1762, those at Milan by
Mai (the discoverer) and Castiglione in 1819-39.
Fragments of two leaves of a Latino-Gothic MS., found
at Antinoe in Egypt, and containing a few verses of
Luke xxiii, xxiv., are at Giessen, and have been edited
by De. \PivGlaue: and Dr. .K:, Helm;(Giessen, \1910));
cf. /.7.S. xi.612. Some quotations from the Gospels,
occurring in a commentary, were published from a
Vatican MS. by J. F. Massmann in 1834, and a few
leaves of the Pauline Epistles from a Turin MS. by
the same editor in 1868. The principal editions of
the version are those of Gabelentz and Loebe (1836—
1843), of G. H. Balg, published in America in 1891,
and of W. Streitberg, Heidelberg, 1909.

CEVAP TER Wa

PATRISTIC QUOTATIONS

[Authorities : Gregory, off. citt.; Harnack, Geschichte der altchristlichen
Litteratur bis Eusebius (Part i. 1893, Part ii. vol. i. 1897); Smith and
Wace, Dictionary of Christian Biography; The New Testament in the
Apostolic Fathers (Oxford, 1905).

Two of the three classes into which the authorities for
the sacred text may be divided have now been passed
in review. The third remains, namely, the quotations
from it which are found in the works of early ecclesi-
astical writers. That these may provide useful evidence
is obvious. If we know how Clement of Alexandria,
or Origen, or Athanasius, or Jerome, quoted certain
passages of the Scriptures in their writings, we know
(subject to limitations which will be mentioned below)
how those passages stood in manuscripts of the second,
third, or fourth century—that is, in manuscripts as
early as, or earlier than, the most ancient which we now
possess. The limits and value of this class of evidence
must consequently be investigated. In doing so, the
term “patristic quotations,” which is commonly used
to describe it, must be given the widest possible inter-
pretation. It is not only the writings of those who
are specially regarded as the Fathers of the Church
that are useful for this purpose. The writings of any
author who quotes the Scriptures at all must be
242

CH. VI PATRISTIC QUOTATIONS 243

taken into consideration. So far as we possess them,
the quotations of heretics or of non-Christians are
evidence, as truly as those of orthodox Christians. All
we have to ask is, Do the quotations which we find
in any given author enable us to know how these
passages stood in the manuscripts of the Bible current
in his day?

The question is not so easy to answer as it perhaps
appears. Before we can accept the Scriptural quota-
tions which stand in our texts of the early ecclesiastical
writers as faithfully representing the manuscripts which
these writers used, there are several deductions to be
made. In the first place, the true text of the writer in
question has to be ascertained, just as the text of the
Bible or of the classical authors has to be ascertained,
by the comparison of authorities. The texts of the
Fathers, as they have generally been read _ until
recently in the editions of the Benedictines or Migne’s
Patrologia, were based (like the received text of the
New Testament itself) upon comparatively few and
late manuscripts. It has been the work of modern
scholars to lay the foundation for this department
of textual criticism by producing editions of the princi-
pal ecclesiastical writers, accompanied by a sufficient
apparatus critezcus. The work is by no means complete
yet, but much has been done. The Imperial Academy
of Vienna has made considerable progress with a Corpus
of Latin ecclesiastical writers; and the Academy of
Berlin has embarked on a similar undertaking with
respect to the Greek Fathers. When these two great
enterprises are completed, it will be possible to handle
the raw material of patristic quotations with far more
confidence than hitherto.’

1 A gigantic work was undertaken by the late Dean Burgon, with a
view to making the evidence of patristic quotations more accessible ;
namely, an index of all Biblical quotations in the principal ecclesiastical

244 CRITICISM OF NEW TESTAMENT cu.

Secondly, even when the earliest manuscripts of an
author have been consulted, we cannot always be sure
that we have his Scriptural quotations in their original
form. In no part of his text is corruption so likely to
creep in as here. A scribe who recognised a quotation
from its first words would be only too likely to write it
down from memory, without looking too closely at the
MS. before him, and so would give it in the form in
which it was current in his own day, instead of in that
which his author actually used. Or, supposing he
noticed that the form of the quotation was unfamiliar,
he might very probably alter it into what he believed
to be the true form. In either case, it will be seen,
this class of corruption consists in the substitution of
the familiar Textus Receptus in the place of an earlier
type of text; therefore, without any prejudice against
the received text, it must be recognised that, where
two alternatives are open, the one which diverges from
the received text is more likely to be the one originally
used by the Father in question. Voluntarily or in-
voluntarily, the scribe is more likely to alter from an
unfamiliar form to a familiar form than wéce versa.
The only cases in which we can ascertain without
doubt the form in which an ecclesiastical writer made a
quotation are when the context points decisively to
one reading or another; when this is not the case, we
are reduced to a balance of probabilities.

But, thirdly, even when we can ascertain, beyond
reasonable doubt, the form in which a quotation was
made, we may still doubt whether this form was
actually derived from a manuscript lying before the

writers, which is now preserved in manuscript in the British Museum in
sixteen huge volumes (Add. MSS. 33421-33436). The references are to
comparatively uncritical texts of the Fathers (generally those in Migne),
but they could of course be used also in connexion with later editions,
where such exist. Considerable use was made of this work in the edition
of the Gospels commenced by E. Miller (see below, p. 308).

vi PATRISTIC QUOTATIONS 245

author. Authors at all times and in all countries
have been apt to quote from memory, and memory
plays strange tricks. Dr. Salmon adduces from E.
Abbot (Authorship of the Fourth Gospel, p. 39) a
remarkable instance of this in no less a person than
Jeremy Taylor, who quotes the text “Except a man
be born again he cannot see the kingdom of God” nine
times, yet only twice in the same form, and never once
correctly. How often, too, does one see the misquota-
tions “ Whatsoever thy hand findeth to do, do it with
all thy might,” or “to give a reason for the fazth that
is in you”? And if this is the case nowadays, when
books are plentiful and verification of references easy,
how must it have been in the earliest centuries of our
era, when a copy of the Bible would not always be at
a writer’s elbow, and when (in the absence of divisions
into chapters and verses, and even of separation of
words) it can have been by no means easy to turn up
a given passage quickly? Quotations (especially short
ones) must often have been made from memory, and
sometimes without any intention of giving more than
the general sense. An especial source of error was
present in the case of quotations from the Synoptic
Gospels. An author might very easily amalgamate
parallel passages in two or more Gospels, and so pro-
duce his quotation in a form which correctly represented
none of them. Therefore, just as the second class of
considerations tells against indiscriminate acceptance
of quotations which agree with the Textus Receptus,
so this third class tells against quotations which differ
from it.

Under these circumstances, it may fairly be asked
whether it is worth while paying any attention to
patristic quotations at all. And the answer would be
that their testimony is strictly limited in scope, but

246 CRITICISM OF: NEW (TESTAMENT δε

that within these limits it is not only real but of very
great importance. Where a patristic quotation stands
alone as evidence for a particular reading, its testimony
must be regarded with the gravest suspicion, unless the
context renders it quite explicit; but where there is
other evidence for the reading, the quotation is of great
value in fixing the time and the place at which this
reading was current. Thus five cursive MSS., from
the tenth to the fourteenth century, have the remark-
able reading Ἰησοῦν τὸν Βαραββᾶν in Mt. xxvii. 16;
but we could know nothing about the age or locality
of this reading were it not mentioned by Origen, which
proves that it existed in MSS. accessible in either
Egypt or Palestine in the first half of the third century.
As will be seen later, the localisation of certain types of
text in different periods and districts is a matter of the
greatest importance when we come to try to reconstruct
the history of the Biblical text, and to see which type
of text has the strongest claims on our acceptance ;
and here the evidence of patristic quotations is indis-
pensable. It will be found that the ascertainment of
their precise nature is quite worth the trouble which
it involves; and a short indication of the principal
sources of this evidence will not be out of place here.
Quotations from the New Testament are found in
the earliest writers of the sub-apostolic age, but they
are so scanty as to be of little service for our present
purpose. Their importance lies in another direction,
namely, as evidence of the existence of the New
Testament books at a period earlier than that which
some critics have sought to assign to them. The
Epistle of Clement to the Corinthians (written about
A.D. 93--95 ἢ) contains two passages which appear to

1 The dates assigned in this chapter (so far as the first two centuries
after Christ are concerned) are generally those given by Harnack
(Geschichte der altchristlichen Litteratur bis Lusebius, Part 11. vol i. 1897).

v1 PATRISTIC QUOTATIONS 247

be quotations from memory of Luke vi. 36-38 and
of Luke xvii. 1, 2, combined with Matthew xxvi. 24 ;
but they are too inexact to be used for textual pur-
poses. The so-called Second Epistle of Clement (a
homily written about the middle of the second century)
contains several quotations from Matthew and Luke,
as well as passages which must have come from a lost
Gospel; and though these are useful as far as they go,
the amount of evidence which they contribute is not
large. The Epistle of Barnabas (written early in the
second century) contains what is apparently a direct
citation of Matthew xxii. 14, and several reminiscences
of New Testament language; but again precision in
quotation is wanting. The Shepherd of Hermas
(about A.D. 140, but embodying writings of rather
earlier date) contributes nothing to our purpose. The
Epistles of Ignatius (about A.D. 110-117) contribute
only a few quotations from the Pauline Epistles, the
references to the Gospels being in no case verbally
exact ; and much the same may be said of the Epistle
of Polycarp (same date). The Teaching of the
Twelve Apostles offers more plentiful material, includ-
ing a considerable extract from the Sermon on the
Mount, but its evidence is discounted by the doubt
attaching to its date; for while the prevalent view
assigns it to the beginning of the second century, some
critics have put it as early as A.D. 70, and one com-
petent editor (Dr. Bigg) carries it down to the fourth
century.’

A fuller stream of tradition is reached with Justin
Martyr, in the middle of the second century, from
whose writings a very complete outline of the Gospel
narrative can be put together; but it is rather with the
facts than the words of the Gospels that he is con-

1 See also Dean Armitage Robinson, /. 7..S. xiii, 339.

248 CRITICISM OF (NEW TESTAMENT cu

cerned, so that the total amount of properly textual
material to be derived from him is not great. It is,
however, enough to enable us to come to some con-
clusions as to the type of text used by him, and it is
noteworthy that the readings given by Justin agree
predominantly with those which are characteristic of
the Old Syriac and Old Latin versions ; that is, they
belong to the 6-type of text. This is a fact which will
be useful presently. Probably somewhat later than
Justin are the Clementine Homilies, an Ebionite (or
Jewish - Christian) work which falsely claimed the
authorship of Clement of Rome; but they do not
provide so much material, and for textual purposes fall
into the same category as Ignatius or Barnabas. More
important are the writings of Justin’s pupil Tatian, not
so much, however, his Apology for Christianity as his
Diatessaron, which has been described and discussed in
the preceding chapter. Akin to Tatian, as providing
a continuous Gospel text rather than isolated quota-
tions, is the work of the heretic Mareion, who produced
an edition of St. Luke’s Gospel and the Pauline
Epistles (about the middle of the second century) in
accordance with his own peculiar doctrines. Marcion’s
writings, as separate literary entities, are lost; but
much of them can be restored from the references and
quotations in the works of the orthodox controversialists
who opposed him, notably Tertullian and Epiphanius.
The contents of Marcion’s Gospel are in this way
known with practical certainty, and in many cases the
actual text adopted by him. This as a rule resembles
that of Tatian and Justin in belonging to the same
group as the Old Syriac and Old Latin versions ; but
as Marcion’s original home was in Pontus, while his
principal work was done in Rome, it is difficult to say
from what locality he derived his Gospel text. Justin’s

vI PATRISTIC QUOTATIONS 249

evidence is equally hard to locate, since by birth he
was a Samaritan, but after his conversion resided first
in Ephesus, whence he proceeded to Rome, where he
ultimately suffered martyrdom. Tatian’s activity, as
we have seen, lay in Syriac-speaking regions, and his
Diatessaron is believed to have been composed in that
language.

So far, then, the results arrived at may be thus
summarised. Up to A.D. 150 the quotations in extant
ecclesiastical writers, though important in their bearing
on the questions of the date and acceptation of the
New Testament Scriptures, are of little value for purely
textual purposes. From 150 to 175 textual evidence
of value is forthcoming, but either in a somewhat
fragmentary condition, or (in the case of Tatian and
Marcion) in a form which can hardly be described as
quotation, At the same time the evidence, so far as
it goes, tends to show that the 6-type of text was
prevalent in this early period. With the last quarter
of the second century the position changes; and
henceforth we have a succession of ecclesiastical writers
whose works are preserved with substantial complete-
ness, and who quote the New Testament Scriptures,
not tacitly or in paraphrase, but fully and explicitly.
It is from this point that the evidence of patristic
quotations in the full sense may be said to begin; and
the father of it is Irenaeus. It is impossible within
the limits of this chapter to examine the evidence, even
of the more important Fathers, at any length; but a
brief indication of their date and country, with references
to the more important modern works dealing with this
aspect of their writings, will serve to show the general
bearing of their testimony.

Irenaeus was probably born about A.D. 135-140
(according to some, as early as 115), and the home of

250 CRITICISM OF NEW TESTAMENT - cu.

his youth was in Asia Minor, where as a boy he saw
and heard Polycarp. From Asia he is believed to have
gone to Rome, and thence (probably after only a brief
stay) to Lyons, to join the Church already founded
there by missionaries from Asia Minor. After the
great persecution in Gaul in 177 he was commissioned
to carry the letters of the Church of Lyons to Rome, .
and shortly afterwards he was made Bishop of Lyons
in succession to his martyred countryman, Pothinus.
This office he held until his death about 202. Irenaeus’
early training consequently belongs to the East, his
literary activity to the West; and these facts are of
importance in estimating his testimony on textual
subjects. His principal work, the "EXeyyos καὶ ἀνατροπὴ
τῆς ψευδωνύμου γνώσεως, Was written between 181 and
189; only fragments of it exist in the original Greek,
but the whole is preserved in a Latin translation, which
is believed to be nearly contemporary with Irenaeus
himself." It abounds with citations from the New
Testament ; a full examination of them has been taken
in hand by Professor Sanday, with the assistance of
Mr. C. H. Turner, but this long-announced book has
not yet appeared. Until the results of such an ex-
amination are published, it must be sufficient to say
that Irenaeus (as might be expected from his personal
history) is to be reckoned among the authorities that
fall into the same group as the Old Syriac and Old
Latin versions.

A different quarter of the Roman Empire is repre-
sented by Clement of Alexandria, a contemporary of
Irenaeus, though somewhat junior to him. The dates
of his birth and death are unknown, but the former

1 Hort placed it as late as the fourth century, but his view has not been
generally adopted. The issue depends mainly on the question whether the
Latin version of Irenaeus was or was not used by Tertullian in his treatise
adv. Valentinianos.

vI PATRISTIC QUOTATIONS 2.51

probably occurred between 150 and 160, the latter
between 212 and 220. Of Greek nationality (his
Roman name, Titus Flavius Clemens, indicates that
he was a freedman of a Roman family), and born
either in Athens or in Alexandria, he studied philosophy
in Greece, Italy, and the East, and ultimately was led
from Stoicism into Christianity. He became a pres-
byter in the Alexandrian Church, and about 190
succeeded his last and best teacher, Pantaenus, as
head of the Catechetical School in that city, but was
compelled to retire thence by the persecution in 202—
203. Of the remainder of his life little is known.
Of his works a considerable portion has been pre-
served, notably his Στρωματεῖς, or Miscellanies, in
eight books, written in the last years of the second
century. He had a wide acquaintance with Greek
literature, and quotes from innumerable authors, pagan,
Jewish, and Christian; hence he must have had the
use of a large library, and must have been accustomed
to look up quotations in it. For the New Testament,
his quotations must be taken as representing a class of
text which was at any rate current, and perhaps pre-
dominant, in the great literary capital of Egypt at the
end of the second century. An examination of them
in the Gospels has been made by Mr. P. M. Barnard
of Christ’s College, Cambridge, which shows that the
text used by Clement was of the 6-type, akin to the
Old Latin and Old Syriac.’ This is an important
piece of evidence for the early history of the New
Testament text, which will have to be referred to again
in a subsequent chapter.

Important as Clement is, he is followed in his own
country by a scholar and divine of far greater import-
ance. In textual scholarship, indeed, Origen has no

1 Texts and Studies, v. § (Cambridge, 1899). See below, p. 339.

252) CRITICISM: OF, NEW) TESTAMENT cx:

rival among ancient writers, and no single individual
has exercised so wide an influence upon the Biblical
text as he. It is with regard to the Greek text of the
Old Testament that the precise character of his work
is most fully known; but there can be little doubt
that his critical labours on the New Testament were
almost equally epoch-making. Born about 185, he
was educated from the first as a Christian, and trained
by his father in the study of the Scriptures. His
father perished in the persecution of 202-- 203, which
also created a vacancy in the headship of the Cate-
chetical School, through the retirement of Clement ;
and to this Origen, though barely eighteen, was
appointed. Here he laboured, learning and teaching,
for many years. About 213 he visited Rome; about
215 he was compelled to leave Egypt, and lived for
four years at Caesarea in Palestine. He returned to
Alexandria in 219; but the following years were full
of difficulties with his ecclesiastical superiors, and in
231 he left Alexandria finally. For the remainder
of his life, which lasted till 253, his home was at
Caesarea; and here the greater part of his literary
work was done. With his philosophical and exegetical
writings we are not here concerned. His textual
labours on the Old Testament were embodied in the
Hexapla, a colossal undertaking which coloured the
whole subsequent history of the Septuagint. For the
New Testament we know of no such formal edition of
the whole text; but he wrote commentaries on most,
if not all, of the books of which it is composed, and
numerous passages in his writings show that he had
examined and compared manuscripts, and considered
the weight of the evidence for various readings. He is,
indeed, the first textual critic of the New Testament,
and when we consider the age of the manuscripts he

νι PATRISTIC QUOTATIONS 253

must have used, the value of his evidence is obvious.
In many cases he mentions various readings, and states
which is found “in most MSS.” or “in the oldest
MSS.” or “in the best MSS.” No doubt we are
not always bound to accept his opinion as to what
were the best MSS., but at least we learn what
readings were extant in Egypt and Palestine in
his lifetime, and what readings were preferred by a
trained scholar and textual student of unusual ability.
The character of the New Testament text used
by him differs in different works. In some it is
of the 6-type, but as a rule his preference is for
the @-type, of which he is the most eminent ally
among the Greek Fathers. Whether his connexion
with the 8-type of text goes even further than this, so
that he may be regarded as largely responsible for its
preservation, or even for its existence in its present
form, is a point which will have to be considered more
at length in the final chapter. It should be added
that Origen’s services to criticism did not end with his
life ; for copies of his writings formed the nucleus of
the library collected by Pamphilus at Caesarea, which
was thenceforth a recognised centre of textual research.’
Many of Origen’s works are now lost; but some are
still extant in the original Greek, and many more
in Latin translations (largely the work of Rufinus),
Critical texts of nearly all the Greek remains have

1 See above, pp. 65, 106, for records of MSS. copied from, or collated
with, MSS. in the library of Caesarea. The description which Jerome
gives of the founding of the library of Pamphilus is as follows (Z/. cxli.
[xxxiv. in Migne], written A.D. 384): ‘‘ Beatus Pamphilus martyr, cuius
vitam Eusebius Caesariensis episcopus tribus ferme voluminibus explicavit,
cum Demetrium Phalereum et Pisistratum in sacrae Bibliothecae studio
vellet aequare, imaginesque ingeniorum, quae vera sunt et aeterna monu-
menta, toto orbe perquireret, tunc vel maxime Origenis libros impensius
prosecutus, Caesariensi ecclesiae dedicavit; quam ex parte corruptam
Acacius [successor of Eusebius in the see of Caesarea about 340] dehinc
et Euzoius, eiusdem ecclesiae sacerdotes, in membranis instaurare conati
sunt,

254 CRITICISM OF NEW TESTAMENT ὦ

been produced in modern times, as of the Hexapla by
Field, the Philocalia (a volume of selections from
Origen, prepared by Basil and Gregory Nazianzen) by
Armitage Robinson, and the remains of the Commen-
tary on St. John by A. E. Brooke; while all, except
the Commentary on St. Matthew, have already made
their appearance in the Berlin Corpus of Greek ecclesi-
astical writers.

We have a little departed from strict chronological
order, in taking Origen in connexion with his fellow-
Egyptian Clement, and must return to consider the
great African writer whose life overlaps those of both
of them, Tertullian. Born about 150, and trained
probably as an advocate, he became after his conversion
the great controversialist of the early Latin Church.
He was probably a presbyter, in spite of the fact that
he was undoubtedly married. About 203 he seceded
from the orthodox church, joining the ascetic and
enthusiastic sect of the Montanists. He died at some
uncertain date after 220. His writings are vehement,
rhetorical, unrestrained in style and language; and
though several are lost, many still remain. He quotes
freely from the Scriptures; but for textual purposes
his evidence is impaired by the doubt whether he is
(1) quoting from memory, (2) translating direct from
the Greek for himself, or (3) using an early Latin
version of the New Testament. Not being a scholar,
such as Clement and Origen were, he has not the
scholar’s accuracy ; and his evidence must consequently
be used with caution. An examination of it has been
made by H. Ronsch (Das Neue Testament Tertullians,
1871), while the new edition of his works by Reiffer-
scheid and Wissowa in the Vienna Corpus Scriptorum
Ecclestasticorum Latinorum sets out the available textual
material for the future use of students.

v1 PATRISTIC QUOTATIONS 255

For the outlying portions of the Roman Empire—
for Asia Minor, Egypt, North Africa, and Southern
Gaul—we have, as has now been shown, excellent
witnesses for the period 175-250; but for Rome itself
we are very scantily supplied with material. The
principal author whose works remain to us is Hip-
polytus, a writer almost as prolific as Origen, and
unquestionably the most prominent theologian in the
Roman Church during this period. The details of
his life are obscure. So much is known, that he
flourished about 220, that he was a presbyter of the
Church of Rome, and that he appears to claim for
himself the position of a bishop; according to some
he was bishop of Portus, according to others a schis-
matic bishop of Rome, in opposition to Pope Callistus ;
while others have doubted whether he is correctly
described as bishop at all.’ In any case, he was the
last and greatest Greek-writing divine of the Roman
Church ; and his writings included commentaries upon
many books of the Old Testament, and on St.
Matthew and the Gospel and Apocalypse of St. John
in the New. By far the greater part of his voluminous
works has perished. Perhaps the most interesting of
all (his Refutateon of all the Heresies) has been in large
measure recovered from a late MS., brought to Paris
from Mt. Athos in 1842, but not published or identified
till 1851; but his commentaries exist only in frag-
ments. All that remains is in course of publication
in the Berlin Corpus. His quotations from the New
Testament, which are fairly numerous, have not been
specially investigated ; but they are often too inexact
to be useful for textual purposes.

The only important writer of the third century who
remains to be discussed is Cyprian, the great bishop of

1 See Dr. Salmon’s article in Dect. of Christian Biography.

256 CRITICISM OF NEW TESTAMENT . cu.

Carthage, who must be coupled with Tertullian as the
main source of our knowledge of the early African
Church. Born about 200 in a pagan family, and
trained in law and rhetoric, he was converted to
Christianity about 245, and was almost immediately
elected, much against his will, Bishop of Carthage.
This position he held from about 248 until his martyr-
dom in 258, during Valerian’s persecution. Cyprian’s
importance lies more in his greatness as a bishop and
an administrator than in his literary ability, and his
writings deal with matters of Church discipline and
morals rather than with textual or exegetical questions ;
but his quotations from Scripture are plentiful, and
since he had more of the scholar’s instincts than Ter-
tullian, he is the most important patristic witness to
the text of the Latin Bible in use in Northern Africa
during the third century. For the conclusions to which
a study of his Biblical quotations leads, see above,
p. 203. His writings have been edited by Hartel in
the Vienna Corpus (1868-71); but in the work of
most importance for textual purposes (the Zestimonza,
a classified selection of Biblical passages) Hartel un-
fortunately followed the least trustworthy of the extant
MSS."

Of the other ecclesiastical writers of the third
century, brief mention need only be made of a few.
Gregory Thaumaturgus, Bishop of Neocaesarea in
Cappadocia from about 240 to his death in 265, wrote
a panegyric on Origen, a paraphrase of Ecclesiastes,
and some dogmatic treatises. Contemporary with him
is Dionysius Alexandrinus, a pupil of Origen, and like

1 See Sanday in Old Latin Biblical Texts, ii. p. xliii. A new edition of
the Zestimonda, and also of the Ad Fortunatum, is contemplated by Pro-
fessor Sanday and Mr. C. H. Turner. Meanwhile the subject has been
fully dealt with by Hans von Soden, Das Jateintsche Neue Testament in
Afrika zur Zeit Cyprians (Leipzig, 1909).

vr PATRISTIC QUOTATIONS 257

him head of the Catechetical School in Alexandria
(from 233), and subsequently bishop of that see (247—
265). Unfortunately of all his many writings only
fragments and a few letters now remain; for most of
these we are indebted to Eusebius. Methodius, Bishop
of Olympus in Lycia or of Patara (or of both in suc-
cession) towards the end of the third century, martyred
under Diocletian in 311, wrote many treatises (including
polemical criticisms of Origen’s theology), some of which
survive in a Slavonic version ; in Greek nothing remains
but a single dialogue and some fragments. Finally,
Pamphilus of Caesarea (ob. 309) deserves mention less
on account of his literary labours (a defence of Origen,
and an edition of the Septuagint extracted from
Origen’s Hexapla, which he prepared with the assist-
ance of Eusebius) than for the theological library which
he established in Caesarea, and which, as we have
already seen, played an important part in the textual
history of the New (as well as of the Old) Testament.
Passing to the fourth century, the sphere of patristic
quotations widens. As the Church became recognised
by the State, the number of ecclesiastical writers natur-
ally increased, and their works were less exposed to
destruction. Only the more prominent among them
can be mentioned here. Few of them have been
critically edited in modern times; but as the Bible
text in current use approximates more to the Textus
Receptus, it is less liable to corruption. Eusebius of
Caesarea, the great historian of the early Church,
carries on the sequence with those whom we have
named above, since he was the friend and colleague of
Pamphilus. His life extends from about 270 to about
340, and for the last twenty-seven years he was Bishop
of Caesarea, With Pamphilus he prepared for separate
publication Origen’s text of the Septuagint; and in
9

2:8. CRITICISM (OF NEW 'THSTAMENT ὍΝ

the library of Pamphilus he found most of the material
for his works. These (so far as they are still extant)
fill six volumes in Migne’s Patrologia, the most im-
portant being the Chronicle (though this contains
nothing to our present purpose), the AHzstorza Ecclesz-
astica, Praeparatio Evangelica, and Demonstratio Evan-
gelica. Some commentaries on certain books of the
New Testament are lost. The fact that Eusebius
certainly had a great library at hand, and was
accustomed to the use of it, adds much weight to
his evidence on textual matters. He had, moreover,
a direct influence on the dissemination of the Scriptures
in the Roman Empire, since it was to him that Con-
stantine, after his conversion, applied to furnish fifty
copies of the Bible for use in the churches of his new
capital. See above, p. 83.

Nearly contemporary with Eusebius, and a member
of the neighbouring Syriac Church, was Aphraates,
bishop in the monastery of Mar Matthaeus near ancient
Nineveh about the year 340. Twenty-two homilies
by him are extant in Syriac, most of them also in an
Armenian version, in which their author’s name appears
in the form of “ Jacob of Nisibis.” The importance of
Aphraates’ evidence with regard to the Biblical text
current in the Church of Syria lies in the fact that
he evidently used the Old Syriac version. Mention
has already been made of Ephraem, a later con-
temporary of Aphraates, who died in 378. His
sermons and theological treatises contain many quota-
tions from the Scriptures ; but of more importance are
his commentaries on various books of the New Testa-
ment, and (as has been shown above, p. 148) on
Tatian’s Diatessaron. The most recent examinations
of his quotations go to show that the text used by him
was some form of the Old Syriac (p. 162).

tn

νι PATRISTIC QUOTATIONS 260

The Churches of Asia Minor during the fourth
century produced several notable writers, among whom
may be named Basil (329-379), Bishop of Caesarea in
Cappadocia, author of sermons, epistles, and ecclesias-
tical treatises; his brother, Gregory of Nyssa, who
wrote commentaries and apologetic works, besides
other theological treatises ; and his namesake and con-
temporary, Gregory of Nazianzus, the great preacher,
for a short time bishop of Constantinople. In Pales-
tine and Syria, besides Eusebius, we have Euthalius,
whose work has been described above (p. 82), Cyril of
Jerusalem, bishop of that see from 351 to 386, and
especially Chrysostom. Born at Antioch about 347,
it was in Antioch that he lived and laboured until 398,
when he was removed to become Patriarch of Con-
stantinople, and to enter on that career of struggles
with principalities and powers which only ended with
his death in exile in 407. His voluminous works
(filling thirteen volumes in Migne’s Patrologza) furnish
ample evidence of the New Testament text which he
used; especially his commentaries, from which the
complete text of the book commented on could in
some cases be restored. He marks a distinct epoch
in the history of the New Testament text, since we
find in him the a-type of text already firmly established.
Mention should also be made of another Antiochene
writer, Theodore of Mopsuestia, a presbyter in the
Church of Antioch contemporary with Chrysostom, and
like him removed (about 392) to a see in a different
country, in this case Mopsuestia in Cilicia. He wrote
commentaries on the Pauline Epistles, which survive in
fragments and in a Latin version; but he is of far
greater importance from a doctrinal than from a
textual point of view.

Egypt contributes during the fourth century no such

260 CRITICISM ΘΕ NEW TESTAMENT © cu

important authors (for the purpose of textual criticism)
as Clement and Origen; but in Athanasius, Bishop
of Alexandria from 328 to 373, it has a theologian of
the first rank, many of whose works survive. Slightly
later than him is Macarius Magnus, an Egyptian pres-
byter, and author of theological treatises and homilies.
Finally, to conclude the evidence of the Eastern
Churches, mention must be made of Epiphanius,
Bishop of Salamis in Cyprus from about 368 to 402.
His great work is his treatises against heresies. Pre-
vious to his appointment to the see of Salamis, his
home was in Palestine, so that his evidence with regard
to the Scriptural text is probably to be credited to
that locality.

In the Latin Churches during the same period, the
earliest name that need be mentioned is that of
Lactantius (about 260 to 326), an African by birth,
but resident in Gaul during the latter part of his life.
His Biblical quotations have been examined by Ré6nsch,}
but as all but three of them are from the Old Testa-
ment, they do not concern us here.

Gaul is also represented by Hilary, Bishop of
Poitiers from 354 to 368, and author (among other
works) of a commentary on St. Matthew. His quota-
tions must be taken as evidence of the Old Latin text
current in Gaul about the middle of the fourth century.
Lucifer (ob. 371), bishop of Cagliari in Sardinia, is a
writer of some textual importance, whose works have
been edited (in 1886) from the only extant MS. by
Hartel in the Vienna Corvfus. As has been stated in
the preceding chapter, his quotations represent the
European type of the Old Latin version. Ambrose,
the great bishop of Milan from 374 to 397, wrote
many theological treatises, which likewise witness to

1 Zeitschrift fiir die historische Theologie, 1871, pp. 531 ff.

VI PATRISTICVOUOTATIONS 261

the pre-Vulgate text of the Scriptures. The two
greatest representatives of the Latin Churches in this
century and the beginning of the next are, however,
Jerome and Augustine. The life of Jerome (circ. 345—
420) has been sketched above (p. 216) in connexion
with his great work, the production of the Latin
Vulgate. Like Origen, he was a textual scholar by
profession ; and his travels and his acquaintance with
Greek manuscripts give his evidence special weight as
that of a trained student of textual criticism, while they
deprive him of the character of representative of any
local type of text. His great contemporary Augustine
(354-430), on the other hand, was a theologian rather
than a scholar; but his very numerous works provide
plentiful evidence for textual purposes. The first
twenty-nine years of his life were passed in Africa,
where he received the ordinary training of the best
schools, and entered on the profession of teacher of
rhetoric. In 383 he visited Rome, and shortly after-
wards was appointed professor of rhetoric at Milan,
Here he came under the influence of Ambrose, and
here it was, in 386, that the long period of moral and
intellectual disturbance through which he had _ been
passing culminated in his conversion to Christianity.
Having been baptized by Ambrose at Easter 387, he
returned to Africa, where the rest of his life was spent.
At Hippo in Numidia he was ordained priest in 390,
and in 395 was consecrated bishop of the see, which
he held until the close of his life. His writings, con-
troversial, dogmatic, and devotional, are too many to
enumerate, the Confessions and the Czty of God being
the most famous. The Biblical quotations which occur
plentifully throughout his works still need scientific
examination. In his earlier books he must necessarily
have used the Old Latin Bible, and it has been usual

262 CRITICISM OF ‘NEW ‘TESTAMENT © cu:

to suppose that he continued to do so even after the
publication of the Vulgate. This belief is strengthened
by the fact that he expressed strong disapproval of
Jerome’s revised text; but this disapproval was con-
fined to the thorough-going revision of the Old Testa-
ment, and does not apply to the more conservative
treatment of the New, with which, on the contrary, he
expressly said that he had no fault to find. Hence
there is nothing @ prvzort improbable in the opinion
maintained by Prof. Burkitt,’ that while Augustine
habitually used the Old Latin to the end of his life in
short phrases which he would naturally quote from
memory, in longer citations, at any rate from the
Gospels, he used (after about 400) the revised version
of Jerome. Prof. Burkitt has made this clear in the
case of two treatises (the Contra Felicem and the De
Consensu Evangelistarum—the latter a work of especial
value for textual purposes), and it is probable that a
full examination of his later works would show the
same result.

An African contemporary of Augustine of some im-
portance is Tyconius, whose period of activity belongs
to the latter part of the fourth century. He was a
member of the sect of the Donatists, whose stern views
on the treatment of those who had given up the sacred
books or vessels under persecution led them to hold
aloof from the rest of the Church. Refusing all
association with those who held the more lenient views
adopted by the heads of the Church, they adhered to
the Old Latin version of the Bible long after the Vulgate
had been generally accepted elsewhere. ‘Tyconius was
the author of a commentary on the Apocalypse, which
is known to us only in fragments; and of a Book of

1 The Old Latin and the Itala (Texts and Studies, iv. 3), pp. 57-593
and see p. 213, above.

vI PAT RISTICV QUOTATIONS 263

Rules (44. rules of interpretation of the Old Testament
prophets), which has survived and has been edited with
a most interesting introduction by Prof. Burkitt.’
Prof. Burkitt's conclusion is that the text used by
Tyconius was substantially the same as that of Cyprian,
slightly altered in Latinity, but not revised from the
Greek.

Finally, mention must be made of Priscillian, a
wealthy Spanish layman who adopted a species of
Gnostic-Manichean heresy, which he propagated with
creat ardour from about 380 till his execution by
Maximus in 385. Until recently his writings were
supposed to be lost; but in 1885 a manuscript (prob-
ably of the sixth century) was discovered by G. Schepss
at Wiirzburg, containing eleven treatises from his pen.”
He was also the author of a system of canons of the
Pauline Epistles, somewhat similar to those of Eusebius
for the Gospels, the text of the Epistles being divided
into numbered sections, and the numbers classified
under ninety heads in accordance with their subjects.
This system is found in some of the Spanish MSS. of
the Vulgate.’

Beyond the end of the fourth century it is not
necessary to pursue the subject. By this time a
substantially uniform type of text had been adopted
throughout the Christian world, and the vast majority
of later patristic quotations are of the same type as our
Textus Receptus. Here and there exceptions may be
found, just as they are found in the case of manuscripts,
where a divergent text has survived into a later age.
Of these it must be sufficient to name Primasius,

1 Texts and Studies, iv. 1 (Cambridge, 1895).

2 Edited by their discoverer in the Vienna Cores, vol. xviii. (1889).

3 The total number of sections in the several Epistles is as follows :—
Romans 125, 1 Cor. 105, 2 Cor. 61, Gali'28) Eph. 41, Phil. 25,, Col: 34,
ὙΠ ΠΞΞ 22,2) Thess. 10, 1 Pam. /4ry2him,., 26, Tit. 15; Philemy'5,
Hebrews 28.

264 CRITICISM OF NEW TESTAMENT cu. v1

bishop of Hadrumetum in Africa in the middle of the
sixth century, whose commentary on the Apocalypse
contains an almost complete text of that book in an
Old Latin version of the African type (see above,
p- 210). But cases such as these are rare, and as a
general rule uniformity, tempered only by the errors
of scribes and the occasional efforts of editors, settles
down upon the Greek and Latin texts of the New
Testament alike, to be stereotyped ultimately by the
invention of printing in the form of our Textus
Receptus.

NOTE TO CHAPTER VI

The following are the numbers of quotations from the N.T. given in
Burgon’s index, in the case of a few of the earlier and more important
writers.

Gospels. | Acts. fon Ton. Apoc. Total.
Justin Martyr. . 268 IO 6 43 3 330
(besides 266
doubtful)
Irenaeus. 5.7%.) . 1075} to4 23 499 65 1819
Clement Alex. . | 1017 44 | 207 1127 II 2406
Origen . ~ . . | 9231 | 349 | 399 7778 =| 165 17922
Tertolliane τ τ 1 3622. ΠΟΣῚ 1 120 2609 205 7258
Hippolytus. . . 734 42 27 387 188 1378
ΞΕ ies). ai 3256 emer 88 1592 27 5176

CHART ER: Vit

TEXTUAL CRITICISM: IN THE PAST

[Authorities.—Gregory, off. cztt.; Scrivener-Miller, of. czt.; 5. P. Tregelles,
An Account of the Printed Text of the Greek New Testament (1854) ;
Reuss, Bibliotheca Novi Testamenti Graecit (1872) ; Schaff, Companion
to the Greek Testament (1883); British Museum Cazalogue of Printed
Books, s.v. Bible; Westcott and Hort, of. cit. ; B. Weiss in Texte und
Untersuchungen, vols. vii. viii. ix. xiv., and N.F. iv. (1892-99). ]

THE preceding chapters have described the material
which lies to the hand of the critic who would restore
the true text of the New Testament—the manuscripts,
the versions, and the patristic quotations. It remains
to consider how these materials should be used. And
the nature of this problem will be clearer, if some
account be given of the work which scholars have
already done in this field of criticism. In the present
chapter, therefore, it is proposed to sketch the history
of textual criticism in the past, and thereby to show
the questions and the difficulties which now confront
scholars at the beginning of this twentieth century, the
fuller consideration of which will form the subject of
the next and concluding chapter.

Textual criticism is not an invention of modern
times, but is far older than the books of the New
Testament. The scholars of Alexandria in the third
century before Christ busied themselves with the text

265

266 CRITICISM OF NEW TESTAMENT cu.

of the ancient Greek poets and prose writers, and
established a tradition which made Alexandria the
home of scholarship for many a century after their
day—in fact until the conquest of Egypt by the
Mohammedan Arabs. Therefore it is not strange
that Alexandria should also be the home of the
textual criticism of the Greek Bible. Its father was
Origen, whose work has been briefly described in
the preceding chapter. Nor was his work barren.
His own great Hexapla, with the separate edition
of the Septuagint text prepared by his followers,
Pamphilus and Eusebius, testifies to the results of
his labours on the Old Testament ; while yet another
Greek native of Egypt, Hesychius, produced another
edition of the Septuagint. For the New Testament
we can point to no such definite embodiment of
Origen’s textual researches; but it is becoming con-
tinually more evident that both Alexandria and the
school of Caesarea exercised considerable influence
on the textual history of the New Testament, and it
is difficult not to recognise here the results of Origen’s
teaching and traditions. It was to Caesarea, too, that
the great textual scholar of the West came to study
and to gather materials, and so the torch was handed
on from Alexandria to Rome, and from Rome, in the
form of the Vulgate, its effects spread over all the
Western Church.

It would not be profitable to dwell here on the
textual criticism of the Middle Ages; and indeed the
principal points in it have been mentioned already.
We have described the attempts of Alcuin and Theo-
dulf, who may be regarded as the most prominent
representatives of textual scholarship in the West, to
purify the text of the Vulgate, and their little success ;
while in the East even less was done. It is only with

Se ον ἐς πιῶ, “- Ἴδαν

vi LEX? UAL CO RTRICISM IN THE | PAST: 267

the revival of learning in the West, and with the inven-
tion of printing, that the period of modern criticism
begins ; and it is to this that we must now pass. It
was in 1454 that the first printed document made its
appearance in Europe, in 1456 that the first printed
book (the great Latin Bible known as the Mazarin
Bible) issued from the press of Fust and Gutenberg at
Mainz; but it was not until sixty years later that even
the New Testament was obtainable in Greek.’ The
word “obtainable” is used advisedly ; for though the
first Greek New Testament—that of Erasmus—was
published in 1516, another edition of it had already
been in type for two years, though still held back from
the public. This formed a part of the great Complu-
tensian Polyglott, which should be regarded as the
parent of the textual criticism of the printed Bible.

: The Complutensian Polyglott owed its existence to
Cardinal Francisco Ximenes de Cisneros, Archbishop
of Toledo, who undertook it in 1502, in honour of the
birth of the child who was afterwards the Emperor
Charles V. It was printed at Alcala, from the Latin
name of which town (Complutum) it has received its
title ; and it contained the entire Bible, the Old Testa-
ment in Hebrew, Greek, and Latin, the New in Greek and
Latin. The New Testament was first printed, the volume
containing it being completed on January 10, 1514;
but its publication was delayed until the Old Testa-
ment should be ready to accompany it. This occupied
no less than four volumes, the last of which is dated
July 10, 1517; yet even then a long delay took place
before the work was actually given to the world.

1 Some small extracts were printed earlier. The Magnificat and Bene-
dictus were printed with other Canticles in a Psalter at Venice in 1486 ;
the first six chapters of St. John, also at Venice, in 1504 by Aldus Manu-
tius; and the first fourteen verses of the same Gospel at Tiibingen in

1514.

268 CRITICISM OF NEW TESTAMENT «cu.

Ximenes died in November 1517, and it was not
until March 22, 1520, that Pope Leo X. authorised
the publication ; and it seems not to have been actually
issued until 1522. The principal editor employed
by Ximenes for the New Testament was Lopez de
Stunica; and the preface and dedication state that the
text was derived from MSS. lent by Pope Leo from the
Vatican Library. No more precise identification of the
MSS. is possible ; but although the editors affirm that
their authorities were the oldest and most accurate
obtainable (which implies that they exercised some
selection and critical judgment), there is no trace of
their having used the great Codex Vaticanus (B).
The terms of their preface also make it possible that
they used other MSS. as well as those of the Vatican ;
and since Leo (who lent the Vatican MSS.) only
became Pope less than a year before the completion
of the New Testament volume, it is probable that the
work was begun in the first instance with other authori-
ties, not now identifiable. In a few instances (as in
the insertion of a Greek version of 1 John v. 7, 8) the
Greek text has been adapted to the Latin, which
Stunica deliberately affirmed to be the more pure; but
generally the two texts are distinct. In spite of the
editors’ assertion as to the age of the authorities
employed by them, the Complutensian text appears
to be substantially of the type found in Greek MSS.
of comparatively late date.

It has already been said that the Complutensian
Polyglott, though the first Greek New Testament to
be printed, was not the first published. That honour
belongs to the New Testament of Erasmus, [15 origin
was due to the energy of the printer Froben of Basle,
who, hearing of the work which Ximenes had in hand,
conceived the idea of anticipating it, and applied to

a

wi TEXTUAL (CRITICISM:IN THE \RAST 269

Erasmus, the first scholar of the day, to furnish him
with a Greek New Testament as speedily as possible.
The application was made in April 1515, and so
quickly did editor and printers work that the edition
was ready: bys the tstyor March) 1516... The (Greek
text was accompanied by a Latin translation and
some notes, which Erasmus had had in hand before
Froben’s proposal. Work so rapidly produced could
not rest on any great accumulation of material, and
although the publisher’s preface speaks of the use of
many ancient MSS. and of the quotations of all the
most important Fathers, it would appear that in reality
only a few manuscripts were employed—those, namely,
which lay ready to the editor's hand at Basle.’ These,
for the most part, were neither ancient nor good, and the
single MS. employed for the Apocalypse was deficient
in the last six verses of the book, which Erasmus accord-
ingly supplied by re-translation from the Vulgate.
Some words of this re-translation, which occur in no
MS. whatever, still linger in our Textus Receptus to the
present day. Similar re-translations, to supply real or
supposed lacunas, were also occasionally made in other
places.

The edition of Erasmus consequently has little
critical value, and is inferior in this respect to the Com-
plutensian ; yet it has exercised a far greater influence
on the history of the New Testament text. In the first
place, it had six years’ start of its rival; and being
issued in a single volume of reasonable size and price,
it had a far wider circulation” than the six-volume |
Complutensian, of which only 600 copies were printed.

1 Now known as I, 2, 23P, 42P, τ, Of these 2 (of the fifteenth century)
was principally used for the Gospels, 28Ρ (of the thirteenth or fourteenth
century) for the Acts and Epistles, and 1° (of the twelfth century) alone for
the Apocalypse. 1, a better class MS. of the eleventh century, was only
occasionally followed.

* Erasmus states that 3300 copies were issued of his first two editions,

2700. CRITICISM OF NEW TESTAMENT.” ex,

Hence it formed the foundation of the editions which
followed it at short intervals during the next genera-
tion. Aldus reprinted it at Venice in 1518, in con-
junction with the Septuagint. In 1519 Erasmus
issued a revised edition, correcting many misprints, and
inserting improved readings from Evan. 3; and three
more editions appeared in his lifetime, in 1522,' 1527,
and 1535. Each of these contains some alterations,
that of 1527 being noticeable for its use of the Com-
plutensian edition (mainly in the Apocalypse) and for
its introduction of the Vulgate text by the side of the
Greek and Erasmus’s Latin. This edition of 1527 may
be considered as Erasmus’s definitive text, that of 1535
showing but very few alterations.

Other publishers followed in the footsteps of Ximenes
and Froben in issuing editions of the New Testament
in Greek, but for the most part they contented them-
selves with reproducing the text of Erasmus, and
the next that deserves mention is Robert Estienne, of
Paris, whose name is Latinised by himself as Stephanus,
and quite unnecessarily Anglicised by some as Stephens.
His first edition appeared in 1546, his second in 1549,
both being pretty little volumes in 16mo, printed from
a new fount of small Greek type, with a text com-
pounded from Erasmus, the Complutensian, and fifteen
MSS., mostly at Paris. Only the two latter authorities
are acknowledged in the preface. His third edition, a
folio, published in 1550, was a more elaborate under-
taking, containing a revised text, and giving in the

1 This edition is notable for its introduction of the passage relating to
the Three Ileavenly Witnesses (1 John v. 7, 8). In controversy with
Stunica, Erasmus had promised to insert it if any Greek MS. could be pro-
duced in which it occurred. It was found (in a clumsy form) in a MS. in
England (61, now at Dublin), and Erasmus, though rightly supposing that
it was due merely to re-translation from the Latin, inserted it in fulfilment
of his promise. Hence the passage (for which there is early Za¢zm authority)
found its way into the Textus Receptus.

Ee τς αὐ

vr ΤΕ RIC ISM IN: THE BAST yt

margin various readings from his fifteen MSS. and the
Complutensian. One of his MSS. was the Codex
Bezae ; most of the rest have been identified with
minuscule MSS. in the Paris Library. The text itself
shows greater approximation to that of Erasmus than
its predecessors. It is from this third edition of Steph-
anus that the Textus Receptus found in our ordinary
Greek Testaments is derived, with some slight altera-
tions ; so that its importance in the history of the Bible
text is very great. A fourth edition was produced by
Stephanus in 1551 (a 16mo, printed at Geneva), but it
practically reproduces the text of 1550, with the
addition of the Vulgate and the Latin version of
Erasmus ; its only important feature being the division
of the text for the first time into verses.

The work of Stephanus was carried on by the Protes-
tant scholar, Theodore Beza, who published no less
than nine editions of the New Testament between 1565
and 1604. Five of these are, however, small reproduc-
tions of the larger volumes, and do not represent inde-
pendent recensions. As we have seen above (pp. 89, 97),
Beza was the owner of two very important MSS., viz.
the Codex Bezae (D) of the Gospels and Acts, and the
Codex Claromontanus (D,) of the Pauline Epistles ; and
he also had access to the collations made by his printer,
Henri Estienne, for his father Robert. Nevertheless
Beza’s editions do not embody much work of a textual
kind, differing but slightly from the fourth edition of
Stephanus and from one another. Their importance
lies in the extent to which they, with Beza’s name and
fame to back them, tended to popularise and to stereo-
type the Textus Receptus.

The last stage in this process is represented by the
Elzevir editions, the first of which appeared at Leyden
in 1624. The objects of the Elzevirs were commercial,

272° CRITICISM: OF NEW TESTAMEN LD)” ten:

not critical or literary ; and their editions, though neat
and handy, and consequently popular, have little textual
value in themselves. No editor’s name is attached to
the edition of 1624, nor to any of its successors ; and
its text appears to be substantially that of Beza’s first
edition. A second edition appeared in 1633, and five
others (from either the Leyden or the Amsterdam
branch of the firm) between that date and 1678; the
variations in these later editions are, however, slight.
The popularity of the Elzevir publications led to their
text being widely adopted for common use; and if the
Stephanus of 1550 set up the standard which has been
generally followed in England, the Elzevir of 1624
performed the same service for the Continent.’

Here, then, ends the first stage in the history of the
printed text of the New Testament, with the establish-
ment of a standard or generally accepted text, which
has continued to form the common basis of criticism
from that day until this. It will have been seen, how-
ever, that but little critical value can be attached to it.
The number of MSS. consulted for its production, in
all the century from Erasmus to Elzevir, is very small ;
few of these were of early date, and they were but
slightly used ; in the main, the text rested upon a few
late minuscule MSS. which happened to be accessible
to the editors. It must be plain, therefore, that so far
as human agency is concerned, the received text (which
of course formed the basis of our Authorised Version,
as well as of our current Greek Testaments) has no
commanding claims upon our acceptance, and, indeed,
that it would be contrary to all the ordinary canons of
textual criticism if it did zo¢ need considerable correc-

tion by the use of earlier and better authorities.

1 The phrase ‘‘ Textus Receptus’’ may be traced back to the second
Elzevir edition (1633), the preface to which has the words ‘‘Textum ergo
habes nunc ab omnibus receptum.”

wm) LEXTUAL CRETICISMOUIN THE PAST (273

The second period,'on which we now enter, is that
of the accumulation of evidence for the improvement of
the received text. It covers a space of well-nigh two
centuries, throughout which time, with a few exceptions,
it is the collection of evidence, and not its application,
that occupies the energies of Biblical scholars. The
labours of the sixteenth and the first part of the seven-
teenth century had given Europe its Bible, alike in the
original Hebrew and Greek and in the vernacular
tongues of the Western nations. The practical needs
of the churches and the peoples were adequately
supplied ;' and the preparation of editions of the
sacred text now passed into the hands of scholars, who
sought to improve upon the legacy of their predecessors.
As new manuscripts came to light and were gathered
into the libraries of Europe, scholars extracted from
them the readings which differed from the received
text, examined them, classified them, formed provisional
estimates of the comparative value of the various
authorities, and so, with many imperfections but witha
steadily increasing standard of accuracy and complete-
ness, compiled a vast body of materials for their suc-
cessors, the scholars of the nineteenth century, to use.
It is with the labours of accumulation that we have now
to deal; the critical application of these labours will
fall next to be considered, and so lead up to the state-
ment of the present position of the subject, and of the
problems which confront the textual student of to-day.

The first collection of various readings (apart from
those given in the margin of Stephanus’s 1550 edition)

1 The British Museum Catalogue contains forty-five editions of the
Greek New Testament between 1516 and 1624, besides forty-seven in which
the Greek is accompanied by a Latin text, and twelve complete Greek
Bibles. Nearly thirty of these were printed at Basle alone. Critically,
these are all substantially reproductions of Erasmus, Stephanus, or Beza ;

but their number shows the demand for copies of the Scriptures in the
original tongue.

sity

274 CRITICISM OF WEW TESTAMENT). cx

was made in England, in the Polyglott Bible edited by
Brian Walton, afterwards Bishop of Chester. The fifth
of his six great volumes (published in 1657) contains
the New Testament in Greek, Syriac, Latin, Ethiopic,
and Persian. The Greek text was that of Stephanus,
to which were added, at the foot of the page, the
readings of the Codex Alexandrinus (A), the recently
acquired treasure of the Royal Library; while in the
sixth volume a collation was given of fifteen other
authorities, in addition to the sixteen cited by
Stephanus. Among these were D and D,, the rest
being minuscules, of which the best is that now known
as Cod. 59. These collations were made by Arch-
bishop Ussher.

Walton’s work was extended by John Fell, Dean of
Christ Church and afterwards Bishop of Oxford, who
in 1675 printed a Greek Testament from the Elzevir
of 1633 with a critical apparatus in which he claims
to give variants from more than 100 MSS. Most of
these were appropriated from Stephanus, Walton, and
other collections; but Fell added the readings of
eighteen MSS. (mostly in the Bodleian) and of the
Coptic (4. Bohairic) and Gothic versions.

Fell not only produced an edition of the New
Testament himself, but also contributed largely to the
production of a work which far eclipsed his own and
all that had gone before. This was the edition of Dr.
John Mill, Fellow of Queen’s College, Oxford, and
Principal of St. Edmund’s Hall. Mill began to make
collections for the purpose about the date of the
appearance of Fell’s edition; and Fell, hearing of his
work, not only encouraged him to proceed, but under-
took to defray the expenses of publication. Mi§ill,
however, was a conscientious worker, and in his zeal
for the collection of all available evidence his work

τ TEXTUAL CRITRCISM IN THE PAST (275

progressed but slowly; so that when Fell died in
1686, only the first twenty-four chapters of St. Matthew
had been printed. The loss of Fell’s pecuniary aid,
and the ever-increasing growth of materials, still further
delayed the progress of the edition, and it was not
until 1707 that it at last saw the light. Mill’s text
was that of Stephanus’s edition of 1550, but he appended
to it collations of seventy-eight MSS. (besides those
quoted by Stephanus), and of all the versions to which
he could have access (including the Old Latin, Vulgate,
and Peshitto) ; and he was the first editor to take the
pains to collect the evidence of patristic quotations to
any noteworthy extent. Moreover, he prefixed to his
work, when it was complete, valuable Prolegomena,
which showed that he knew how to use his materials
as well as collect them. The MSS. collated by Mill
imelude the uncials’ Ay By Dy) EY Bop Ἐς WK and “the
good minuscules 28, 33, 59, 69, 71. His collations
do not come up to the modern standard of complete-
ness and precision, but they are far in advance of
anything that had been done previously, while his
Prolegomena include a mass of invaluable material ;
and the importance of his work, as showing both what
had to be done and the method of doing it, can hardly
be over-estimated.

A somewhat rearranged edition of Mill, with addi-
tional collations of twelve MSS. (mostly at Paris, and
including the uncials C and G,), was issued by L.
Kuster at Rotterdam in 1710; but this hardly breaks
the sequence of English contributions to textual criti-
cism, since Kuster himself had been a student at
Cambridge, and his enterprise may have been due to
the fact that he was there a pupil of the great scholar
Richard Bentley, whose labours on the text of the
Bible rivalled those of Mill in extent, though not in

276 CRITICISM OF NEW TESTAMENT ‘cx:

successful achievement. A more noteworthy under-
taking, though less so in performance than in idea, was
that of Dr. Edward Wells, who, between the years
1709 and 1710, issued at Oxford a revised text of the
Greek Testament, with English translation and notes.
The attempt was no doubt prompted by Mill’s great
work, and was rendered possible by the wealth of
various readings therein first given to the world. Wells’
edition was, in fact, the forerunner of that third stage
of textual criticism which only reached its full develop-
ment in the nineteenth century, the reconstruction of a
purer and more ancient text of the New Testament.
The attempt was premature, since fuller materials had
still to be collected, and the principles of their applica-
tion discovered ; but the intention was right.

Mill’s work was, however, fruitful in another direc-
tion, in bringing into the arena of textual criticism the
greatest scholar of that age, and one of the greatest
of any age, Richard Bentley. Mill had contemplated,
in addition to his main work, the publication of com-
plete texts of the more important MSS. then known,
namely, the Codices Alexandrinus, Bezae, Claromon-
tanus, and Laudianus; and as early as 1691 Bentley
had addressed a Latin epistle to him,’ urging the
performance of this undertaking. Again, after Mill’s
death, when his edition of the New Testament was
assailed by many (notably by Dr. D. Whitby) on the
ground that the mass of various readings collected by
him cast doubt on the integrity of our Bible text,
Bentley issued a tract” in which he defended the true
principles of textual criticism, and the importance of

1 Epistola ad Johannem Millium.

2 Remarks upon a late Discourse of Free Thinking, in a letier to F. 227.»
D.D., by Phileleutherus Lipstensts (1713); written in the character of a
German student, and provoked primarily by the work of one Collins, who
had taken up Whitby’s arguments and used them with a sceptical purpose.

vic TEXTUAL; CRITICISM IN THE) PAST 277

studying it. And not only did he teach this precept
—he also devoted many years of his life to practising
it. About this time he began to employ scholars,
among whom the most notable were John Walker and
the Swiss J. J. Wetstein, to make collations for him in
foreign libraries; and in 1720 he issued formal pro-
posals for printing a revised text of the Greek Testament
and of the Vulgate, with a critical apparatus. His
first examination of the MS. evidence showed him the
similarity of the oldest authorities, Greek and Latin ;
and he believed that a full comparison of them would
enable him to restore the original text with almost
absolute certainty, and with almost absolute identity
between the two languages. This belief, which is the
keystone of his undertaking, is thus expressed in the
third paragraph of his Proposals for Printing :—

“ The author believes that he has retrieved (except
in very few places) the true exemplar of Origen, which
was the standard to the most learned of the Fathers, at
the time of the Council of Nice and two centuries after.
And he is sure that the Greek and Latin MSS., by
their mutual assistance, do so settle the original text
to the smallest nicety, as cannot be performed now in
any classic author whatever ; and that out of a labyrinth
of thirty thousand various readings, that crowd the
pages of our present best editions, all put upon equal
credit, to the offence of many good persons, this clue
80 leads and extricates us, that there will scarce be two
hundred out of so many thousands that can deserve the
least consideration.”

As a specimen, the last chapter of the Apocalypse
was printed with the proposals, but the apparatus
criticus was only given in skeleton outline. The
materials had still to be collected and arranged, and in
the end the vastness of the task he had undertaken, in

278 CRITICISM OF NEW TESTAMENT ocu.

re-editing both the Vulgate and the Greek Testament,
proved too much for him. A large number of collations
were made for him, including several of the most
valuable Latin MSS., and two of the Codex Vaticanus ;
but Walker, who had been chosen as co-editor, died in
1741, and Bentley in 1742, and the edition remained
unexecuted. Only the materials were left, and these
were eventually bequeathed by his nephew to the
library of Trinity College, Cambridge, where they still
remain.’ It has been suggested that the fuller know-
ledge of the complexity of the problem, brought to
him by the increase of his materials, convinced him that
his great principle of the identity of the Greek and
Latin texts, and the resultant certainty of his con-
clusions, was untenable, and so indisposed him to
continue his undertaking; but it is probable that the
quarrels which filled the latter part of his life at Cam-
bridge had even more to do with his failure to bring
his great work to the birth.

So far the study of New Testament textual criticism
had been practically confined to England ; now (in part
through Bentley’s own impulse, as will be seen) it
passed out into the wider sphere of Europe in general,
and for a time was little practised in the land of its
birth. In 1734 J. A. Bengel published at Tiibingen
an edition of the New Testament which marks an era
in the history of textual criticism. His text is mainly
that of the Textus Receptus, only altered when the
reading which he believed to be the true one had
already appeared in some printed edition. In other
cases the reading which he preferred is indicated in the

1 For an account of them see Bemileit Critica Sacra, by A. A. Ellis
(Cambridge, 1862), and, on the Vulgate MSS., Wordsworth and White’s
Vulgate, pp. xv-xxvii. The date of Walker’s death, which caused the
collapse of the edition after the death of Bentley, was established by Bishop
Wordsworth (Old Latin Biblical Texts, i. p. xxv).

wr TEXTOAL: CRITICISM IN ‘THE! PAST 279

margin, together with such other readings as he thought
sufficiently important to be thus distinguished. A select
apparatus criticus, drawn mainly from Mill, was added
at the end of the volume. What makes Bengel’s edition
specially noteworthy, however, is the fact that he was
the first to attempt any classification of his authorities
—a principle which has proved very fruitful in the
present generation. He divided his authorities (MSS.
and versions) into two groups, which he called African
and Asiatic, the former including the few most ancient
authorities, the latter the great majority of later date ;
and he gave the preference to the former.

These novel principles asserted by Bengel met with
sharp criticism ; and one of his most prominent oppon-
ents was J. J. Wetstein, Bentley’s former assistant, who
was himself engaged on an edition of the Greek
Testament. His study of textual matters began when
he was quite young, and he spent many years in
collating manuscripts in France, England, and Switzer-
land, before settling down in his native town, Basle,
where he entered the diaconate. Here his relatives
and acquaintances at first urged him to complete and
publish the results of his collections ; but subsequently
he became suspected of heresy, his work was opposed
and thwarted, and eventually, in 1730, he was ejected
from the ministry. In the same year he published the
Prolegomena to his proposed edition; but the change
in his circumstances obstructed any further publication,
and the delay, though it led to an increase in his
materials, led also to a change in the plan of his work.
He had at first thought of printing the text of the
Codex Alexandrinus, and next of putting forth a revised
text of his own; but ultimately he decided to print the
received text, with an apparatus criticus indicating what
was, in his opinion, the true reading. His edition

280 CRITICISM, OF (NEW TESTAMENT | cx.

appeared eventually at Amsterdam in 175 I—5 2,equipped
with prolegomena, various readings, and a valuable
series of illustrative passages from sacred and profane
literature (including Jewish), which is a very distinctive
feature of his work. His apparatus criticus is remark-
able for its introduction of the system of manuscript-
notation which has continued in use ever since; and
it was enriched by readings from many MSS. not
previously collated. Inthe Gospels his list includes the
uncials A to O, and cursives I to 112; in the Acts and
Catholic Epistles, A to G,1 to 58; in the Pauline
Epistles, A to H, I to 60; in the Apocalypse, A to Ὁ,
I to 28; besides twenty-four Evangeliaria and four
Apostoli. Inthe accumulation of materials, accordingly,
Wetstein’s edition did considerable service ; but his
contributions to textual theory were very retrograde in
character. Largely, as it would appear, out of opposi-
tion to Bengel, he propounded the theory that all the
most ancient MSS. had been contaminated from the
Latin, and that consequently the later authorities, from
which the Textus Receptus was more immediately
derived, contained the more authentic tradition. This
hostility to the earliest authorities has found echoes in
our own time, but has never been carried so far as by
Wetstein ; and the general sense of textual critics has
condemned it unreservedly.

A far more valuable contribution to textual theory
was made by J. S. Semler, who reprinted Wetstein’s
Prolegomena at Halle in 1764, with comments of his
own, and subsequently published other treatises on the
same subject. Adopting at first Bengel’s classification
of all ancient authorities, which he called Eastern
(= Bengel’s Asiatic) and Western (= Bengel’s African),
and assigning them respectively to the recensions of
Lucian and Origen, he subsequently (in 1767) expanded

ym CTEXTUAL CRITICISM (IN THE: PAS! 51

this into a threefold division—(a@) Alexandrian, derived
from Origen, and found in the Syriac, Coptic, and
Ethiopic versions; (ὁ) Eastern, in vogue at Antioch
and Constantinople; (c) Western, embodied in the
Latin versions and Fathers. Semler’s theory derives
special importance from its adoption and extension by
his pupil J. J. Griesbach, whose name ranks with the
highest in the history of textual criticism. Griesbach’s
activity ranged over a space of nearly forty years, and
he produced three editions of the New Testament.
The first of these was published at Halle in parts in
1774-75, containing a revised text and select apparatus
criticus;' a new issue of the first part, giving the
Synoptists in their usual order, instead of in the form
of a harmony, as previously, appeared in 1777. The
apparatus was taken mainly from Wetstein, with cor-
rections and additions due to Griesbach’s own collations.
His second edition appeared in two volumes, published
respectively in 1796 and 1806, by which time the
available material had been greatly increased by the
industry of Matthaei and others who will be mentioned
below. The plan of this edition is the same as that of
its predecessor, but several changes are made in the text
adopted. A small edition published in 1805 introduces
yet further changes into the text of the Gospels, and
must be taken as representing Griesbach’s final judgment,
though he never altered the text so largely as would
have been necessary if he had carried out his own
textual theories thoroughly. These theories had already
been stated in the Prolegomena to the several parts
of his first edition. The authorities for the Pauline

1 Besides the apparatus criticus properly so called, Griesbach gives a list
of select readings, with symbols denoting the degrees of probability which
he believes to attach to them. This system will be familiar to many
students in the small pocket edition of Griesbach’s New Testament which
used to be rather commonly in circulation.

2382 CRITICISM OF (NEW TESTARIENT > cu:

Epistles were classified into two families, the Alexan-
drian and the Western ; those for the Gospels into three
families, Alexandrian, Western, and Constantinopolitan.
This is practically the division of Semler, but Griesbach
gave greater fulness to it by his precise assignment of
the principal authorities to the several families. In the
Alexandrian family he placed the uncials C, K, L, the
important cursives 1, 13, 33, 69, 106, 118, the Coptic
(44. Bohairic), Ethiopic, Armenian, and Harkleian Syriac
versions, and the quotations in Origen, Clement of
Alexandria, Eusebius, and a few other Fathers ; in the
Western, D, the Latin versions, and sometimes the
Peshitto Syriac; in the Constantinopolitan, which he
regarded as a later compilation from the other two, A,
and the great mass of late uncials and cursives. To
the evidence of this third group, numerically preponder-
ant though it is, he attached but little weight in
comparison with the two older groups. It will be
seen later how nearly Griesbach’s theory anticipated
that which holds the field among New Testament
students to-day.

Meanwhile, in the interval between δεῖ: first
and second editions, much had been done to increase
the store of materials available for the textual critic.
C. F. Matthaei produced a New Testament in twelve
parts between 1782 and 1788, containing the Greek
text with the Latin Vulgate and a large apparatus
criticus. His text, which is based upon late MSS.,, is
of little value ; but his collations are of considerable
importance. Being a professor at Moscow (though by
birth a Thuringian) he had access to many MSS. which
had never previously been examined ; and his collations
are generally good and accurate. Fifty-seven MSS.
were thus added by him to the available stock of
evidence, including the uncial V; and he also edited

vip TEXTUAL CRITICISM IN THE’ PAST 283

with great accuracy the Codex G, at Dresden. Like
Wetstein, he was wholly opposed to the critical theories
of Griesbach, whom he attacked bitterly, holding that
the text of the later MSS. was much to be preferred
to that of the most ancient. A smaller edition of the
New Testament was published by him in 1803-7, in
which collations of some additional MSS. were used.
Almost simultaneously with Matthaei’s first edition
F. K. Alter, Professor of Greek at Vienna, published
in full the text of a manuscript in the Imperial Library
of that city (218) with collations of twenty-one other
MSS5:,) in) the isamen library «(178687 )) .Aclarger
addition to the stock of materials was made by three
Danish professors, Birch, Adler, and Moldenhauer,
who were sent by King Christian VII. to examine
MSS. throughout Europe, especially in Italy, Germany,
and Spain. The results of their labours were em-
bodied in an edition of the New Testament by Birch
(1788-1801). Stephanus’ text of 1550 was taken
as the basis of this edition, and collations were given
of a very large number of manuscripts. The full
list includes 172 MSS.; but many of these were only
partially examined by Birch and his colleagues, and
some had been more or less collated by other scholars
previously.

The materials available for the purposes of textual
criticism had thus been very greatly increased during
the last quarter of the eighteenth century, and at the
same time some very valuable work had been done
in the way of publishing in full the texts of some
of the most important MSS. It will be remembered
that Mill had contemplated such a scheme, and Bentley
had advocated it; but neither of these great critics
was able to perform the task. Hearne had indeed
published the Codex Laudianus (E,) in full in 1715,

284 CRITICISM ‘OF (NEW TESTAMENT” cu:

but his example was not followed by any one for
nearly half a century. Then, in 1762, Knittel pub-
lished the Wolfenbiittel palimpsests P and Q; in
1786 Woide edited the New Testament portion of
the Codex Alexandrinus (A); in 1789 Giorgi pub-
lished the Graeco-Sahidic fragments known as T;
in 1791 Matthaci, as just mentioned, published G, ;
and in 1793 Kipling published the very important
Codex Bezae (D). Thus by the close of the century
a considerable body of evidence was at the disposal of
any scholar competent to use it.

The early years of the nineteenth century were not
favourable to works of scholarship; and although the
French invasion of Italy led to the temporary sojourn of
the Codex Vaticanus in Paris, and thereby to a better
knowledge being obtained of its character, yet on the
whole the Napoleonic wars interposed a serious obstacle
to the development of textual criticism. Nearly a
quarter of a century separates the last edition of
Griesbach from the next works of importance in this
department of knowledge. When at last, in the early
thirties, the stream burst out again, it was in the
works of two German scholars, Scholz and Lachmann ;
but though their publications were nearly contem-
poraneous, the positions which they hold in the history
of textual criticism are far apart. Scholz represents
the end of the old period, Lachmann the beginning of
the new.

The importance of J. M. A. Scholz lies in the fact
that the list of manuscripts attached to his edition of
the New Testament gathers up, with large additions,
all the lists of his predecessors, and forms the basis
of the numerical catalogues of MSS. which appear in
our books of reference to-day. He travelled through
Europe, cursorily examining all the manuscripts of

wi, LEXTUAL CRITICISM IN THE PAST 2365

the New Testament which he could find, and so
compiling a long list, not indeed of collated MSS.
but of MSS. known to be in existence, thus pointing
the way to others to carry on a work which one man
alone could not do. His catalogue (which continues
the system of numeration initiated by Wetstein) in-
cludes the uncials from A to A (A to H of Acts, A to
I of Paul, A to C of Apoc.) and the minuscules Evan.
1-469, Act. 1-192, Paul. 1-246, Apoc. 1-88, Evl.
1-181, Apost. 1-58. The list abounds in mistakes,
as later scholars have shown; but as pioneer work
it did excellent service. In other respects Scholz’s
edition (which was published in 1830-36) is less
valuable. His text is not very different from that
of Griesbach, although his critical principles were the
reverse of Griesbach’s. Accepting Bengel’s classification
of MSS. in two families, which he calls Alexandrian
and Constantinopolitan, he selected for preference, not
the former, which includes all the earliest MSS. and
versions, but the latter, which comprises the great
mass of later authorities. His extensive examination
of minuscule MSS. in the libraries of Europe had
impressed him with the general uniformity of type in
their texts, which he regarded as evidence of authen-
ticity ; and the representation that all the most ancient
MSS. exhibited a different type of text was met by
him with the argument (revived half a century later
by Dean Burgon) that these MSS. had only survived
because, being erroneous, they had been less used.
Nevertheless he did not form the text of his edition
consistently with this theory, retaining many “ Alex-
andrian ” readings which, according to his own principles,
he should have discarded.

With Scholz ends the second stage in the history
of textual criticism, that of which the chief feature was

286 CRITICISM OF NEW TESTAMENT cu.

the accumulation of evidence. Throughout this period
the Textus Receptus had held its position practically
unshaken, while the evidence which was ultimately to
overthrow it was being sedulously brought to light by
the labours of successive scholars. The third period,
which may be dated from 1831 (the year which re-
formed parliamentary government in England), is the
period of the application of critical principles to the
accumulated mass of materials. It is true that the
collection of materials has by no means ceased since
1831; and it is true that some application of critical
principles had been made before that date, notably by
Bengel and Griesbach; but the broad features of the
two periods are quite distinct. Neither Griesbach nor
Scholz, the last representatives of the earlier period,
had the full courage of their opinions. Griesbach,
believing that the older MSS. and versions were of
quite preponderant value, yet made relatively few
alterations in the received text. Scholz, believing that
the older authorities were wholly unreliable, yet re-
tained many readings which had been adopted from
them by Griesbach.

The new era begins with the name of Karl Lach-
mann, who illustrates, not for the last time, the
stimulus which may be given to Biblical criticism by
the appearance in the arena of a scholar trained in
other studies. Lachmann was a great classical scholar
before he turned his attention to the text of the
New Testament; and when he did so, he applied
to it the critical principles which he had practised
in editing the classics of Greek and Roman literature.
For the first time in the history of New Testament
criticism, he cast aside the received tradition altogether,
and set about reconstituting the text from the most
ancient extant authorities. His first edition appeared

vi LEXTUAL (‘CRITIGISM IN THE PAST \'287

in 1831, unaccompanied by any statement of the
authorities used or the principles followed, except a
brief note stating that he had aimed at reproducing
the text of the most ancient Churches of the East, and
where there was doubt upon this head he had preferred
that reading which was supported by the Italian and
African Churches, and that consequently he had
ionored the evidence! ofthe: Textus.Receptus.’ For
further information the reader was referred to an
article in a German periodical (ZTheologesche Studien
und Kritiken, 1830). The natural consequence of
this reticence was that his work was misunderstood,
even by those who would have been most likely to
sympathise with it. When, however, its character
came to be appreciated, he was urged to produce
another edition with a full statement of the principles
followed and the authorities consulted for the con-
stitution of the text. This edition, in which he had
the assistance of P. Buttmann, appeared in 1842-50,
and differs from the first mainly in including the
Vulgate text as well as the Greek; and in the fact
that the evidence of the Latin authorities was taken
into account throughout, and not only when the
Eastern authorities disagreed among themselves.

The method of Lachmann, as practised in both
editions and expounded in the second, was this. Putting
entirely aside the whole mass of later authorities, he
confined his attention to a small group of very ancient
manuscripts, versions, and Fathers, and by their aid he
believed that he could recover, not indeed always the
actual words of the authors of the sacred books, but
the earliest form of text which enjoyed wide circulation
among the Churches, which might probably be assigned
to the end of the fourth century. The authorities which
he regarded as available for this purpose were, from the

288 CRITICISM OF NEW TESTAMENT (ἢ

East, the manuscripts ABCH,PQTZ and the writings
of Origen ; and from the West the bilinguals DD,E,G,,
the Old Latin MSS. adcg, the Codices Amiatinus and
Fuldensis of the Vulgate, and the writings of Irenaeus,
Cyprian, Lucifer, Hilary, and (for the Apocalypse)
Primasius. This list may seem fairly long, but it
must be observed that of the Greek MSS. which
formed the first line, of evidence, C \is) ‘imperfect,
H;POTZ are only fragments, and B was at that
date only very inadequately collated. For the best
MS. of the Vulgate too, the Amiatinus, Lachmann
only had access to an imperfect collation ; while the
Syriac and Coptic versions were not utilised by him,
on account of his ignorance of these languages. He
also tied his own hands unnecessarily, by binding him-
self to follow the majority of his authorities, without
regard to the internal probabilities of the rival readings ;
thinking that thus he would eliminate altogether the
“personal equation,” while if he thereby sometimes
introduced erroneous readings, they were readings
which must have enjoyed some considerable circula-
tion in the early Church. Some deduction has con-
sequently to be made from the value of his edition, on
account of the insufficiency (in many parts of the New
Testament) of his authorities and the inelasticity of his
principles ; but his work is nevertheless epoch-making,
from its courageous rejection of the Stephanus-Flzevir
text and deliberate application of critical principles,
which in the main were sound, to the recovery of a
more authentic text from the most ancient authorities.

We come now to the name which probably is the
best known of all the scholars who have devoted them-
selves to the restoration of the Bible text, Constantin
Tischendorf. His life’s work was of two kinds, as a
publisher of the exact texts of ancient MSS. and asa

wo TEXTUAL CRITICISMAIN ΤΗΝ PAST 285

critical editor of the Greek Bible. His achievements
in the first capacity have been repeatedly mentioned
in Chapter III. He discovered and edited the Codex
Sinaiticus (8); he succeeded in producing a more
accurate edition of B than had previously been in
existence ; he edited C, deciphering much of it for the
first time; he also edited the MSS. then denoted by
the symbols D,E,F*II?LM,N (so far as it was then
known) O70'PP,QRW?W°Y@* Most of these are small
fragments of uncial MSS., many of which he was the
first to discover or to make use of. In addition he tran-
scribed (but without publishing) H,MOII, and minutely
collated, EF,GG,HH,K1,0"O,/STUXW™ToOtA?
and Pap.? Consequently, although other scholars have
also worked on many of these MSS., we are indebted
to Tischendorf for a large portion of our knowledge of
the uncial evidence for the New Testament text. In
addition, he edited the Old Latin MSS. e& and gue
(besides the bilinguals ὦ and e,) and the Codex Ami-
atinus of the Vulgate, and diligently collected evidence
as to the readings of other MSS., versions, and patristic
quotations from all available sources.®

In the use of the materials thus collected for the
recovery of the true text of the New Testament, he
was not less assiduous. No less than eight editions of
the Greek New Testament were issued by him, the two
last appearing in a double form, a larger and a smaller.’

1 Now D,E, (F? is not now included in the lists, being merely an O.T.
MS. with a few verses of the N.T. in the margin), 065, 066, 067, 068, 078,
079, 088, 096, 097, LM3N, Εν]. 1345, Εν]. 1348, PP,QR, 0115, 0130,
0133, 0106.

2 Now H,M, 050, II and EF,GG,HH,KL,, Εν]. 1346-1352, 082, ST,
083, 084, Evl. 1353, UX, 0116, 050, I', 089, 090, 091, 0136, A.

3 For a full account of Tischendorf’s labours as a collector of evidence,
and the sources used by him when he did not work from first-hand know-
ledge, see Gregory, Prolegomena, pp. 1-44.

4 The dates of the several editions are 1841, 1842, 1842, 1849, 1850,
1854, 1859, 1869-72. ‘The fifth and sixth (the latter containing also the
Latin and German texts of the New Testament) were frequently reprinted.

U

290 CRITICISM (OF (NEW (TESTAMENT?) cu:

In all these editions a revised text was given, with a
constantly growing critical apparatus. The earlier
editions were not of much permanent value; but the
last embodies the fullest critical apparatus hitherto
available for scholars, and may consequently be re-
garded as the standard critical edition of the Greek
Testament.’ This text is formed by the free exercise
of his own judgment on the evidence provided by the
various authorities. He does not tie himself down to
a limited number of authorities, like Lachmann ; neither
does he adopt any classification of families, like Gries-
bach. On the other hand he has no prejudice in favour
of the Textus Receptus. The traditional claims of the
Stephanus-Elzevir text had been shattered by Lach-
mann; and Tischendorf, like Lachmann, attached a
preponderant weight to the testimony of the oldest
authorities. Unfortunately his critical judgment was
not so sound or stable as could be wished; and he
was liable to be over-influenced by the witnesses which
he had last studied. According to Scrivener, his seventh
edition differs from his third in 1296 readings, nearly
half of these variations being reversions to the Textus
Receptus, due to the study of the cursive MSS. which
he had undertaken in the interval. Between his seventh
and eighth editions came his great discovery of the
Codex Sinaiticus ; and consequently his latest text, in
which the new evidence was, naturally enough, allowed
a somewhat excessive weight, differed from its pre-
decessor in no less than 3369 places. Tischendorf’s
method left too much to the personal equation of the
critic; and consequently, valuable as his text is as
representing the opinions of one who gave a strenuous
lifetime to the study of the subject, it could not in any

1 A new critical edition, with full apparatus, is in preparation by von
Soden, and it is understood that another is contemplated by Gregory.

ἡ TEXTUAL CRIEICISM' IN THE PAST 291]

sense be final, or even mark a striking epoch in the
history of New Testament criticism.

Tischendorf died (in 1874) before completing, or
even advancing far in, the intended Prolegomena to his
last edition; and the task of supplying the omission
was committed to Dr. C. R. Gregory, aided by his
fellow-American Dr. Ezra Abbot, who, however, also
died before the work was concluded. These prole-
gomena, which appeared in the course of 1884-94,
contain an enormous mass of information with regard
to the authorities for the New Testament text, and
form (with Scrivener’s work, to be mentioned later) the
standard book of reference on the subject.

Tischendorf’s fame as the discoverer of the Codex
Sinaiticus, and his energy in the acquisition and publi-
cation of other early MSS., or fragments of MSS., gave
weight and popularity to his texts of the New Testa-
ment, and did more than anything else to familiarise
the educated public in general with the idea of the in-
sufficiency of the Textus Receptus. But he was not in
reality more energetic in his labours than his English
contemporary, 5. P. Tregelles. Incited to the task
by observing the persistency of Scholz in rejecting the
evidence of the earliest authorities (his “ Alexandrian ”
group), Tregelles embarked about 1838 on the prepara-
tion of an edition based upon the opposite principle,
namely, the ignoring of the claims of the Textus
Receptus (in which he did not then know that he had
been anticipated by Lachmann) and the determination
of the text de novo, in accordance chiefly with the most

1 Dr. Gregory’s new work, Zexthritik des Neuen Testamentes, which has
been published in three volumes (Leipzig, 1900-1909) is substantially a
German translation of the Prolegomena, with additions to bring it up to
date. Its chief new feature is a somewhat fuller discussion of the Lection-
aries ; but much additional information is contained in it. The revised

notation of the MSS. subsequently introduced by Gregory has been
described above (p. 56).

οι CRITICISM ‘OF NEW (TESTAMENT \ cx:

ancient MSS., versions, and Fathers. A text of the
Apocalypse was published in 1844; but before pro-
ceeding with the rest of the New Testament, Tregelles
set himself to collate personally all the most important
MSS. throughout Europe, a task which occupied the
greater part of his remaining years. All the extant
uncial MSS. were examined by him, and all that had
not previously been published were collated by him,
together with some important cursives;* and by an
interchange of collations in several instances with
Tischendorf the accuracy of both was tested, and
their results raised to a very high degree of probability.
His edition appeared in parts between the years 185 7—
1872, illness making it necessary for him to accept help
from friends in the preparation of the last part. The
Prolegomena which he had intended were still unwritten
at his death in 1875, but a sketch of his views on the
principles of textual criticism was compiled out of his
other writings, together with important addenda to the
apparatus criticus, by the Rev. A. W. Streane, under
the direction of Dr. Hort.

Tregelles’ edition, which contains the Vulgate (from
the Codex Amiatinus) as well as the Greek text, has
no such elaborate apparatus criticus as Tischendorf’s,
the cursive MSS. being wholly neglected, with the
exception of) 1, 33,and, 60. The) uncial MSS. θη
the other hand, are fully represented, together with
the Syriac, Latin, Coptic, Ethiopic, and Armenian
versions, and the early patristic quotations, the im-
portance of which in fixing the date of the types of
text represented in them he recognised and emphasised.
In general, his text is constructed on the principles
advocated by Bentley, Lachmann, and Tischendorf,

1 For details, see Tregeiles’ Account of the Printed Text of the Greek
New Testament, pp. 151-174.

vt) EEXTUAL CRITICISMOIN ‘THE’ PAST 293

namely, in dependence on the most ancient authorities
without reference to the Textus Receptus; but he
resembled Tischendorf rather than Lachmann in not
tying himself down to hard and fast rules, while he
did not adopt the system of classification advocated by
Griesbach. On the whole his text does not differ very
creatly from that of Tischendorf, their general prin-
ciples of criticism being much the same; and the
resemblance would probably have been greater but
for the fact that his edition of the Gospels had been
published before the discovery of x or the appearance
of Tischendorf’s edition of B.

Another English scholar whose name requires men-
tion at this place is Ἐν H. A. Serivener. It is true
that he constructed no new text of the New Testament,
but he did much to collect material for others and to
popularise a knowledge of the principles and most im-
portant results of Biblical criticism. He edited Codex
Bezae (D) and Codex Augiensis (F,), collated some
hifty MSS., made minute and careful examinations of
the earliest printed Greek Testaments and of the Eng-
lish Authorised Version, published the text of Stephanus
with a collation of the readings of Lachmann, Tischen-
dorf, and Tregelles for the use of students, and, above
all, wrote his Plazn Introduction to the Criticism of the
New Testament, the fresh and readable style of which
has done much to popularise a knowledge of the sub-
ject, and which is still the fullest English authority on
the materials of textual criticism. Three editions of
this work appeared during its author’s lifetime, in
1861, 1874, and 1883; and after his death a fourth
edition appeared (in 1894), under the editorship of
the Rev. E. Miller, the chief features in which are a
great extension of the list of cursive MSS., and an
admirable series of chapters on the Versions, contri-

294 CRITICISM OF NEW TESTAMENT cu.

buted by scholars with special knowledge of each
version in question.

We come now to the two scholars whose joint work
has been epoch-making, in the literal sense of the word,
in the history of New Testament criticism, the two
Cambridge friends, B. F. Westcott, late Bishop of
Durham, and Ἐκ J. A. Hort, Professor of Divinity in
the University of Cambridge. The epoch- making
character of their work lies not so much in any
absolute novelty in their views as in the thoroughness
with which they were elaborated, and the influence
which they have exerted on all subsequent criticism
of the New Testament. It has coloured all that has
been written on the subject for the last thirty years,
and supplies the basis of all work done in this field
to-day. Indeed, it is the chief defect of Scrivener’s
Introduction, regarded as an aid to students, that,
having been originally written before the promulgation
of Westcott and Hort’s theory, it has never in its later
editions taken it fairly and fully into account.

Westcott and Hort’s edition of the New Testament
was published in 1881, and contains a revised Greek
text without apparatus criticus, but with critical notes
on special passages and a volume of elaborate pro-
legomena treating of the principles upon which the
text is constructed. It is in these that the importance
of the work lies. Westcott and Hort did not them-
selves collate or edit manuscripts, but devoted them-
selves to the study of the materials collected by others,
and to the elaboration of a theory of the history of the
early transmission of the New Testament text, which
might serve as a guide to the discovery of the true
text among the multitude of divergent witnesses that
have come down to us. Briefly, this theory is a
revival of Griesbach’s classification of all textual authori-

Tie tea ORE Ieee iN: ΤΗΝ RAST 295

ties into families, with a decided preference for one
which, though very early in point of date, is numerically
insignificant in comparison with that which includes the
great mass of later witnesses; but in view of the
present importance of Westcott and Hort’s restatement
of this theory it is advisable to set it out at greater
length, following the lines adopted by themselves in
their exposition of it.’

The principles upon which the system of classifica-
tion of families rests are briefly these. The simplest
method, frzma facie, of procedure when various read-
ings are offered by different authorities is to adopt that
reading which most commends itself to your instinct
or common sense ; but this will lead to different results
with every critic and carries no weight of proof with it.
Moreover, the reading which commends itself to your
instinct as probable may similarly have commended
itself to the ancient scribe as an improvement on the
text which he had before him ; it is indeed a recognised
fact in textual criticism that errors are often introduced
through a scribe misunderstanding his text and altering
it to something that seems to him easier and more
natural. Hence the reading which prema facie is less
probable may often turn out on examination to be
more probable, as affording an explanation of the
other ; for that reading must be considered the most
probable which provides the best explanation of the
origin of its rivals. But even here we must depend
much on the taste and judgment of the individual
critic. We still have to make allowances for the per-
sonal equation, and that is exactly what we wish to

! The Introduction to Westcott and Hort’s New Testament was written
by Hort, and for brevity’s sake it is convenient to refer to it by his name ;
but it must be understood that Bp. Westcott fully shared and adopted the
views expressed in it, which had been elaborated by the two friends in
thirty years of close intercourse.

206 CRITICISM OF NEW TESTAMENT > ‘cx!

eliminate. A first step in this direction is made if we
can pass from simple criticism of the various readings
as they arise, to the criticism of the documents in
which they are contained. By a study of those cases
in which intrinsic probability allows a fairly certain
judgment to be formed, we come to see which of our
authorities usually have good readings, and which the
reverse ; and we can then apply this knowledge of the
character of our authorities to cases in which, so far as
intrinsic probability goes, a choice might be difficult.
The evidence of a few witnesses who have generally
been found trustworthy will naturally outweigh that of
many witnesses whose character has stood the test less
well. Knowledge, therefore, of the documents must
precede a final judgment upon divergent readings.
But a still higher degree of certainty, and a fuller
exclusion of the element of personal prejudice, may be
obtained if we can classify our authorities into groups
descended from a common ancestor. The testimony
of individuals is thereby checked and corrected, and
their evidence carried back to the date of their ancestor,
possibly a century or more behind their own date.
Moreover, it may be possible to distinguish between
the characters of the several groups. One group may
be shown to go back to an earlier date than another,
another to owe its origin to the revising hand of a
particular critic (like the MSS. which represent Lucian’s
edition of the Septuagint, or Alcuin’s of the Vulgate),
or to belong to a certain country or town (like the
MSS. of the Vulgate which can be traced to North-
umbria or St. Gall). Such a discovery will affect, for
better or worse, our opinion of the authorities contained
in each group, and will help us to understand and
estimate the value of their readings. We shall see
further into the history of the transmission of the text

vit) TEXYPUAL CRIDTRISM IN THE PAST 297

of the author with whom we are dealing; and our
knowledge will enable us to judge between the claims
of rival readings on some definite and general principle,
instead of upon the vagaries of individual taste.

It is the business, therefore, of the critic of any
ancient author, first to study the individual readings
and the authorities for them ; then to form an estimate
of the character of the several authorities ; then to see
how far these authorities can be grouped as descendants
of a common ancestor, and which family has the
greatest claims to respect ; and finally to return to the
individual readings, and revise, in the light of his
acquired knowledge of the value and inter-relation of
the several authorities, his first provisional estimate of
their comparative probability. This is the method
which is applied to all textual criticism ; and what we
have now to see is, how does it apply to the text of
the New Testament ? ,

Now when the textual authorities, which have been
enumerated in the previous chapters, are examined,
they are found to fall into three or four groups, more
or less clearly markéd;' that is) certain “MSS; and
versions are habitually found in agreement with one
another, and opposed to certain other groups of MSS.
and versions. One group is formed of the codices
ΝΣΦ and the great mass of later uncials and cursives,
with considerable support from the Peshitto Syriac and
(in the Gospels) from the uncials A and C. Another
group, much smaller in numbers, but eminent in point
of age, includes the uncials sBLT# (with occasional
support from others, such as PORXZ, A in St. Mark,
and A and C in the Acts, Epistles, and Apocalypse)
and the Coptic versions (especially the Bohairic) ;
some of the cursives also are frequently found in this
group, notably 33 and 81. Yet a third group is com-

298 CRITICISM OF NEW TESTAMENT) | ‘cx.

posed of the uncials DD,E,F,G,, some of the cursives
(Ki: 28, '23 550888) 505, 614; .700,4876),the (Ola
Syriac and Old Latin versions, and sometimes the
Sahidic. These groupings are not indeed constant, all
MSS. occasionally deserting the family to which they
properly belong and allying themselves with their
habitual opponents ; but in general it is found to be
observed. Other authorities join themselves now to
one group, now to another, and can only be described
as possessing mixed texts.

These three groups or families are those which we
have above provisionally denoted by the letters a, 8, and
6; and to these Hort adds a fourth, which is a kind of
subdivision of the second, not found wholly in any one
MS.,.but to be discerned when some members of that
group, notably xnCLX 33 and the Bohairic version,
differ from the other members headed by B. The
readings referred to this group, which may be indicated
by the letter y, are not generally of great importance,
consisting mainly of slight verbal alterations, such as
might be prompted by a desire for correctness of style.

How, then, are we to judge between these rival
families, so as to know to which we ought normally to
give the preference? The answer is, By the evidence
of the Fathers, whose quotations enable us to locate
these groups approximately both in time and in space.
The key to Hort’s whole theory lies in the proposition
that πο reading strictly belonging to the a-famuly ts found
nm any Father before Chrysostom. From Chrysostom
onwards, this type of text becomes frequent, until it
almost monopolises the cursives, and (in a somewhat
corrupted form) provides our Textus Receptus. To
this family, the establishment of which he traces to
the neighbourhood of Antioch in the latter part of the
fourth century (the time and the place of Chrysostom’s

vin TEXTUAL CRITICISM IN THE PAST 299

principal literary activity), Hort gives the name of
Syrian; and on account of the comparative lateness of
its origin, “ Syrian” is with him a term of the utmost
reproach. The β, y, and 6-types of text, on the other
hand, find attestation among the Fathers of earlier date.
The 8 and y-types are found pre-eminently in Origen,
and to a considerable extent in Clement of Alexandria
and Eusebius ; the 6-type in a the Fathers before the
end of the third century (including at times the three
just mentioned), but notably in Justin Martyr, Tatian,
Irenaeus, Tertullian, and Cyprian. To this last-named
family, which monopolises the early Latin authorities,
the name of Western has been given, though it will be
seen that it includes also authorities from the East, in
the shape of the Old Syriac version and Tatian. The
y-type, the evidence for which is wholly of Egyptian
origin, is styled by Hort A/evandrian ; while the §-type,
which alone remains, receives the name of Wewtral.
Between these three families, the patristic evidence for
all of which is of very early date, a decision must be
made on the grounds of internal probability ; and here
Hort’s verdict is emphatically in favour of 8. The
readings of the y-family appear to be due to deliberate
corrections in style and language, not important in
substance, but showing less signs of originality and
authenticity than those which they supplant. On the.
other hand the variants of the 6-family are very exten-
sive and important, and seem (in the opinion of Hort
and of most other critics) to be due to an extremely
free handling of the text at some early date, when
scribes apparently felt themselves at liberty to vary the
language of the sacred books, and even to insert addi-
tional passages of considerable length. As compared
with the readings of the -family, these vagaries of ὃ
lack authority and probability ; but in @ Hort finds all

300. CRITICISM: OF NEW TESTAMENT: > \ ‘cx:

the signs of authenticity and probability, and to that,
which he regards as neutral and substantially uncor-
rupted, he pins his whole faith.

Hort’s general view, then, of the textual history of
the New Testament is as follows. Corruption began
to leave its mark upon the tradition at a very early date.
Absolute accuracy of transcription was little prized, and
scribes felt themselves free to amend or extend the
text before them by additions or supposed corrections.
This was especially the case with regard to the Gospels,
since here there was the temptation, on the one hand, to
incorporate incidents of our Lord’s life which were
recorded in other writings or in oral tradition, and on
the other to correct one of the Evangelists from another.
Hence there came into existence, as early as the second
century, a type of text characterised by very free
departures from the true tradition. This type took
root in the Syriac Church, and was embodied in the
earliest known version of the Gospels in the Syriac
language ; but it was carried from the East to the West,
and being best known from its appearance in the Old
Latin version it may be named Western. In spite of
its very early origin, its testimony is not to be highly
regarded, on account of the liberties which it takes with
the text ; indeed, almost the only cases in which much
weight is to be attached to purely Western testimony
are those of omissions. A notable series of such
omissions occurs in the last chapters of St. Luke; and
the non-appearance of these passages in the Western
authorities seems to indicate that they were absent from
the original work, and have found their way into all
other authorities from some other source. In general,
however, a reading attested wholly or mainly by Western
evidence must, according to Hort, be regarded with the
gravest suspicion.

wm TEXTUAL CRITICISM IN (THE PAST οί

While this corruption of the sacred text was taking
place in the East and in the West, another kind of
modification, of a much less serious character, was being
introduced in the South. In Egypt, and especially in
Alexandria, the capital of Hellenistic literature, the
books of the New Testament were looked on witha
critical eye. Accustomed to literary Greek, accustomed
also to criticise classical authors, and it may be even at
times to correct them in accordance with their own
canons of style, the more cultured among the scribes of
Alexandria were continually under the temptation to
introduce verbal alterations into the compositions which
they had before them. Thus on the one hand the
literary training and traditions of Alexandria operated
in favour of accuracy in transcription, and so contri-
buted greatly to preserve the true text, which the
“Western ” scribes were imperilling ; and on the other,
those same literary instincts tended to produce altera-
tions in the text, but of a verbal character only, not
affecting the substance. To readings of this class the
name Alexandrian is given.

Such were the conditions of the New Testament
text during the early days of the Church, and indeed
nearly up to the date when Christianity became the
religion of the empire. As, however, the multiplication
of copies went on, and the divergences of text became
more marked, an attempt seems to have been made to
rectify the evil by an authoritative revision. The
principles upon which this revision was conducted con-
sciously or unconsciously, were (1) to combine divergent
readings when possible, (2) to smooth away rough-
nesses, to remove obscurities, and generally to produce
an easy and flowing text. The first of these principles
is seen at work in what are called (in Hort’s nomen-
clature) conflate readings, where two readings, each

302. CRITICISM OF (NEW "TESTAMENT. | cx

separately attested by a group of earlier authorities,
are combined into a single reading. A simple instance
may be found in Acts vi. 8, where one group of
authorities gives πλήρης χάριτος, another πλήρης
πίστεως, while the Codex Laudianus (E,) has πλήρης
χάριτος Kal πίστεως. In this case, however, the
authorities for the three readings do not fall into the
usual groups. In Mark ix. 49, on the other hand, we
have the readings πᾶς yap πυρὶ ἁλισθήσεται supported
by x (substantially) BLA, a few cursives and the Coptic
versions ; πᾶσα yap θυσία ἁλὶ ἁλισθήσεται supported
by D, the Old Syriac and Old Latin; and the conflate
reading πᾶς yap πυρὶ ἁλισθήσεται καὶ πᾶσα θυσία
ἁλὶ ἁλισθήσεται by ACNS and the great mass of later
uncials and cursives, the Peshitto Syriac and some other
versions. Or again, in Luke xxiv. 53, 8BCL Boh.
συγ read εὐλογοῦντες τὸν θεόν, D several Old
Latin and Vulgate MSS. and Augustine αἰνοῦντες τὸν
θεόν, and A and the great mass of later authorities
αἰνοῦντες καὶ εὐλογοῦντες τὸν θεόν. Hort examines
eight such instances from the Gospels, besides briefly
referring to others; and more (though probably not
very many) could be produced if necessary."

The revision was not confined, however, to the
manufacture of conflate readings, but included a general
softening and smoothing away of difficulties. | Con-
junctions are inserted to avoid harshness, proper names
substituted for pronouns for the sake of greater clear-
ness, and unfamiliar phrases turned into more familiar

1 A new instance has been indicated by Lake (Journal of Theo-
logical Studies, i.. 291) in Mark xiii. 11. Most authorities have μὴ
προμεῤιμνᾶτε μηδὲ μελετᾶτε, while xBDL and the cursives I, 33, 209 have
μὴ προμεριμνᾶτε alone. The other half of the conflation is provided by
W, which has μὴ προσμελετᾶτε, evidently for προμελετᾶτε. A considerable
number of passages in which the MS. evidence falls into these three groups
have been collected and set out by the Rev. E. A. Hutton (4m Atlas of
Textual Criticism, Cambridge, 1911), but not all these are examples of
conflation.

wf LEAP UAk CRITICISM IN THE) VAST 40g

forms. In Hort’s words, the authors of the revision
“were apparently desirous that the reader should have
the benefit of instructive matter contained in all the
existing texts, provided it did not confuse the context
or introduce seeming contradictions. New omissions
accordingly are rare, and where they occur are usually
found to contribute to apparent simplicity. New inter-
polations on the other hand are abundant, most of them
being due to harmonistic or other assimilation, for-
tunately capricious and incomplete. Both in matter
and in diction the Syrian text is conspicuously a full
text. It delights in pronouns, conjunctions, and ex-
pletives and supplied links of all kinds, as well as in
more considerable additions. As distinguished from
the bold vigour of the ‘ Western’ scribes, and the refined
scholarship of the Alexandrians, the spirit of its own
corrections is at once sensible and feeble. Entirely
blameless on either literary or religious grounds as
regards vulgarised or unworthy diction, yet showing no
marks of either critical or spiritual insight, it presents
the New Testament in a form smooth and attractive,
but appreciably impoverished in sense and force, more
fitted for cursory perusal or recitation than for repeated
and diligent study.” ἢ

As has been indicated above, Hort believed this
revision to have taken place at or about Antioch,
whence he dubs it Syvzaz, in spite of the obvious danger
of confusion with the term Syrzac. He admits freely
that no mention of such a revision occurs in ancient
Christian literature, and does not attempt to assign to
it any specific author, beyond a bare reference to the
possibility of Lucian (whose edition of the Septuagint
was conducted on somewhat similar lines) having had
a hand in it? It is noteworthy also that he regards

1 Hort, Lvtroduction, pp. 134, 135. 4 Ibid. p. 138:

304 CRITICISM OF NEW TESTAMENT cu.

the revision as having taken place in two stages. The
ground for this belief is found in the evidence of the
Peshitto version, which holds a somewhat intermediate
position between the more ancient texts and the fully
developed Syrian revision. Although as a whole it
belongs to the a-family of texts (Hort’s “Syrian ”),
nevertheless in a considerable number of instances it
agrees with the earlier texts against that which we find
in the Antiochene Fathers of the age of Chrysostom.
Hort consequently considers that “(1) the growing
diversity and confusion of Greek texts led to an authori-
tative revision at Antioch, which (2) was then taken
as a standard for a similar authoritative revision of the
Syriac text, and (3) was itself at a later time subjected
to a second authoritative revision, carrying out more
completely the purposes of the first; but that the
Vulgate Syriac [the Peshitto] did not undergo any
corresponding second revision.”’ The final revision
was apparently completed by about 350, while the
earlier stage may be placed about half a century earlier;
but we have no means of fixing it precisely. Once
completed, it rapidly gained universal popularity, no
doubt owing largely to the smoothness, fullness, and
easiness, which are its chief characteristics; and it is
consequently found in an overwhelming majority of the
later uncials and the cursives.

It will be seen, therefore, that the main result of the
Westcott-Hort theory is the total rejection of the great
mass of authorities, and a complete reliance on a rela-
tively small group, composed of the earliest uncials and
versions, with a few later MSS. which preserve the
same type of text. If.a reading is “attested by the
bulk of the later Greek MSS. but not by any of the
uncials sBCDLPORTZ (A in St. Mark) & (also 33) in

1 Hort, Zntroduction, p. 137.

Wii) ΕΑ CRITICISM: IN ‘THE: PAST 305

the Gospels, sABCDE, (also 13, 61 [now 33, 817) in
Acts, 8ABC (also 13 [33]) in the Catholic Epistles, or
NABCD,G, (also 17, 67** [33, 424]) in the Pauline
Epistles, and not by any Latin authority (except the
latest forms of Old Latin), the Old or the Jerusalem
Syriac, or either Egyptian version, and not by any
certain quotation of a Father earlier than 250, there
is the strongest possible presumption that it is dis-
tinctively Syrian, and therefore to be rejected at once
as proved to have a relatively late origin.”' Even this
list of favoured authorities is open to considerable
reductions. The secession of CLPOR 33 in the Gospels,
ACE, 13 [33] in the Acts and Catholic Epistles, or
AC 17 [33] in the Pauline Epistles, would not be held
materially to weaken the presumption in favour of the
more ancient witnesses, all of these MSS. being con-
siderably affected by Syrian influences. Further, when
the Syrian readings have thus been eliminated, we still
have to be on our guard against Western corruptions.
Therefore if D (sometimes in combination with other
uncials, such as xXI', and cursives such as I, 13, 22,
25, 81, 15,7), the. Old) Latin, and Old. Syriac versions,
and the Ante-Nicene Fathers generally (with the partial
exception of those of Alexandria), separate themselves
from the group of authorities above mentioned, they
must be allowed to go, as more than suspect of error ;
in the Acts and Epistles the same family is represented
by DD,E,G, and various cursives, Alexandrian read-
ings are to be found chiefly in CL and the Bohairic
version, reinforced sometimes by xXZ (Matthew) A
(Mark) ΞΕ (Luke) 33, the Sahidic version, and the
Alexandrian Fathers.

The result of all these deductions is to leave B almost

1 Hort, Zxtroduction, p. 163. By 67** he means the marginal readings
of Paul. 67 [now Cod. 424].

Χ

206. CRITICISM ‘OF 'NEW TESTAMENT > > ‘cu:

alone; and Hort does not shrink from this conclusion.
It is better, no doubt, to have the support of other
ancient and trustworthy witnesses. The combination
ΝΒ, in particular, is very strong, since the two MSS. are
sufficiently alike to show that they descend from a
common ancestor, and yet sufficiently unlike to show
that this common ancestor must lie a considerable dis-
tance behind them, and consequently not far from the
autographs themselves. But the superiority of B is such
that no reading attested by it (obvious slips of the pen
of course excepted) can safely be passed over without
the most careful attention ; and in the majority of cases
its evidence must be regarded as decisive.

The learning and conviction displayed in Westcott
and Hort’s work would have secured full attention to
it at any time ; but special prominence and importance
were lent to it by its influence on the Revised Version
of the New Testament. The Revised Version was pub-
lished, it is true, a few months before Westcott and
Hort’s volumes (in May 1881), but a statement of
their theory and its results was communicated to the
Revisers in advance, and the presence of both authors on
the Revision Committee (in which they, with Scrivener,
were by far the most experienced textual scholars)
ensured full attention for their views ; and in point of
fact the new translation displays the effect of their in-
fluence on every page. No doubt the text adopted by
the Revisers differs in many details from that which
appears in Westcott and Hort’s own edition; but the
principle of rejection of authorities of the a-type under-
lies it, and the greatest respect is manifested for the
evidence of » and B, especially the latter. The depar-
ture from the Textus Receptus, the basis of our vene-
rable Authorised Version, was complete ; and the results
of modern textual research, and especially of the theory

Mm Leet UAL CRIEICISM IN DRE Piast 307

of Westcott and Hort, were thus brought forcibly to
the notice of all intelligent readers of the English
Bible.

Naturally so great a shock to established tradition
aroused great opposition, which found at once its most
vehement and its most learned advocate in the person
of J. W. Burgon, Dean of Chichester. Always strenuous
in his resistance of change, Burgon threw himself whole-
heartedly into the championship of the traditional text,
maintaining both that it was intrinsically superior to
that adopted by Westcott and Hort and the Revisers,
and that the fact of its universal acceptance by the
Church was (in view of the Divine institution and
inspiration of the Church) a conclusive proof of its
authenticity. His contributions to textual science during
his lifetime included the examination and collation of
many cursive MSS. of the New Testament, the results
of which were placed at the disposal of Scrivener for
the third edition of his /ztroduction ; the preparation of
a vast index to the quotations from the New Testament
to be found in the Fathers (in MS. only, acquired since
his death by the British Museum) ; an elaborate defence
of the concluding section of St. Mark’s Gospel (The
Last Twelve Verses of the Gospel according to St. Mark
vindicated against recent objectors and established, 1871);
and a vehement, at times even intemperate, assault
on the Revised Version and Westcott and Hort (716
Revision Revised, 1883). He had also contemplated
a more deliberate and constructive work on the textual
criticism of the New Testament, together with a revised
text, exhibiting the true form of the “traditional ” text,
apart from the blemishes which are found in the Textus
Receptus of Stephanus and Elzevir. For both these
works considerable materials were left behind him at
his death, which were subsequently arranged and

308 CRITICISM OF NEW TESTAMENT cu.

supplemented by Prebendary Miller. Two volumes
contain their joint defence of the traditional text, and
a beginning was made with their revised text of the
Gospels." Some examination of the position assumed
in these volumes will be found in the next chapter.
Westcott and Hort’s theory was epoch-making in
the fullest sense of the term. In spite of certain criti-
cisms and modifications, which appear to be well
founded, and of which mention will have to be made
below, this theory holds the field among the scholars
of to-day, and is presupposed as the starting-point of
nearly all the work that is being done in this depart-
ment of New Testament criticism. It is worth noticing
that its main conclusions were reinforced, from a
different standpoint, by the work of the veteran German
scholar, Bernhard Weiss, in a series of studies of the
text of the several portions of the New Testament,
beginning with the Apocalypse and ending with the
Gospels.” These studies took the form of an ex-
amination and classification of the various kinds of
mistakes which characterise the principal MSS. The
result of this examination, which deals with each read-
ing mainly on the ground of internal probability (and
therefore rests to some extent on the “personal equa-
tion” of the critic), is an emphatic verdict in favour of
B, which, though disfigured by many obvious blunders

1 The Traditional Text of the Holy Gospels vindicated and established
(1896); Zhe Causes of the Corruption of the Traditional Text of the Holy
Gospels (1896); A Textual Commentary upon the Holy Gospels, Part I.,
St. Matthew, Division 1, i.—xiv. (1899). The most noteworthy feature of
the last-named work is its copious references to patristic evidence, derived
from Burgon’s above-mentioned index. [The death of Mr. Miller, while
the first edition of this volume was passing through the press, put an end to
this edition, and it is permissible now to say that, in spite of Mr. Miller’s
devotion to the task he had taken up, his scholarship and judgment were
not equal to the demands made on them. ]

2 Published in Gebhardt and Harnack’s Zexte und Untersuchungen,
vii. 1 (the Apocalypse, 1892), viii. 3 (Catholic Epistles, 1892), ix. 3, 4
(Acts, 1893), xiv. 3 (Pauline Epistles, 1896), N.F. iv. 2 (Gospels, 1899).
A text of the Gospels, embodying Weiss’ results, was published in 1900.

ye TEXTUAL CRITICISM IN THE PAST 309

in transcription, shows far less than any other MS.
the signs of deliberate revision or substantial cor-
ruption. Weiss’ work, which takes little account of the
versions or patristic quotations, does not amount to
a complete theory of the textual history of the New
Testament ; but such an examination of documents as
he has conducted forms, as Westcott and Hort them-
selves contend, the essential basis of textual science,
and his conclusion in favour of B, reached as it is rather
through exegetical than transcriptional considerations,
affords a valuable reinforcement to the views of the
Cambridge scholars."

In conclusion it may be useful to mention some of
the more handy critical editions of the New Testament

1 In the Apocalypse Weiss examines the five uncials sACP,Q [=046 in
our list above]. Of these sAC form the earlier group, PQ the later. The
earlier group has many errors and verbal corrections (such as a scribe may
make 672 passant, sometimes involving a misunderstanding of the text), but
shows no signs of systematic revision in editing. The later group, on the
other hand, does show signs of such deliberate editing, especially Q. Of
the individual MSS. A is the best, representing the oldest text most
accurately, and standing alone in about sixty correct readings, while x has
only eight peculiar correct readings, and C four. C (which lacks about a
third of the book) is closely akin to A, but shows some traces of being
affected by the later emended text. This is still more the case with x,
which is much more corrupted than AC, and not infrequently agrees with
PQ. Of the later group, Q has been the most fully emended, but the text
upon which it ultimately rests must have been good, and akin to that of A.

In the Catholic Epistles the text is better preserved than in the Apoca-
lypse. Here the later group is represented by K,L,P,, which show distinct
signs of deliberate emendation, KL somewhat more so than P. xKAC
represent an older text, but still somewhat affected by emendation. B in
general goes with the older group, but differs from it in being practically
free from deliberate emendation. The true reading is preserved in B alone
twenty-four times, never in any of the other MSS. alone; and in combina-
tion with other MSS., while xis right 150 times, A 274 times, and C (which
lacks a quarter of the book) 196 times, B is right 400 times. Nevertheless
B has faults of its own, due to careless copying, and no single MS. can be
trusted implicitly.

In the Acts, the later group is represented by H,L,P,. Of these P has
the fewest peculiar readings, and is probably the purest representative of
the emended text. DE, form a separate sub-group, having many special
variants (especially D, which has 1600 variants as against 440 in E), due
mainly to wilful and thoughtless alteration ; but the deliberate variants are
of the same type as in HLP, so that the basis of the text is the same in
both cases. The earlier group, xAC, is influenced by the emended text (x

310 CRITICISM:OF NEW TESTAMENT οἱ.

for the use of students. The Cambridge Greek Testa-
ment, edited by Scrivener, gives the Textus Receptus,
with a critical apparatus showing the readings of Lach-
mann, Tischendorf, Tregelles, Westcott and Hort, and
the Revised Version, and will be found very serviceable
by those who would rather have the opinions of editors
than the evidence of manuscripts. The Oxford Greek
Testament, on the other hand, which also gave the
Textus Receptus (as printed by Mill) as its main text,
was provided in 1889 with an admirable series of

less so than AC), though not to the same extent as DEHLP. B, on the
other hand, though it has many mistakes of carelessness, shows no sign of
deliberate emendation. It has forty-eight correct readings peculiar to
itself, while x and A have only one each.

In the Pauline Epistles the text is relatively well preserved, and the
groups substantially as in the Acts. The emended text is represented by
K,L,P, (as in Cath.), with the fragment M. The Graeco-Latin MSS.,
ΠΕ, ΕΝ ας (E, being a copy of Dz, and F, and Gg very nearly related), form
a group akin to this, but containing many aberrations peculiar to them-
selves. Both groups derive from an earlier emended text. D in particular
has a good basis, though considerably altered. The older group, xAC, is
also influenced by the emended text, though to a less extent ; but B stands
far ahead of all in purity of text. It shares some twenty-five mistakes
with xAC, about eighty with KLP, and seventy with both groups, these
being mistakes going back to a very early text. On the other hand it has
no less than eighty-five correct readings peculiar to itself, whereas x has
only three and A one.

Finally in the Gospels B again stands out by itself, not for freedom from
mistakes, since it has over 400 errors peculiar to itself, but because its errors
are not due to deliberate emendation. As a rule they are merely scribe’s
blunders ; rarely conformations of the text to suit the context; still more
rarely, conformations with parallel passages elsewhere. All the other
uncials show signs of deliberate revision, but of a superficial kind. A
stands at the head of these emended texts, to which C also belongs ; DLA
are often found with them, DA oftener with A, L oftener with C. D,
however, has 4300 peculiar readings (1700 in Luke, 1150 in Mark, 775 in
Matthew, 655 in John), as against 600in A. These readings are often very
old, but nevertheless false, being due to free handling of the text, and
clearly of a secondary character. x shares 1350 errors with the emended
MSS. (many of them plainly early, as they occur in early versions), and 600
with D, besides having 1350 peculiar to itself, many of which are similar in
character to those in D. But it also shares many of the errors of B, so that
the text represented in B lies also at the base of x. Of the later MSS. L
is the most free from emendation. The younger group (DLA and the frag-
ments RXZ3Z) shares many genuine readings with B, as does the older group
ΝΑῸ ; but B is independent of both, and in 280 places has the right read-

ing alone.

wi ΤΕ CRITICISM IN | THE ‘PAST: 414

appendices by Prof. Sanday, containing (1) a colla-
tion of Westcott and Hort’s text, (2) a select apparatus
criticus, giving all the more important variants, with the
authorities for them, (3) select readings from the Bohairic,
Armenian, and Ethiopic versions, with a brief account of
the MSS. consulted forthem. These appendices can be
obtained in a separate volume, and will be found exceed-
ingly useful by students; the second in particular
provides a most useful small apparatus criticus. A
somewhat similar apparatus is furnished for English
readers in the notes to the excellent Variorum Bible
published by Messrs. Eyre and Spottiswoode, the
New Testament portion of which is edited by Prof.
Sanday, Mr. RK.) 1, Clarke,,and Mr., A. Goodwin.
Those who would prefer to see the results of modern
criticism incorporated in the text itself will find the
texts of Tischendorf and Westcott-Hort in the smaller
editions of these scholars. These, however, represent
only the results arrived at by a single editor or partner-
ship of editors, and it is not likely that either of them
would ever be accepted as a standard text for general
use. The best text for this purpose is probably 7he
Greek Testament with the readings adopted by the Revisers
of the Authorised Verszon, published at Oxford in 1881
under the editorship of Archdeacon Palmer.’ In this
edition the text adopted by the revisers is printed in
full, with a brief apparatus criticus giving the alternative
readings of Stephanus, the Authorised Version of 1611,
and the margin of the Revised Version. In the first
edition of the present work, this edition was strongly
recommended as the best for the use of students,
especially in combination with Prof. Sanday’s critical

1 Simultaneously Dr. Scrivener published at Cambridge an edition of the
Textus Receptus, with the Revisers’ readings in the margin ; but this does

not answer the purpose of providing a revised text of the Greek Testament
for ordinary use.

312 CRITICISM OF NEW TESTAMENT © cu.

appendices, mentioned above. Since then a great step
in advance has been made, the Oxford University
Press having commissioned Prof. A. Souter to equip
the Revisers’ text with a select apparatus criticus.
Prof. Souter has done his work with admirable skill
and zeal, and his edition (Oxford, 1910) is now
incomparably the best for general use. He makes use
of all the uncials (except small fragments), of 11
papyri, of 202 minuscules, and practically all the
versions and Fathers; but only noteworthy variants
are taken into account, so that the textual apparatus
rarely occupies more than a few lines on each page.
The numeration of the MSS. is that of Gregory’s
revised system. The student thus has a text which,
though not perfect, rests upon a much sounder basis
than the text of Stephanus and Elzevir, while it yet
does not represent solely the views of any one critic ;
and a textual apparatus which gives all the variants
which for most purposes are of any importance. It is
much to be hoped that this edition will be generally
adopted in schools and colleges in place of the Textus
Receptus.

As an alternative, mention may be made of the
edition produced by Dr. E. Nestle of Maulbronn
(Stuttgart, 1898), a scholar who has done invaluable
work on the text of the Old Testament as well as on
that of the New. His text (in its original form) is
based upon the texts of Tischendorf and Westcott- Hort,
and upon that produced by Mr. R. F. Weymouth (Zhe
Resultant Greek Testament, 1886), which is itself the
result of a comparison of the texts of Stephanus, Lach-
mann, Tregelles, Tischendorf, Lightfoot, Ellicott, Alford,
Weiss, the Basle edition of 1880, Westcott and Hort,
and the Revised Version. In later editions (1901,
etc.) Weiss has been substituted for Weymouth. Of

wr InRATOAL. CREMCISM IN THE PAST 411

these three editions, Dr. Nestle follows the verdict
of the majority, placing the reading of the minority
in the margin. Another set of footnotes gives some
of the more remarkable variants contained in MSS.,
especially those of the 6-type of text, but without
stating the evidence for them. When once the rather
complicated system of critical symbols is mastered,
this will be found to be a useful pocket edition of the
Greek Testament, with instructive textual material.
Nestle’s text has since 1904 been adopted by the
British and Foreign Bible Society, with a different
apparatus, giving every variation of any importance
irom! the Textus Recepfus and’ the’ Revisers’ text:
This also makes a convenient volume.

With these alternatives before him, any student of
the New Testament can obtain a far purer and more
authentic text (unless modern textual criticism is wholly
and fundamentally at fault) than that which has been
in possession of the ground for the last three centuries
and a half; and he will also be able to follow intelli-
gently the discussion of the textual problems which
still occupy the attention of Biblical critics.

The latest development of textual theory, that of
von Soden, is described at the end of the following
chapter.

CHAPTER | Vii

THE TEXTUAL PROBLEM

[Authorities : Westcott and Hort, of. cit.; Nestle, of. cz¢.; Burgon and
Miller, Zhe Traditional Text of the Holy Gospels (London, 1896), and
The Causes of the Corruption of the Traditional Text of the Holy Gospels
(London, 1896); Salmon, Some Points in the Textual Criticism of the
New Testament (London, 1897); Sanday and Headlam, 4 Critical and
Exegetical Commentary on the Epistle to the Romans, pp. Ixiii-lxxiv
(Edinburgh, 1896); F. Blass, Acta Apostolorum secundum formam Ro-
manam (Leipzig, 1896); Evangelium Lucae secundum formam Romanam
(Leipzig, 1897), and The Philology of the Gospels (London, 1898) ;
Weiss, of. czt., and Der Codex D in der Apostelgeschichte (Leipzig,
1897); K. Lake, The Text of the New Testament, pp. 64-91 (London,
1900); C. H. Turner, Historical Introduction to the Textual Criticism
of the New Testament, in Journal of Theological Studies, 1908-10; H.
von Soden, Die Schriften des Neuen Testaments (Berlin, 1902-10). }

THE data of New Testament textual criticism have
now been set forth,—on the one hand the materials
with which the critic has to deal, and on the other the
various ways in which critics in the past have dealt
with them. Evidence has been collected with great
assiduity for more than two hundred years, and highly
trained and gifted scholars have applied themselves to
the interpretation of it; yet the whole problem is not
solved, and students are by no means agreed, even
upon points of fundamental importance. It is not the
office of a handbook such as this to advance any new
solution, or to aspire to make any noteworthy addition

314

CH. VIII FHE TEXTUAL PROBLEM 418

to textual theory. It is probable that whatever
advance is made in the immediate future will take the
form of elucidation of special and particular points
rather than of a general restatement of the whole
subject ; but a survey and knowledge of the whole field
is necessary as a foundation for such special work. The
object, therefore, of this concluding chapter is the state-
ment of the textual problem as it confronts the New
Testament critic to-day, and the indication of the lines
along which progress is most to be desired and most
to be expected.

With this view we shall examine in succession the
claims and character of the several types of text which
the preceding chapters have shown to be presented by
our textual authorities.

§ ΤΙ. The a-text

It will have been seen from the last chapter that
the uniform tendency of modern criticism has been to
discredit and dethrone that type of text which has
held possession of our Bibles since the invention of
printing,—nay, if we go back to manuscript books, for
a thousand years before that date,— and which is
consequently known as the Textus Receptus or the
Traditional Text. Commonly accepted, however,
though this doctrine is among scholars, the realisation
of it is hardly yet popular and general; and the
reasons for it need to be set forth for the instruction of
students. It will therefore be best, in the first place,
in order to clear the ground, to re-consider the claims
of the Traditional Text, as stated by its last whole-
hearted advocates. These are Dean Burgon and his
continuator, Mr. E. Miller, and in their criticisms of
Westcott and Hort’s theory may be found the fullest

316 CRITICISM OF NEW TESTAMENT cu.

statement of the claims of the Textus Receptus, and
the most trenchant attack on the views which, since
their promulgation by Westcott and Hort, have been
either accepted, or adopted as the basis for yet more
advanced views, by nearly all textual scholars.

The propositions upon which Burgon and Miller
based their defence of the Traditional Text, reduced to
their simplest form, are two in number: (1) that the
universal acceptance of it by the Church from the
fourth century to the nineteenth is in itself proof of its
superiority, since the Church must have been Divinely
guided in its dealings with the sacred Word of God;
(2) that, apart from such considerations, it can be
shown to be both older and intrinsically better than its
rival, which they call the “neologian” text.1 Now if
the first of these propositions is true, fiuzta est quaestto ;
for Hort admits, no less than Burgon claims, that the
Traditional or Received Text has been at first pro-
minent and ultimately dominant in the Church from at
least the end of the fourth century. There is, more-
over, this much at least to be said in favour of sucha
contention, that God, Who instituted the Church to be
the guardian and teacher of His Word, would surely
not have allowed that Word to be propagated in a
corrupt or seriously mutilated form. Further still,
there is the analogy of the establishment of the doctrine
of the Church, which received its final formularisation
in the course of this same fourth century ; and much
the same may be said with regard to the determination
of the Canon of the New Testament. As, therefore,
we believe that the Church was Divinely guided in its
assertion of the doctrine of Nicaea and Constantinople,
and in its choice of the books which it regards as

1 By this term Burgon presumably meant to associate modern textual
criticism with the rationalistic interpretation of the Bible, to which the
same term was applied in the eighteenth century.

VIII THE ΤΕ PROBLEM 419

especially inspired, are we not also called upon to
believe that it was Divinely guided in its choice of the
text of these books, and that the type which it selected
must be regarded as thereby stamped with the Divine
approval ?

The analogy, however, with the establishment of
doctrine and the canon in the fourth century (or about
that date) is very imperfect. It is not contended that
any oecumenical council selected the Traditional Text
for universal use, or even so much as considered the
question of an authorised text at all. The doctrines
of the Church were established by councils, and by
councils only; the limits of the canon, if not so wholly
dependent on the decisions of councils, were yet con-
sidered and ratified by them; but the text of the
sacred books never formed the subject of their delibera-
tions. The selection of the traditional type of text by
the Church was gradual and informal, and therefore
cannot claim the sanction of a deliberate decree. Nor
is the argument that God would certainly secure the
preservation of the true form of His Word much more
pertinent. We may indeed believe that He would not
allow His Word to be seriously corrupted, or any part
of it essential to man’s salvation to be lost or obscured ;

not one of doctrine. No fundamental point of doctrine

rests upon a disputed reading: and the truths of
Christianity are as certainly expressed in the text_of

Westcott and Hort as in that of Stephanus.’

1 Burgon and Miller did indeed at times accuse the codices x and B of
being tainted with sceptical tendencies, and especially with minimising the
Divinity of our Lord ; but the evidence adduced in support of this charge
is wholly inadequate. No doubt the traditional text contains many more
phrases in which the Divinity is implied,—the natural amplifications of
scribes writing after the phraseology of the Church had become more fixed;
but the language of x and B as they stand is wholly inconsistent with such
supposed heretical revision.

318 CRITICISM/Or NEW TESTAMENT | | ce

It is, moreover, a perversion of the facts of history to
speak of the text of the Scriptures as preserved in a
uniform shape from the fourth century to the present
day, as the argument of Burgon requires. While the
substance of the sacred text, and its general type, have
been so preserved, a very great amount of variation in
detail has been admitted. The manuscripts of the
Greek Testament differ very considerably from one
another. The manuscripts of the Vulgate differ from
those of the Greek Testament, and have suffered even
more corruption among themselves. We have seen in
an earlier chapter how the history of the Vulgate text
is one of widespread depravation and of repeated
attempts at restoration. The Syriac and Coptic texts,
again, differ in many particulars from both Greek and
Latin. Still more great and deep-seated are the differ-
ences in the Text’ of the Old Testament. The textier
the Septuagint, which was and is the Bible of the Greek-
speaking Churches, differs widely from the Massoretic
Hebrew. In short, the first of Burgon’s main proposi-
tions is neither convincing @ 271072 nor in fact reconcil-
able with history. History makes it clear that God in
His wisdom has permitted great deviations in the tradi-
tion of the sacred text through the frailty of its human
trustees, though always so that its substance was not
lost or seriously endangered.

Dismissing, then, the ὦ przort argument that the
Church would certainly be Divinely guided in her
choice of a text, we are forced to deal with the
problem in accordance with the established principles
of textual science. Here too Burgon and Miller claim
a verdict, and that principally on the ground of the
enormous numerical preponderance of witnesses in
their favour. Again and again they contrast the
hundreds of manuscripts found upon the one side

VIII THE TEXTUAL PROBLEM 310

with the mere handful which is opposed to them,
and to which modern editors have almost unanimously
pinned their faith. “Is it likely,’ says Burgon,’ “is
it in any way credible, that we can be warranted in
rejecting the testimony of (suppose) 1490 ancient wit-
nesses, in favour of the testimony borne by (suppose)
ten?” “What,” asks Mr. Miller,? “would an editor
of Sophocles do under such circumstances?” The
answer to this query is simple. He would do pre-
cisely as Hort and the majority of editors have done.
There are about 104 MSS. of Sophocles; yet the
evidence of a very large majority of these is wholly
disregarded by all editors. One manuscript (L, in
the Laurentian Library at Florence) is of predominant
authority ; two others (A and I’) are of considerable
value ; the rest have little independent worth, but only
support one or other of the leaders, or diverge into
palpable error. The weight attached by all editors to
LAT in comparison with the remaining 1o1 MSS. is
even greater than that which most modern editors
attach to 8BDLT and the early versions in the Gospels.
So too with Virgil; out of the hundreds of existing
MSS., even Henry (who devoted far greater pains to
the collection of evidence than any other editor of the
poet) only quotes the seven great MSS. and some
seventy minor MSS. (and these last only in numerical
sroups, not as individuals). So, in fact, with every
other classical author; in every case where any con-
siderable number of MSS. exists, it is found that

1 Traditional Text, p. 45.

2 The Oxford Debate on Textual Criticism, p. 6: ‘‘Suppose you are
sitting at the elbow of an editor of Agamemnon, or the Zrachinzae, or
whatever it may be of Sophocles, you would see that in his very wildest
dreams he would never conceive on any difficult passage of such an
immense mass of evidence being at hand as we have in this case on the one
side set aside by those few.” The form of expression is odd, but the
intention is clear.

320 CRITICISM OF NEW TESTAMENT ὁπ

nearly everything depends upon a few leading authori-
ties, all the rest being relegated to the background and
consulted only under special circumstances.

When, therefore, Burgon and Miller condemn the
modern editors of the New Testament, from Lachmann
to Hort, for their preference of a few generally early
MSS. and versions to the great mass of later authori-
ties, they are in fact impugning the universally accepted
principles of textual criticism. The earliest printed
texts of the classical authors were in nearly all cases
based upon comparatively late manuscripts, because
these were the most numerous and accessible at the
time ; but scientific criticism has uniformly shown that
the texts so obtained are unsound, and that recourse
must be had to a select group of a few authorities,
generally those of earliest date. In some instances
a single MS. is held to outweigh all its rivals, except
where it is manifestly corrupt. The Laurentian MSS.
of Aeschylus, Sophocles, and Aristotle’s Eihzcs, the
Paris MSS. of Plato’s Republic and of Demosthenes,
the Urbinas of Isocrates, enjoy a pre-eminence over all
other authorities in their respective spheres, which even
Hort would hardly claim for the Codex Vaticanus, In
short, what Burgon and Miller persist in regarding as a
paradox is in fact a commonplace of textual criticism.

Of course it is still possible for the advocates of the
Traditional Text to maintain that the case of the New
Testament is unlike that of all other books, and that
Lachmann and his followers have erred in the group of
witnesses which they have selected as the best. No
doubt 8 and B are older than any other MSS. which
we possess, and as a rule the earlier authorities are
ranged upon the same side; but age, though it raises
a presumption in favour of superior accuracy, is not
decisive, and there would be nothing prima facte con-

VIII THE TEXTUAL PROBLEM 321

trary to sound textual criticism in preferring A and C,
or N and ®, to x and B as standards of the true text.
The comparison between these types of text must be
made upon their merits, and any one is perfectly within
his rights in defending the Traditional Text as zuirznsz-
cally superior to its rival. Here the question of relative
originality comes in. So long as opposing critics con-
tent themselves with asserting their preference for this
reading or for that, on grounds of internal fitness, little
progress can be made. The personal equation is too
hard to allow for in such a controversy. But if it can
be shown that one type of text goes back to an earlier
date than another, or represents a primary as opposed to
a secondary stage of development, then the presumption
is very strong in favour of the text so guaranteed.

It is on this crucial point of the controversy that
the patristic evidence becomes of decisive value. Hort,
as we have seen, appeals to it as showing that the
Traditional Text is characterised by many readings
which cannot be traced back farther than the fourth
century,—readings which, moreover, have in his eyes
the appearance of a secondary character, as derived
from pre-existent readings which are found in the other
groups of authorities. Here is a plain issue. If it can
be shown that the readings which Hort calls “ Syrian ”
existed before the end of the fourth century, the key-
stone would be knocked out of the fabric of his theory ;
and since he produced no statistics in proof of his
assertion, his opponents were perfectly at liberty to
challenge it. It must be admitted that Mr. Miller did
not shirk the test. A considerable part of his work as
editor of Dean Burgon’s papers took the form of a
classification of patristic quotations, based upon the
great indices which the Dean left behind him,’ according

1 See above, p. 243, note.
δ

322 CRITICISM) OF ΝΕ TESTAMENT | cx:

as they testify for or against the Traditional Text of
the Gospels. |

The results of his examination are stated by him as
follows." Taking the Greek and Latin (not the Syriac)
Fathers who died before A.D. 400, their quotations are
found to support the Traditional Text in 2630 instances,
the “neologian” in 1753. Nor is this majority due
solely to the writers who belong to the end of this
period. On the contrary, if only the earliest writers be
taken,from Clement of Rome to Irenaeus and Hippolytus,
the majority in favour of the Traditional Text is pro-
portionately even greater, 151 to 84. Only in the
Western and Alexandrian writers do we find approximate
equality of votes on either side. Further, if a select
list of thirty important passages be taken for detailed
examination, the preponderance of early patristic evi-
dence in favour of the Traditional Text is seen to be
no less than 5 30 to 170, a quite overwhelming majority.

Now it is clear that if these figures were trustworthy,
there would be an end to Hort’s theory, for its premises
would be shown to be thoroughly unsound. An ex-
amination of them, however, shows that they cannot be
accepted as representing in any way the true state of
the case. In the first place, it is fairly certain that
critical editions of the several Fathers, if such existed,
would show that in many cases the quotations have
been assimilated in later MSS. to the Traditional Text,
whereas in the earlier they agree rather with the
“Neutral” or “Western” witnesses. For this defect,
however, Mr. Miller cannot be held responsible. The
critical editions of the Greek and Latin Fathers, now
in course of production by the Academies of Berlin and
Vienna, had covered very little of the ground at the

1 Traditional Text, pp. 94-122. The examination was confined to the
Gospels, the textual problem being both harder and more important in these
books.

VIII THE: TEXTUAL: PROBLEM 323

time when his materials were compiled, and meanwhile
he might legitimately use the materials accessible to
him; and the errors arising from this source would
hardly affect the general result to any very serious
extent. The real fallacy in his statistics is different,
and is revealed in the detailed examination of the thirty
select passages. From these it is clear that he wholly
misunderstood Hort’s contention. The thirty “ tradi-
tional” readings, which he shows to be so overwhelmingly
vindicated by the Fathers, are not what Hort would
call pure “Syrian” readings at all. In nearly every
case they have Western or Neutral attestation in addition
to that of the later authorities. Thus the insertion of
Matthew xvii. 21 is supported by DL and the Old
Latin version; Matthew xviii. 11 by D, the Old Latin
and Curetonian Syriac; ἀγαθέ in Matthew xix. 16 by
the Old Latin, Curetonian and Sinaitic Syriac, Bohairic
and Sahidic ; ἔρημος in Matthew xxiii. 38 by xD, the
Old Latin, and most Coptic MSS.; the last twelve
verses of δε, Mark; by 19, the’ Old) Latin (except 2),
Curetonian Syriac, and most Bohairic MSS.; Luke
xxivi 40. by NBL, the: Bohairic; etc: ; John, xxi.25
by every authority except x, and every editor except
Tischendorf. In short, Mr. Miller evidently reckoned
on his side every reading which occurs in the Traditional
Text, regardless of whether, on Hort’s principles, they
are old readings which kept their place in the Syrian
revision, or secondary readings which were then intro-
duced for the first time. According to Hort, the
Traditional Text is the result of a revision in which
old elements were incorporated ; and Mr. Miller merely
points to some of these old elements, and argues there-
from that the whole is old. It is clear that by such
argumentation Hort’s theory is untouched.

So far, then, as the central point of Hort’s theory

324° CRITICISM (OF NEW TESTAMENT cx

is concerned, namely, the secondary nature of the
Traditional Text, it has stood the test of thirty years’
criticism, and is now taken for granted by most scholars.
The discoveries which have been made since the theory
was put forth, such as the Sinaitic Syriac version and
the Diatessaron, have fallen into line precisely as Hort
would have wished, and have supplied a most valuable
test, because one which Hort could not have reckoned
upon when writing his /ztroduction. The more the
evidence as to the earliest texts of the New Testament
is examined (and much has been done in this direction
in the last thirty years), the more certain does it appear
that the type of text to which we are accustomed did
not come into existence until the fourth century. The
texts in use before that date show great variety and
fluctuation; but the characteristic features of the
“Syrian” text are not yet visible.

There is, however, still room for question as to the
manner in which the a-text came into existence. Hort
holds that it is the result of deliberate revision: “ The
Syrian text must in fact be the result of a ‘recension’
in the proper sense of the word, a work of attempted
criticism, performed deliberately by editors and not
merely by scribes” ;* and he divides it into two stages,
in order to account for the phenomena presented by the
Peshitto version, which seems to offer the “ traditional ”
text in a somewhat earlier form than the majority of
Greek MSS. In connexion with the earlier stage, he
mentions tentatively the name of Lucian [ob. A.D. 311],
whom we know to have been the author of a revision of
the Septuagint conducted upon similar lines ; but there
is no direct evidence to associate him with the New
Testament, and Hort does not press the suggestion.
Indeed the absence of evidence points the other way ;

1 [ntroduction, p. 133.

VIII THE: TEXTUAL PROBLEM 325

for it would be very strange, if Lucian had really edited
both Testaments, that only his work on the Old Testa-
ment should be mentioned in after times. The same
argument tells against any theory of a deliberate re-
vision at any definite moment. We know the names
of several revisers of the Septuagint and the Vulgate,
and it would be strange if historians and Church writers
had all omitted to record or mention such an event as
the deliberate revision of the New Testament in its
original Greek. It seems probable, therefore, that the
Syrian revision was rather the result of a tendency
spread over a considerable period of time than of a
definite and authoritative revision or revisions, such as
produced our English Authorised and Revised Versions.
We have only to suppose the principle to be established
in Christian circles in and about Antioch, that in the
case of divergent readings being found in the texts
copied, it was better to combine both than to. omit
either, and that obscurities and roughnesses of diction
should be smoothed away as much as possible. Such
a principle is a natural one in an uncritical age, and
this hypothesis accounts not only for the absence of
specific reference to a revision, but also for the Peshitto
evidence above mentioned. The process would no doubt
be’ assisted) and accelerated; if, as’ Dr,’ Salmon Was
suggested,’ the texts current in any district depended
largely upon the bishop or clergy who regulated the
lessons to be read in church, and who could thereby
familiarise the congregation with the type of text pre-
ferred by them. The point is that the Syrian revision
was a long-continued process, not a single act. Nor is
it clear that Hort meant much otherwise, though on a
first reading his words convey the impression that he
did. He speaks of deliberate criticism, of the work of

1 Some Points in the Textual Criticism of the New Testament, pp. 77-79.

326 CRITICISM OF NEW TESTAMENT αἵ

editors as opposed to scribes, and he refers to two
stages in the work separated by an interval of time.
But he does not say that these two stages were the
result of two definite and authoritative revisions ; his
words need mean no more than that we happen to have
evidence, in the Peshitto and the Greek MSS., which
shows us the extent to which the continuous process
had gone at two particular moments. At any rate it
involves very little modification of Hort’s theory to treat
it in this way ; and this modification is now commonly
made by students of textual criticism.

With regard to the authorities in which the a-text
is to be found, it will have been seen from the pre-
ceding discussion that no full enumeration is possible.
The “ Syrian” or “traditional” text may be presumed
to be found in any MS. of which the contrary is not
stated. In the Gospels the list of authorities of this
class is generally headed by A and C, though both
are free from many later corruptions. The purple
MSS., N=®, represent a further advance in the tradi-
tional direction ; but the most typical members of the
a-text are the late uncials EFKMSUII. With these go
the great mass of the minuscules, and the later Fathers.
Of the versions, the Peshitto generally belongs to this
family, but (as has been stated above) to a relatively
early stage in its development; while all the later
versions, and most late MSS. of the early versions, are
more or less affected by its influence. Readings
attested only by the authorities here enumerated may
almost certainly be regarded as “Syrian.” In the
Acts and Catholic Epistles, A and C cease to belong
predominantly to this type, and the other uncials
above mentioned do not contain these books. The
leading representatives of the a-type here are H,K,
(the Catholic Epistles only) L,P,, with the large

ὙΠ] THE TEXTUAL PROBLEM 327

majority of minuscules, and the versions and Fathers
as before. For the Pauline Epistles the available
uncials are K,L,P,; for the Apocalypse 046 P.,
Small fragments and MSS. which have not yet been
adequately examined (such as Q 047 049) are not
taken into this reckoning, though the age of these
three uncials makes it probable that they belong to
the same class, Further details as to the text of the
uncials of the a-type will be found in the treatises of
Weiss, summarised on pp. 309-10 above; but a real
attempt to classify the whole mass of MSS., uncial and
minuscule alike, has for the first time been made by
von Soden. Of this more will be said below.

§ 2. The B-text

Acceptance of Hort’s theory of the secondary
nature of the a-text does not, however, necessarily
involve acceptance of his views with regard to the
-text, to which he gives the designation of “ Neutral” ;
and the character of this type of text must be separ-
ately examined. Hort finds this text pre-eminently in
the Codex Vaticanus (B); indeed one may almost
define his “Neutral,” text. as the text of B (scribe's
blunders excluded), and of other MSS. so far as they
agree with B, The most notable exception to this
rule occurs in the case of certain verses (mostly in the
latter chapters of St. Luke) which are found in B and
the great majority of MSS., but are omitted in D and
other authorities of the d-class. These verses Hort
believes not to have formed part of the original
text, and designates as “ Western non-interpolations.”
Except in this special case, the authority~of B is
predominant with him; if with other support, well
and good (unless the company is very suspicious), but
if, alone, still, it, cannot \,safely) be... rejected,» The

bt

Ἶ

———

es

328 CRITICISM (OF NEW TESTAMENT. (cu:

authority found oftenest among its allies is δ, and
these two MSS. must have had some common ancestor
short of the original autographs; but the differences

/ between them are sufficiently numerous and important

to show that this common ancestor was not a very
near one. Other authorities, of which more will be
said below, add their testimony from time to time to
the 8-text ; but the predominant element in it is always
the Codex Vaticanus.

Now this predominance assigned to one manuscript
among thousands (though we have shown parallels to
it on a smaller scale in the case of many classical
authors) is very striking, and has naturally been fixed
on as a point of attack by the opponents of Hort’s
system. The independent student also may well
hesitate before he admits it. It is therefore important
to notice that Hort does not stand alone in his
preference for B. Weiss, who will have nothing to
do with Hort’s classification, or with any far-reaching
classification, of authorities, is not less positive in his
exaltation of B above all other MSS. As has been
stated above (pp. 309-10) he regards it as the only MS.
of the New Testament which has escaped deliberate
revision, and estimates that in no less than 437 places
it has the true reading alone (alone, that is, among
the uncials, to which his examination is confined).
This result of a wholly independent examination,
coincident as it is with the general judgment of textual
scholars in the last century, may go far towards
reconciling the student to the idea of this marked
superiority on the part of one MS. among so many.
But it also increases the necessity of considering care-
fully the history of this MS., and the extent to which
this admission of its excellence carries us.

What, “in ‘short, is ‘the, @-text' (zz, (the text of Ὁ,

VIII THE TEXTUAL PROBLEM 329

purged of its obvious errors and amended in the
comparatively few places in which it seems certainly
inferior to some other authority)? Is it a close
approximation to the original text of the New Testa-
ment as it left its authors’ hands, or is it merely one
among several local texts which the chances of time
have brought déwnto-tis? In Hort’s eyes it is the
former, and Weiss appears substantially to agree with
him. Salmon, on the other hand, considers that it
cannot be shown to be more than an Alexandrian text
of good character, traceable perhaps to the second
century, but not to be carried with any certainty to
an.earlier period than that." He would grant it a
high character for truth, but would not allow it a
monopoly of that virtue. The 6-text, for example,
though no doubt less trustworthy on the whole, may
yet often preserve the true reading when B and its
fellows have gone astray. Its allies, it will be observed,
are mainly MSS. and versions connected with Egypt,
and the Fathers who confirm it most often are Origen
and his followers in Egypt. Is it not therefore prob-
ably a local text,—the text of Egypt, as the 6-text is
the text (according to Salmon) of Rome,—preferable
no doubt as a rule, but not invariably so?

The problem would be nearer solution if we could
determine with any certainty the place in which B was
written, There was a strong tendency, as has been
indicated above (p. 82), to refer it to the library at
Caesarea, founded by Origen’s disciples, Eusebius and
Pamphilus ; but, as has there been shown, the evidence
is far from conclusive. If, however, the statement is
made a little wider, and B and ws are connected with
the Origenian school of textual criticism, whether in
Alexandria or in Caesarea, the evidence in support of

1 Textual Criticism of the New Testament, p. 52 ff.

330 CRITICISM OF NEW TESTAMENT cu.

it is more adequate. Directly or indirectly, then, it
would appear that we must look to Egypt for the
origin of the @-text, of which these MSS. are the
principal representatives. So far, in fact, the view of
Salmon, that it is an Alexandrian text of good char-
acter, would seem to be justified ; but it is less easy to
follow him when he proceeds to treat it as more or
less equally balanced in authority by the “ Western”
text, which he localises in Rome. It is not sufficient
to assign each type of text to a certain locality. The
credentials of the localities must also be examined, as
well as the character of the text associated with them.
Now, in the first place, we have no sufficient proof
that the Western text took its rise in Rome at all; on
the contrary, as will be shown below, the available
evidence connects it rather with other localities of less
imposing station in the ecclesiastical world. Next,
even if the association of the 6-type of text with Rome
be admitted, it gives little or no guarantee for the
quality of the text. Rome had no traditions of textual
scholarship in regard to Greek literature, and scholarly
accuracy is not the predominant characteristic of the
early Fathers of the Western Church. Alexandria, on
the other hand, was the home of textual criticism
and of minute scholarship. There questions of text
would be carefully considered and scientifically decided.
The traditions of heathen scholarship could not but
affect the manner in which the text of the Scriptures was
treated, especially from the moment when Christianity
was recognised by the state. When, then, we say that
the 8-text is an Alexandrian text of good quality,
representing, if not the text as edited by Origen, at
least the kind of text which he selected as the basis of
his labours (and it will be remembered that this is the
conclusion to which the study of the Old Testament text

VIII THE TEA TUAL: PROBLEM 331

leads us, see p. 83, above), we are giving it a very high/

claim to authority. Such a text would not be immacu-
late ; it might have suffered something from the ordinary
risks of transmission; it might have suffered from
deliberate alterations of a too pedantic critic; but it
would probably have been based on an intelligent
comparison of authorities, conducted by a scholar or
scholars accustomed to the scientific criticism of texts.’
Admitting, therefore, Salmon’s contention that the
8-text cannot be shown to be more than an Alexandrian
text, it is still possible to go far in the direction of
giving it the position of supreme authority which is
claimed for it by Hort and Weiss. A 2γοζζξ, such a
text has more chance of accuracy than one produced
in less critical surroundings; and a jposterzort it is
found to have a very high proportion of readings which
textual science pronounces to be authentic. No doubt
we reserve the right to revise the verdict of Alexandria
in cases where we have evidence of the existence of
.other very ancient readings. The hypothetical Alex-
andrian critics, as will be shown later in treating of the
6-text, must have had before them texts of the type
which we now call “ Western”; and we may use our
own judgment as to whether they were always right in
their rejection of them. At the same time we shall
do well to attach considerable weight to the fact that
they did so reject them; for they are not likely to
have done so without reason, and they stood very
1 Of late years it has become not unusual to attribute the Alexandrian
text to Hesychius, just as the Syrian has been attributed to Lucian (see
p- 324). It is known, on the authority of Jerome, that Hesychius and
Lucian produced editions of the Septuagint, the former in Alexandria
and the latter in Syria; but the only evidence that they did the same for
the New Testament is a reference in Jerome’s Epistle to Damasus (prefixed
to the Vulgate N.T.) to certain MSS. which passed under the names of
Hesychius and Lucian, and which he considered to be wholly unreliable.

Bousset, however, has developed the theory that B represents the recension
of Hesychius, and von Soden follows him (see below, p. 365).

<—

332 CRITICISM OF NEW TESTAMENT οἱ

much nearer to the original autographs than we do.
It is also important to remember that, in the judgment
of such a scholar as Weiss, whose life has been spent
in the study of the New Testament, the readings of
B have not the character of a deliberate revision, but
rather of a tradition generally faithful, though marred
by superficial blunders in its later stages; while the
a-text presents throughout the phenomena of deliberate
change. In the #-text, therefore, we may see the
results of conscientious protection of the sacred text ;
in the a-text, the results of injudicious and unscientific
editing ; while the 6-text, of which we shall have to
speak later, if it is free from editorial handling, has
also escaped the protection of a literary conscience and
the environment of textual scholarship.

With regard to the authorities for the -text, it
will have been seen that B is by far the most important,
and next to it is x. Nearest to these come, in the
Gospels, LRTZE (and A in Mark), with the more
mixed texts of PQ and W (especially in Mark), and
with support occasionally from A and C, when these
have escaped the emendation characteristic of the
a-text. Among the minuscules the greatest approxi-
mation to the f-type is shown by 33, the group
f-1118-18 1-200; $9, 17, 421, 496, 802), but anyon
these have a greater or less degree of admixture with
other elements. The whole group headed by A (see
p. 118), which claims connexion with Jerusalem, may
also be reckoned as having some kinship with this type
of text. Of the versions the Bohairic is the leading
representative of this text; but there are considerable
traces of it in the Sahidic. Jerome’s revision of the
Old Latin was also based on MSS. of this type; but
the Vulgate (being a revision of a version of the
6-type, considerably contaminated with the a-type, by

VIII THE TEXTUAL PROBLEM 333

the help of MSS. of the @-type) is of too mixed a
character to be reckoned wholly with any one family.
Finally, of the Fathers Origen is by far the most
important witness to the B-type, though he also exhibits
the 6-type in some of his works. Clement of Alexandria
belongs more decidedly to the 6-family, but is also
found, supporting: the p-text.) Indeed it) must be
remembered throughout that the 8 and ὃ authorities
are often found in alliance, such readings being pre-
sumably parts of the very ancient text common to
both, and therefore possessing an almost irresistible
claim on our acceptance.

In the Acts and Catholic Epistles the authorities
for the 8-text are substantially the same, after deducting
those which only contain the Gospels. A and C,
however, are found oftener in accord with xB. To
these must be added the minuscule 81, and to a less
extent 69 and the Euthalian group 88, 181, 203, 307,
1836, 1845, 1846, 1875.

In the Pauline Epistles’ #B are again the principal
representatives of the §-text, but B has not here quite
the same predominance as elsewhere, and is found not
infrequently supporting variants characteristic of the
6-text. On the other hand, AC definitely part company
with the a-text, and must be reckoned with either the
8 or y-text; so that where B joins the 6-text, ΝΑῸ
may be taken to stand for the §-text. L,M, and
sometimes P, (which has a better text here than in
Acts and Cath.) are also found in the same group.
Further, for most of the Pauline Epistles we have the
evidence of H,, representing the text of Pamphilus in
the library of Caesarea (see p. 105); and with this is
associated the third corrector of x, known 85 8, who

1 On the textual criticism of the Pauline Epistles see Sanday and
Headlam, Romans, pp. Ixili-]xxiv.

334 CRITICISM OF NEW TESTAMENT cu.

similarly claims the authority of Pamphilus (see p. 65),
and whose readings are found to agree generally with
those of H,. If the connexion of the Armenian version
with the text of Euthalius can be made out, it likewise
will have to be added to this group. Among the
CUESIVES αν 34/190, \ 104; 200; 424"") 4321: seem it
deserve most notice, together with the Euthalian 88,
1914, 1916, 1962, 1970, and the upper writing of P.,.

In the Apocalypse B fails us, and the @-text is
represented by RAC, but especially by A. The cursives
which approximate most to this type are 1", 104, 172,
2018) 2018, 2020, 2032, 2030) 2640, 204i:

§ 3. The y-text

Ii, then, the origin »of the) 8-text.| be: traced. te
Alexandria, what is to become of the y-text, which
Hert calls , “ Alexandrian’)? “Has. it any (feal/ine
dependent existence? There is no doubt that the
readings which Hort thus designates occur in MSS. of
Egyptian origin; but it may be questioned whether
they represent any distinct tradition, especially when it
is remembered that this type of text occurs in no single
MS. throughout, but is embodied in isolated readings
in MSS. which otherwise are of “ Neutral” or “Syrian”
or “ Western” character. Dr. Salmon goes so far as
to suggest that “ Alexandrian” readings are merely the
residuum of the Egyptian text which does not happen
to be found in B; and clearly our view of the y-text
depends very largely on our view of the character of
B and the A-text. If that is due to the revision of an
Alexandrian editor or editors, then the y-text is simply
a congeries of readings current in Egypt which did not
commend themselves to that editor or editors ; while if
the B-text, though preserved mainly in Egypt, owes

little or nothing to editorial revision, but is substantially
a pure or “neutral” fount of text, then the y-text may
embody the results of editorial revision in Alexandria.
The last is, of course, Hort’s view, and he would regard
the readings in question as due to the scholarly pre-
possessions of Alexandrian scribes and editors, who
introduced verbal alterations into the sacred text in
accordance with their ideas of correctness and style.
The tendency of recent discoveries is rather to con-
firm the existence of a separate type of text in Egypt,
distinct from that of B. The extant fragments of the
Sahidic version have greatly increased in number since
Hort wrote, and the Middle Egyptian version is
beginning to take concrete shape. In both of these
we seem to see traces of an Egyptian family of text
distinct from that of B. An example may be found in
a bilingual fragment in the British Museum, containing
John iii, 5-iv. 18, iv. 23-35, 45-49, in Greek and
Middle Egyptian, of the sixth century.’ The affinities
of this MS. are emphatically with its fellow bilingual
083 and with L, and with A and C in their non-Syrian
readings, but very much less with B, and decidedly not
mia Dor is~) It is).a text, meither, “ Western.” ‘ner
“Syrian,” and yet not precisely the text of B, though
agreeing with it in the more important variants. Thus
it possesses precisely the characteristics which Hort
would assign to his “ Alexandrian” text, though (as it
comes from Central Egypt and has no demonstrable
connexion with Alexandria) it might perhaps more
properly be called “ Egyptian”; and if we had the
whole MS. we might apparently possess a complete
representative of this type. At any rate it tends to
establish the separate existence of such a type of text ;

1 Published by Crum and Kenyon in /ourn. of Theol. Studies, i.
415-433 (1900).

3360 CRITICISM: OF NEW TESTAMENT. | cm

and all that remains to do is to consider its relations
with the #-text.

After all, there is no insuperable difficulty in
imagining the existence of two or more types of text in
the same country. Indeed this must have been the
case in Egypt, since, as will be shown below, there is
good evidence that in very early times the 6-text was
current in Egypt as elsewhere. Its subsequent dis-
appearance may be ascribed to the higher level of
textual scholarship in that country; but it does not
follow that it left only one uniform text behind. If
the §-text is due to a highly enlightened criticism,
which generally succeeded in selecting authentic read-
ings without a large admixture of editorial revision, the
y-text may be due to a somewhat less successful attempt
in the same direction, or to subsequent modifications of
that rescued text. A somewhat similar phenomenon
is found in the case of the Old Testament Septuagint
text. Here it is known that an edition was prepared
in Egypt by Hesychius, and the question is in what
sroup of manuscripts to look for this edition. Ceriani
finds it in the Codex Alexandrinus (A), the original
text of the Marchalianus (Q), and in certain cursives ;
while Cornill refers it to another group of cursives and
the Coptic versions, akin to the MSS. just mentioned,
but presenting in his view more of the character of a
formal and authorised edition. It is immaterial for
our present purpose to discuss which of the two is
right; all that is important is to observe that we have
here two groups of MSS., both connected with Egypt
and showing clear signs of affinity with one another,
yet recognisably distinct.

This, then, is the view of the y-text to which the
evidence seems at present to point. It is a type of
text demonstrably connected with Egypt, being found

VIII LHEY TEXPUAL PROBLEM 337

in the Coptic versions, the bilingual fragments, and
MSS. which, like L, are associated with that country.
It has affinities with the §-text, which we have more
doubtfully assigned to Egypt also, so that in the more
important variations the two groups ordinarily agree
with one another. It is therefore substantially a sound
and good text, though the scholars who have examined
the evidence most closely regard it as showing some
signs of deliberate revision ; whence, in case of diver-
gence, the presumption is rather in favour of its rival.
Future discoveries of MSS. in Egypt, which may be
confidently looked for, will probably enable this theory
to be tested decisively in the future ; and the increase
of our knowledge of the Sahidic and Middle Egyptian
versions, which may be expected from the same source,
will have an important bearing on the solution of the
problem.

Meanwhile the authorities to which we must look
chiefly for readings of the y-type are nACLTY, the
group of fragments, mostly Graeco-Sahidic, from Upper
Egypt, formerly known as ΤΟΥ, and (in the Acts and
Epistles) 048, with the various Coptic versions, None
of these authorities is wholly of this type throughout,
all being found (as has been seen above) in some cases
with the @-group and in others with the a-group; but
where these groups can be separately distinguished,
then any residuum which is supported by some of the
authorities above enumerated may be safely regarded
as “ Alexandrian.” For instance, if the authorities for
a set of variants in a given place fall into the following
groups’ (1) NB) (2)(CLY Bohy 3 AEM; ete, or (1)
BD Bol, Syrr (2 "ALT iid-He.,(3) EFKU, ‘ete,
there can be no doubt that they represent the β- y- and
a-types of text respectively. Of the Fathers, Origen
and Clement are those in whose writings this text is

Z

338 CRITICISM OF NEW TESTAMENT | cm

most likely to be found; but here again it is inter-
mixed with readings of the β- and 6-types, and can
only be identified when these two possibilities have
been eliminated.

§ 4. The 8-text

The most difficult question relating to the history of
the New Testament text still remains to be answered,
namely, What is the origin of the 6-type of text, or, as
Hort calls it, the Western text? If this question could
be satisfactorily answered, the key to the whole history
would be in our possession. It is a type of text which
it is impossible to believe to be authentic as a whole,
and yet it can be traced back to sub-apostolic times
and is widely spread throughout the Christian world.
How did such a text come into existence, and what
is the value of it to us to-day in our search for the
primitive text of the New Testament ?

The character of the 6-text has been described
already in connexion with the MS. which is its prin-
cipal Greek representative, the Codex Bezae. It is
marked by many additions, great and small, to the
common text, and by a few striking omissions; by a
multitude of small and apparently pointless verbal
variations ; by frequent changes in the order of words;
and by frequent incorporations from, or assimilations
to, the parallel narratives in the other Gospels, In
the Epistles this last form of corruption is of course
absent, and the total amount of variation is much less ;
on the other hand, in the Acts, which equally escapes
assimilation, the total amount of variation is very great.
Indeed it is in this book and its companion, the Gospel
of St. Luke, that the divergence of the 6-text alike from
the a-text and the @-text is most marked: a pheno-

VIII THE TEXTUAL PROBLEM 339

menon which has given rise to a special explanation, of
which more will have to be said below.

The local distribution of the 6-text is a point of
considerable importance. Originally observed most
in the Old Latin version, it received the name of
“Western” ; and by this title it has been very gener-
ally known since the promulgation of Hort’s theory.
The title is, however, inadequate and misleading. The
discovery of the Curetonian Syriac showed that this type
of text was also current in Syria at a very early date ;
and the more recent discovery of the Sinaitic Syriac has
established this fact yet more clearly. It has also been
shown that Tatian’s Diatessaron was compiled from a
text of the same character, and that the primitive
Armenian version was derived from a similar source.
Hence it is clear that this type of text was at home
in the East as well as in the West, and it has been
plausibly suggested that the connecting link may be
found in the Eastern origin of the Church of Gaul.
Irenaeus, whose home was in Asia Minor, was also
subsequently Bishop of Lyons; and his writings show
that he used a Bible text of this character. Is it not
reasonable, then, to suppose that he and his companions
carried this text to Gaul, and that thence it became the
parent of the Old Latin version ?

It is, however, becoming more and more clear that
the range of the 6-text was not confined even to the
Syriac and Latin-speaking Churches. It is found even
in Egypt, the headquarters of scientific criticism. Hort
himself noticed? that Western quotations hold a prom-
inent place in some of the writings of Origen and in
Clement of Alexandria; and a detailed study of the
New Testament quotations of Clement? has confirmed

1 [ntroduction, p. 113.
2 P. M. Barnard, The Biblical Text of Clement of Alexandria (Texts
and Studies, v. 5 ἢ Cambridge, 1899).

340 CRITICISM) ΘΕ NEW “TESTAMENT: © cu

this view decisively. Hence Egypt must be added to
Syria and the West as one of the early homes of the
6-text ; the number of places in which it was not known
is becoming small; and the inappropriateness of the
designation “ Western” becomes increasingly evident.’

It is also clear that its origin must be referred to a
very early date. Both the Old Syriac and the Old
Latin versions are very ancient, and their common
origin must go back almost to primitive times. A
more precise proof is given by the early Fathers. In
all the earliest Christian writers whose works have come
down to us, the 6-text is predominant. It is found in
Justin and in Tatian, in the heretic Marcion and in
Irenaeus, in Origen and in Clement of Alexandria ; and
if the list is not still longer, this is probably due to the
scantiness of the remains of the earliest Christian
literature. Clearly a type .of ‘text. which is donnd
everywhere where we have evidence, in Syria, in Egypt,
and in the Latin-speaking West, is one whose character
must be closely scrutinised, and investigated without
prejudice.

Until recently the claims of the “Western” text met
with scant consideration and universal rejection. Its
variations, alike from the Textus Receptus and from
the leading uncials, were so numerous and apparently
so arbitrary that little or no weight was attached to
them ; and the small band of witnesses (principally D
and the Old Latin version) was treated almost as a
negligible quantity in the constitution of the text of
the New Testament. Even among themselves they
could not agree, the Latin MSS. falling (as we have

1 Mr. Miller tried ‘‘ Syro-Low-Latin,” but this is too cumbrous for
practical use, and still is not exhaustive. A non-committal designation,
such as “‘the 6-text,” seems preferable, if only scholars in general would
agree to adopt it. For a suggestion that Antioch was the original centre of
the 0-text, see p. 199.

VIII THE TEXTUAL PROBLEM 341

seen above) into at least three groups, and the several
individuals in each group differing markedly from one
another. Time and research have not indeed lessened
the amount of internal difference, but they have widened
our general conception of the character of the family
as a whole. They have shown that the 6-text is earlier
in date and more universally disseminated in the
Christian world of the second century than had
previously been realised ; and they have led scholars
to look further for an explanation of its origin.

The most important of these theories with regard to
the “ Western” text is that which was put forward by
the late Professor F. Blass of Halle, the well-known
classical scholar, whose experience in editing the texts
of the Greek classics stood him in good stead in dealing
with the analogous problems connected with the New
Testament.’ His attention was directed mainly to the
two books of St. Luke, in which the variations are the
most strongly marked ; but it is evident that his results
affect the entire question of the 6é-text. Briefly, his
theory was that Luke prepared two editions of each of
his works, and that the 6-text represents one of these
editions, and the #-text the other; the a-text being
obviously secondary and not entering therefore into
consideration. He supposed that Luke wrote his
Gospel in Palestine, and that subsequently, on his
coming to reside in Rome with St. Paul, he was asked
by the Christians there for copies of his work, and
thereupon wrote it out again for them, with such
alterations as an author naturally feels at liberty to
make in transcribing his own books. Similarly, in the
case of the Acts, one copy was no doubt made for

1 For English readers his results will be found most conveniently set
forth, in a fresh and vigorous style, in his Phz/ology of the Gospels (Macmillan,

1898). The fuller statement of his case is contained in his editions of the
Acts (Teubner, 1895 and 1896) and Luke (Teubner, 1897).

342 CRITICISM) OF NEW TESTAMENT cz,

Theophilus, to whom it is addressed, another for the
Church in general; verbal, and occasionally even
substantial, alterations being made in the later copy.
The natural tendency of such a revision, in Professor
Blass’ opinion, would be to abbreviate by the omission
of what seemed to be superfluous; hence in each case
the shorter text is to be regarded as the later. The
result of an investigation upon these lines is to show
that the “ Western” or 6-text, or, as Blass called it, the
Roman text, consists in fact of the later edition of the
Gospel and the earlier of the Acts ; while the “Neutral”
or 8-text, which Blass called Antiochene, consists of the
earlier edition of the Gospel and the later of the Acts.

This theory of a double edition is not wholly new,
the idea having been propounded, more or less inde-
pendently, by different writers at different times. The
first statement of it appears to have been made by
Jean Leclerc (Johannes Clericus) early in the eighteenth
century, and it was mentioned with approval by Bishop
Lightfoot, who, however, did not work it out. Blass’
statement of it was not merely independent of these
precursors, but was also far fuller and more elaborate ;
and he alone deserves the credit of having really
brought the theory into the arena of criticism and
made it an element with which textual scholars were
cempelled to reckon. At the time when the : first
edition of the present work was prepared, Blass’ theory
held a prominent place in textual criticism, and had
received the adhesion of such eminent authorities as
Dr. Salmon and Professor Nestle. At the present time
its vogue is somewhat past; but as it may at any time
be revived, and as the phenomena on which it was
based remain in any case to be explained, it seems
worth while to retain the original examination of it.

A priori, there is no difficulty in accepting the

VIII THE 'THEXTUAL PROBLEM 343

fundamental proposition. It is quite possible that
Luke may have revised his own work, and copies of
both editions might have survived. We know that
certain works of classical literature were issued in two
editions (for example Aristophanes’ Clouds and Plutus),
and in the informal conditions of publication which
must have applied to all Christian books in the earliest
times the simultaneous circulation of different editions
is quite conceivable. Practically, too, the theory fits
in with many of the daza, and satisfactorily explains
some of the principal variations. Many of these are
wholly inexplicable on the ground of any ordinary
scribal error or licence, and appear to postulate an
authoritative revision by some one with special know-
ledge of the facts in question, and some one who
thought he had the right to deal as freely as he
pleased with the text. Thus in Acts v. 29, for the
ordinary ἀποκριθεὶς δὲ ὁ Πέτρος καὶ of ἀπόστολοι εἶπον"
πειθαρχεῖν δεῖ θεῷ μᾶλλον ἢ ἀνθρώποις, the 6-text ' has
ἀποκριθεὶς δὲ Πέτρος εἶπεν πρὸς αὐτόν" tive πειθαρχεῖν
δεῖ, θεῷ ἢ ἀνθρώποις ; ὁ δὲ εἶπεν: Oew. In viii. 24
(the incident of Simon Magus) the 6-text adds the
words καὶ πολλὰ κλαίων ov διελίμπανεν. In x. 25
(the story of Cornelius and Peter) the ordinary narra-
tive ws δὲ ἐγένετο τοῦ εἰσελθεῖν τὸν Πέτρον, συναντήσας
αὐτῷ ὁ Κορνήλιος πεσὼν ἐπὶ τοὺς πόδας προσεκύνησεν
is replaced by a more detailed description, προσεγγίζοντος
δὲ tov Ilétpov εἰς τὴν Καισάρειαν, προδραμὼν εἷς τῶν
δούλων διεσάφησεν παραγεγονέναι αὐτόν. ὁ δὲ Κορνήλιος
ἐκπηδήσας καὶ συναντήσας αὐτῷ, πεσὼν κ.τ.λ., COntinu-
ing in the next verse εἶπεν δὲ αὐτῷ ὁ Πέτρος: Τί

1 Here and elsewhere I have taken the 6-text from Blass’ edition of St.
Luke and the Acts; but it must be remembered that the authorities of the
6-type differ very widely among themselves. In the present instance the
authorities for the reading here given are the Old Latin MSS. g and ὦ

(Blass’ 23) and Lucifer of Cagliari. The readings quoted in comparison
with those of the d-text are those of the Revisers (see p. 311).

344 CRITICISM OF NEW TESTAMENT εὐ

a \ \ / 5. ἸῸΝ \ bY f 5 e
ποιεῖς ; τὸν θεὸν προσκύνει' ἐγὼ yap ἄνθρωπός εἰμι ὡς
καὶ σύ. In xi. 2, where the ordinary text has merely
4 NESE χὴν ῇ ? ς , / \
ὅτε δὲ ἀνέβη Ἰ]έτρος εἰς “Ἱεροσόλυμα, διεκρίνοντο πρὸς
αὐτὸν οἱ ἐκ περιτομῆς, the 6-text has quite a long
addition: ὁ μὲν οὖν Ἰ]έτρος διὰ ἱκανοῦ χρόνου ἠθέλη-

a > € \ / \ > \
σεν πορευθῆναι εἰς “1.. Kai προσφωνήσας τοὺς ἀδελφοὺς

Ν > / 2) \ fen / / /
Kal ἐπιστηρίξας αὐτοὺς ἐξῆλθεν, πολύν TE λόγον ποιού-
μενος ἐπορεύετο διὰ τῶν χωρῶν διδάσκων αὐτούς. ὅτε
δὲ κατήντησεν εἰς ‘I. καὶ ἀπήγγειλεν αὐτοῖς τὴν χάριν
τοῦ θεοῦ, οἱ ἐκ περιτομῆς ἀδελφοὶ διεκρίνοντο πρὸς
αὐτόν. In xii, 10 a topographical detail is inserted
in the narrative of Peter’s miraculous deliverance, καὶ
ἐξελθόντες κατέβησαν τοὺς ἑπτὰ βαθμούς. In xiv. 2
(Paul and Barnabas at Iconium) the ordinary text has
οἱ δὲ ἀπειθήσαντες ᾿Ιουδαῖοι ἐπήγειραν καὶ ἐκάκωσαν
τὰς ψυχὰς τῶν ἐθνῶν κατὰ τῶν ἀδελφῶν, while the 5-text
has οἱ δὲ ἀρχισυνάγωγοι τῶν ᾿Ιουδαίων καὶ οἱ ἄρχοντες
ἐπήγαγον διωγμὸν κατὰ τῶν δικαίων καὶ ἐκάκωσαν
.. ἀδελφῶν" ὁ δὲ κύριος ἔδωκεν ταχὺ εἰρήνην : con-
tinuing in verse 5, καὶ πάλιν ἐπήγειραν διωγμὸν ἐκ
δευτέρου οἱ ᾿Ιουδαῖοι σὺν τοῖς ἔθνεσιν, καὶ λιθοβολή-

Ȣ/ 3 \ b) a / \ /
σαντες ἐξέβαλον αὐτοὺς ἐκ τῆς πόλεως, καὶ φυγόντες
5 > \ / 2 / \ f
ἦλθον εἰς τὴν AvKaoviay, εἰς πόλιν τινὰ καλουμένην
Λύστβαν. In the list of recommendations in the
Jerusalem Council (xv. 29) the 6-text adds a further
clause, καὶ dca μὴ θέλετε ἑαυτοῖς γίνεσθαι, ἑτέρῳ μὴ
ποιεῖν. In xvi. 35, the narrative of the release of Paul
and Silas from Philippi is amplified thus: ἡμέρας δὲ
γενομένης, συνῆλθον οἱ στρατηγοὶ ἐπὶ τὸ αὐτὸ εἰς τὴν
ἀγοράν, καὶ ἀναμνησθέντες τὸν σεισμὸν τὸν γεγονότα
ἐφοβήθησαν, and in verse 39 καὶ παραγενόμενοι μετὰ

,ὔ “Ὁ > \ / / > ΝΥ
φίλων πολλῶν εἰς τὴν φυλακὴν, παρεκάλεσαν αὕτους
> n > / > / \ ? e n v4 > \
ἐξελθεῖν, εἰπόντες" ᾿Ηγνοήσαμεν τὰ καθ᾽ ὑμᾶς, ὅτι ἐστὲ
ἄνδρες δίκαιοι. In xvii. 15 an explanatory clause is
inserted in the account of Paul’s journey from Beroea

vit THE TEXTUAL PROBLEM 345

to Athens; παρῆλθεν δὲ τὴν Θεσσαλίαν: ἐκωλύθη yap
εἰς αὐτοὺς κηρύξαι τὸν λόγον. In xviii. 27 a wholly
different account is given of the cause of Apollos’
journey from Ephesus to Corinth; for whereas the
ordinary text has βουλομένου δὲ αὐτοῦ διελθεῖν eis τὴν
᾿Αχαίαν προτρεψάμενοι οἱ ἀδελφοὶ ἔγραψαν τοῖς μαθηταῖς
ἀποδέξασθαι αὐτόν, the ὃδ-ἰαχί states ἐν δὲ τῇ φέσῳ
ἐπιδημοῦντές τινες Κορίνθιοι καὶ ἀκούσαντες αὐτοῦ,
παρεκάλουν διελθεῖν σὺν αὐτοῖς εἰς τὴν πατρίδα αὐτῶν'
συγκατανεύσαντος δὲ αὐτοῦ, οἱ ᾿Εφέσιοι ἔγραψαν κ.τ.λ.
In xix. I a wholly new detail appears in the 6-text:
θέλοντος δὲ τοῦ ἸΙ]αύλου κατὰ τὴν ἰδίαν βουλὴν πορεύε-
σθαι εἰς Ἱεροσόλυμα, εἶπεν αὐτῷ τὸ πνεῦμα ὑποστρέφειν
Εἰς rain Ασία τ Τα πχίχον δὲ itis) statedithat | Paul
taught in the school of Tyrannus ἀπὸ ὥρας πέμπτης
ἕως δεκάτης. Similarly in xx. 15 (Paul’s voyage to-
wards Jerusalem) between the mention of Samos and
Miletus is inserted the additional fact, καὶ μείναντες ἐν
Τρωγιλίᾳ, which is a noticeable instance of a reading
common to the 6- and a-texts, which is not in the
B-text. In xxi. 16 the disciple Mnason is represented
as living not in Jerusalem but in a village between
that town and Caesarea: οὗτοι δὲ ἦγον ἡμᾶς πρὸς ovs
ξενισθῶμεν, Kal παραγενόμενοι εἴς τινα κώμην ἐγενόμεθα
παρὰ Μνάσωνι Κυπρίῳ, μαθητῇ ἀρχαίῳ κἀκεῖθεν
ἐξιόντες ἤλθομεν εἰς ‘lepocodvpa. From xxii. 29 the
principal Greek witness for the 6-text, the Codex Bezae,
is wanting, but from the remaining witnesses the original
form of this type of text can be at least approximately
and in substance recovered. In xxili. 24, besides
other variations in the narrative of St. Paul’s convey-
ance as a prisoner to Caesarea, the following clause
is inserted: ἐφοβήθη yap μήποτε ἁρπάσαντες αὐτὸν οἱ
᾿Ιουδαῖοι ἀποκτείνωσιν, καὶ αὐτὸς μεταξὺ ἔγκλημα ἔχῃ
ὡς χρήματα εἰληφώς. Several phrases also in the

346 CRITICISM ΟΕ NEW (TESTAMENT. |) cu:

letter of Claudius Lysias are altered. In xxv. 24,25 a
passage of some length is inserted in Festus’ speech,
recapitulating some of the facts already known to the
reader from the preceding narrative. The narrative of
xXVil, 1. 15. worded quite differently)’ In) xxvii, (san
place of the ordinary τό τε πέλαγος τὸ κατὰ τὴν
Κιλικίαν καὶ Παμφυλίαν διαπλεύσαντες, the 6-text has
καὶ μετὰ ταῦτα διαπλεύσαντες τὸν Κιλίκιον κόλπον καὶ
τὸ Παμφύλιον πέλαγος δι’ ἡμερῶν δεκάπεντε. On the
other hand, the ordinary narrative of xxvii. II, 12
is concentrated into the single phrase ὁ δὲ κυβερνήτης
Kal ὁ ναύκληρος ἐβουλεύοντο πλεῖν, and the description
of the harbour of Phoenix, βλέποντα κατὰ λίβα καὶ
κατὰ χῶρον, is omitted. In xxviii. 16 an additional
item of information is given in the 6-text, which has
been preserved in several MSS. (HLMP) which are not
wholly of this type: (ὅτε δὲ εἰσήλθομεν eis Ῥώμην) ὁ
ἑκατόνταρχος παρέδωκε τοὺς δεσμίους TO στρατοπεδάρχῃ,
a title which Mommsen has shown may well be accurate.’
In xxviii. 19, after the words ἀντιλεγόντων δὲ τῶν
Ιουδαίων is added καὶ ἐπικραζόντων, Aipe τὸν ἐχθρὸν
ἡμῶν, and at the end of the verse ἀλλ᾽ ἵνα λυτρώσωμαι
τὴν ψυχήν μου ἐκ θανάτου. Finally in xxvili. 31 the
last words of the book, in place of the usual καὶ διδάσκων
τὰ περὶ τοῦ Κυρίου Ἰησοῦ Χριστοῦ peta πάσης παρ-
ρησίας ἀκωλύτως, the 6-text had something to the
following effect: διισχυριζόμενος καὶ λέγων ἀκωλύτως
ὅτε οὗτός ἐστιν ὁ Χριστὸς ὁ υἱὸς τοῦ θεοῦ, δι’ οὗ
μέλλει πᾶς ὁ κόσμος κρίνεσθαι.

Such are some of the more remarkable variants
presented by this type of text in this single book
of the Acts of the Apostles. In the Gospel of St.
Luke they are for the most part less striking, often
consisting merely of the omission or insertion of

1 Sitsungsberichte d. Berl. Acad., 1895, p. 491.

THE TEXTUAL PROBLEM

VIII

347

pronouns, the substitution of pronouns for proper
names, or vice versa, or the interchange of καί and δέ.
Blass emphasised the fact that the 6-text in the Gospel
is characterised by omissions as compared with the
a or B type, and in the Acts by additions; but on an
examination of his edition, it is difficult to resist the
feeling that this difference is less marked in reality
than he represents it, and is partly due to his attach-
ing special weight in the Gospel to those authorities
which show omissions, and in the Acts to those which
show additions ; and seeing that the authorities for the
6-text are rarely unanimous in support of any given
reading, much is necessarily left to the judgment of
the editor. An example of the kind of verbal variations
characteristic of the 6-text may be given from Luke v.
5-11, where they occur more thickly than usual. The
6-text is given from Blass’ edition, and the 8-text from
the Revisers’ Greek Testament, as before.

B-text

καὶ ἀποκριθεὶς ὁ Σίμων εἶπεν"
᾿Επιστάτα, δι’ ὅλης νυκτὸς κοπιά-
σαντες οὐδὲν ἐλάβομεν" ἐπὶ δὲ τῷ
ῥήματί σου χαλάσω τὰ δίκτυα. καὶ
τοῦτο ποιήσαντες συνέκλεισαν ἰχθύων
πλῆθος πολύ"
δίκτυα αὐτῶν" καὶ κατένευσαν τοῖς
μετόχοις ἐν τῷ ἑτέρῳ πλοίῳ, τοῦ
ἐλθόντας συλλαβέσθαι αὐτοῖς" καὶ
ἦλθον καὶ ἔπλησαν ἀμφότερα τὰ
πλοῖα, ὥστε βυθίζεσθαι αὐτά. ἰδὼν
δὲ Σίμων Πέτρος προσέπεσε τοῖς
γόνασι τοῦ ᾿Ιησοῦ λέγων, Ἤξελθε
am’ ἐμοῦ, ὅτι ἀνὴρ ἁμαρτωλός εἰμι,
Κύριε. θάμβος γὰρ περιέσχεν αὐτόν,
καὶ πάντας τοὺς σὺν αὐτῷ, ἐπὶ τῇ ἄγρᾳ
τῶν ἰχθύων ὧν συνέλαβον, ὁμοίως δὲ
καὶ ᾿Ιάκωβον καὶ ᾿Τωάννην, υἱοὺς LeBe-
δαίου, οἱ ἦσαν κοινωνοὶ τῴ Σίμωνι.
καὶ εἶπε πρὸς τὸν Σίμωνα ὁ ᾿Ιησοῦς,
Μὴ φοβοῦ ἀπὸ τοῦ νῦν ἀνθρώπους
ἔσῃ ζωγρῶν. καὶ καταγαγόντες τὰ
πλοῖα ἐπὶ τὴν γῆν, ἀφέντες ἅπαντα
ἠκολούθησαν αὐτῷ.

διερρήγνυτο δὲ τὰ

6-text

ὁ δὲ Σίμων ἀποκριθεὶς εἶπεν" Διδά-
σκαλε, Ou’ ὅλης τῆς νυκτὸς κοπιάσαντες
οὐδὲν ἐλάβομεν" ἐπὶ δὲ τῷ ῥήματί σου
οὐ μὴ παρακούσομαι. καὶ εὐθὺς
χαλάσαντες τὰ δίκτυα συνέκλεισαν
ἰχθύων πλῆθος, ὥστε τὰ δίκτυα ῥήσσεσ-
θαι. καὶ κατένενον τοῖς μετόχοις τοῖς
ἐν τῷ ἑτέρῳ πλοίῳ, τοῦ ἐλθόντας
βοηθεῖν αὐτοῖς" ἐλθόντες οὖν ἔπλησαν
ἀμφότερα τὰ πλοῖα, ὥστε παρά τι
βυθίζεσθαι. ὁ δὲ Σίμων προσέπεσεν
αὐτοῦ τοῖς ποσίν, λέγων, Παρακαλῶ,
ἔξελθε am’ ἐμοῦ, ὅτι ἀνὴρ ἁμαρτωλός
εἰμι, Κύριε. θάμβος γὰρ περιέσχεν
αὐτοὺς ἐπὶ τῇ ἄγρᾳ τῶν ἰχθύων ὧν
συνέλαβον. ἦσαν δὲ κοινωνοὶ αὐτοῦ
Ἰάκωβος καὶ ᾿Ιωάνης, υἱοὶ Ζεβεδαίου.
ὁ δὲ εἶπεν πρὸς τὸν Σίμωνα, Μὴ
φοβοῦ" ἀπὸ τοῦ νῦν ἀνθρώπους ἔσῃ
ζωγρῶν. οἱ δὲ ἀκούσαντες τὰ πλοῖα
κατέλειψαν ἐπὶ τῆς γῆς καὶ ἠκολού-
θησαν αὐτῷ.

348 CRITICISM OF NEW TESTAMENT cu.

A noticeable abbreviation occurs in v. 26, where the
6-text has only καὶ ἐπλήσθησαν θάμβους, λέγοντες,
Εἴδομεν παράδοξα σήμερον, in place of καὶ ἔκστασις
ἔλαβεν ἅπαντας, καὶ ἐδόξαζον τὸν θεόν, καὶ ἐπλήσθη-
σαν φόβου, λέγοντες ὅτι Hidowev παράδωξα σήμερον.
Similarly the 6-text omits v. 39 altogether: καὶ οὐδεὶς
πιὼν παλαιὸν θέλει νέον" λέγει yap, ‘O παλαιὸς χρηστός
ἐστιν. On the other hand it inserts in place of vi. 5
the story of the man working on the Sabbath day,
which has already been quoted above (p. 93). At ix.
55 the 6-text has the rebuke to the sons of Zebedee, οὐκ
οἴδατε ποίου πνεύματός ἐστε: ὁ γὰρ υἱὸς τοῦ ἀνθρώπου
οὐκ ἦλθε ψυχὰς ἀπολέσαι, ἀλλὰ σῶσαι, which other-
wise is found only in late authorities, being omitted by
NABCL, etc. At x. 41, 42 the 6-text omits the words
μεριμνᾷς καὶ θορυβάξῃ περὶ πολλά" ἑνὸς δέ ἐστι χρεία:
but in xi. 2-4 it has the fuller form of the Lord’s
Prayer, with the further addition (in D only) of the
introductory words (ὅταν προσεύχησθε) μὴ βαττολογεῖτε
ὡς οἱ λοιποί" δοκοῦσιν γάρ τινες ὅτι ἐν τῇ πολυλογίᾳ
αὐτῶν εἰσακουσθήσονται' ἀλλὰ προσευχόμενοι (λέγετε).
Another omission (in Daé) occurs at xii. 21, οὕτως ὁ
θησαυρίζων ἑαυτῷ καὶ μὴ εἰς θεὸν πλουτῶν. The most
important variants, however, whether of omission or
addition, occur in the later chapters of the Gospel.
Thus the narrative in xix. 31-35 of the procuring of
the ass for our Lord’s entry into Jerusalem is reduced
by a series of excisions to the words, καὶ ἐάν τις ὑμᾶς
ἐρωτᾷ, οὕτως ἐρεῖτε, ὅτι ὁ κύριος αὐτοῦ χρείαν ἔχει.
καὶ ἀπελθόντες ἀπεκρίθησαν ὅτι Ὃ κύριος αὐτοῦ χρείαν
ἔχει. καὶ ἀγαγόντες τὸν πῶλον ἐπέριψαν τὰ ἱμάτια
αὐτῶν ἐπ᾽ αὐτὸν καὶ ἐπεβίβασαν τὸν ᾿Ιησοῦν,----ἃ form
which cannot be considered probable. After xxi. 36,
Blass would insert the narrative of the woman taken in
adultery, which is universally admitted to be out of

oh) THE TEXTUAL’ PROBLEM 349

place in St. John; but for this there is no evidence
except that of the Ferrar group of minuscules, which
inserts the passage after xxi. 38. In the narrative of
the institution of the Lord’s Supper, the 6-text omits
the end of xxii. 19 τὸ ὑπὲρ ὑμῶν διδόμενον" τοῦτο
ποιεῖτε εἰς τὴν ἐμὴν ἀνάμνησιν, and the whole of
verse 20, καὶ τὸ ποτήριον ὡσαύτως μετὰ τὸ δειπνῆσαι
λέγων, Τοῦτο τὸ ποτήριον ἡ καινὴ διαθήκη ἐν τῷ αἵματί
μου, τὸ ὑπὲρ ὑμῶν ἐκχυνόμενον. Blass would also
(without MS. authority) eject the rest of verse 10, so
that the institution of the Sacrament would wholly
disappear from this type of the text. On the other
hand the incident of the Bloody Sweat and the Angel
of the Agony is included in the 6-text, though it forms
no part of the @-text. On the saying from the Cross,
Πάτερ, ἄφες αὐτοῖς, οὐ yap οἴδασι τί ποιοῦσι, the
authorities of the 6-family are divided, Blass siding
with the Curetonian Syriac and several Old Latin
MSS. (but against Dad, the Sinaitic Syriac, and the
Sahidic) in retaining the words. At xxiii. 53, how-
ever, he follows Dc and the Sahidic in the extraordinary
insertion καὶ τεθέντος αὐτοῦ ἐπέθηκεν TO μνημείῳ λίθον,
ὃν μόγις εἴκοσι ἐκύλιον, and at xxiv. I, with the same
authorities, he inserts ἐλογίζοντο δὲ ἐν ἑαυταῖς Tis dpa
ἀποκυλίσει τὸν λίθον; In xxiv. 6 the §-text omits οὐκ
ἔστιν ὧδε, ἀλλ᾽ ἠγέρθη: also the whole of xxiv. 12
ὁ δὲ Πέτρος ἀναστὰς ἔδραμεν ἐπὶ τὸ μνημεῖον, καὶ
παρακύψας βλέπει τὰ ὀθόνια μόνα" καὶ ἀπῆλθε πρὸς
ἑαυτὸν θαυμάζων τὸ γεγονός, the end of xxiv. 36 καὶ
λέγει αὐτοῖς, Εἰρήνη ὑμῖν, and the whole of xxiv. 40
καὶ τοῦτο εἰπὼν ἔδειξεν αὐτοῖς τὰς χεῖρας καὶ τοὺς
πόδας. Finally, all express mention of the Ascension
disappears from this form of the text, the words καὶ
ἀνεφέρετο εἰς τὸν οὔρανον being omitted, and only the
ambiguous phrase ἀπέστη am αὐτῶν retained.

380 CRITICISM OF NEW TESTAMENT cu.

Such being, in outline or selection, the most prom-
inent data, the question has to be faced, Does Blass’
theory provide an adequate explanation of them? At
first sight it may seem that no one except the author
himself could ever have taken the liberties with the
text which are involved in the greater changes, and
that the smaller ones are such as no copyist or editor
would have taken the trouble to make. That is, in
brief, the essence of Blass’ case. There are, however,
serious considerations to be taken into account on the
other side. The small verbal changes (individually too
unimportant to be included in the selection given above,
and collectively too numerous) are far from impressing
one as the product of the author's pen. A writer de-
liberately condensing his narrative must have done so
in a less haphazard manner. Very many of the sub-
stitutions of one word for another are inexplicable on
any theory of deliberate revision. It would require a
disproportionate amount of space to examine the evi-
dence in detail here,’ but if the reader will compare the
two forms of text in a chapter or two of the Gospel or
the Acts, he will find it hard to conceive the mind of
the author in the production of this supposed revision.
Changes which are possible to a careless scribe, or to
a writer indifferent as to the precise wording of his
text, are incomprehensible as the work of an author
transcribing his own composition.

Nor are the larger variants, deliberate though they
must have been, satisfactorily explicable on the theory
of a revision by the author himself. The additional
facts contained in the 6-text of the Acts have, no
doubt, the appearance of being due to special know-
ledge, and might reasonably have been the work of

1 A classified enumeration of the variants in D is given in Weiss’ Der
Codex D in der Afostelgeschichte (Texte und Untersuchungen, N.F. ii. 1).

ΠῚ THE TEXTUAL PROBLEM 351

Luke himself; but then the difficulty arises, how to
account for their omission in the later version. The
economy in space is insignificant, the loss in detail and
picturesqueness occasionally great. Some of the sup-
posed omissions are still more improbable, since they
occur not only in Luke’s own narrative, but in docu-
ments which he is quoting. Is it likely that he would
have struck out a clause from the recommendations of
the Jerusalem Council (Acts xx. 29), or that he would
have altered the wording of the letter of Claudius
Lysias (xxiii. 26-30)? Similarly, in the Gospel we
are asked to believe that Luke in his later edition (in
this case represented by the 6-text) altered his narrative
of the evening before the Crucifixion so as to omit all
mention of the institution of the sacrament of the Lord’s
Supper, and perhaps even excised the record of one of
our Lord’s sayings from the Cross. The omissions in
chapter xxiv. are surely inexplicable as the deliberate
excisions of an author bent on improving his literary
style by condensation.’ Nor is Blass’ theory more
satisfactory in dealing with the additions which are
found in the 6-text of the Gospel. The most important
of these is the passage introduced at chapter vi. 5, con-
cerning the man found working on the Sabbath; and
to this Blass would add the story of the woman taken

1 Blass assumes throughout that an author’s revision of his work natur-
ally takes the form of condensation, and states that he finds it to be so in
his own case; but it may be doubted whether the experience of all authors
is the same. A historian is constantly tempted to add fresh facts and
arguments to his narrative; and the tendency of most modern books is to
increase in size in their successive editions. From this tendency Prof. Blass’
own works would not appear to be exempt, his history of Attic Oratory
having increased from 1763 pages to 1863 (so far as the second edition has
yet appeared), while his introductions to his editions of Aristotle’s ᾿Αθηναίων
Πολιτεία and the poems of Bacchylides have grown from twenty-eight pages
to thirty-one in the first case, and from sixty-two to seventy-one in the
second. If, then, expansion rather than condensation be taken as the sign
of a second edition, Blass’ views as to the relative priority of the two texts
of the works of Luke must be inverted ; which will not, however, remove
their difficulties.

352 CRITICISM OF NEW TESTAMENT occu.

in adultery, which the Ferrar MSS. place at the end of
chapter xxi., and which Blass believes to have formed
part of Luke’s “Roman” edition of his Gospel. In
both of these cases Blass holds that Luke deliberately
withheld these passages from the edition which was
intended to circulate in the East, as being likely to
give offence to the Jews; while he included them in
the edition addressed to his Roman readers, who would
not find the same difficulty in them. This explana-
tion, however, overlooks the fact that, according to
Blass’ own theory, the first edition of the Gospel (that
intended for the East) was written at a considerable
time before the second edition, when no idea of a
Roman edition can have been present in his mind.
No deliberate reservation of these narratives for
Western readers can possibly have been intended.

And while Blass’ theory does not appear to fit the
facts with regard to the two books of St. Luke, it is a
further objection to it that it does not account for the
similar phenomena in the other books of the New Testa-
ment. The variants of the 6-text are, no doubt, most
conspicuous in St. Luke’s Gospel and in the Acts; but
they are not confined to these books. Similar additions
and variations appear, though not so frequently, in the
other three Gospels, and to some extent in the Epistles ;
though in the latter the authorities of this type are fewer,
and have been less fully investigated. For instance, the
largest addition of all in the Codex Bezae is the passage
inserted after Matthew xx. 28; and the omission by D
and the Old Latin of Matthew xxi. 44 (“and whosoever
shall fall on this stone shall be broken, but on whom-
soever it shall fall, it shall grind him to powder”) and
of the end of John iv. 9 (“for the Jews have no deal-
ings with the Samaritans’) are quite of the same nature
as the omissions in Luke. Blass, however, is driven

VIII THE TEXTUAL PROBLEM 353

to find different explanations for each of the other
Gospels. In Matthew he finds that there is compara-
tively little divergence between the §-text and the
6-text, and what there is he regards as deliberate inter-
polation in the latter by very ancient readers. In
Mark, on the other hand, where the divergence is greater,
he propounds the bold theory that the evangelist origin-
ally wrote in Aramaic, that Luke translated his work
with additions or alterations of his own, and that this
Lucan version circulated side by side with another
translation made by some one else.’ Finally, the vari-
ants in the case of St. John’s Gospel are assigned to
the disciples of the evangelist, who “took the liberty of
enlarging the text here and there, of course each in a
different way.” °

This multiplication of hypotheses to account for
similar phenomena in the several Gospels is evidently
unsatisfactory, and weakens belief in the principal one,
which necessitates the others. Indeed, the more this
hypothesis is examined, the less does it seem to account
for the phenomena of the type of text with which we
are now dealing. The Codex Bezae, the principal
Greek representative of the 6-text, contains many vari-
ants peculiar to itself, and yet of the same character
with those which are accepted as belonging to the
“Roman” text. Thus it incorporates in Luke the
genealogy of our Lord given by Matthew, and in many
other passages introduces words or incidents from one
Gospel into another. It is quite clear that this cannot
be the work of Luke, and Blass does not adopt them
in his “Roman” text. It is evident, therefore, that

1 For the arguments by which this strange theory is supported, the
reader must consult Blass’ Phzlology of the Gospels, pp. 190-218. They
consist of a succession of hypotheses, each barely possible, and collectively
possessing only an infinitesimal degree of probability.

2 70.:p. 234.

2A

334 CRITICISM OF NEW TESTAMENT cu.

bold revision has been at work on D or its archetype,
just as Blass finds it at work in John; and the question
naturally presents itself, Why may not the same cause
account for the variants which D shares with other
authorities, and which Blass accepts as Lucan ?

What we want, in fact, is some uniform cause appli-
cable to the whole range of phenomena presented by
the 6-text, with some special addition to account for
their special prominence in the two books of St. Luke.
This is not provided by Blass. He merely makes a
selection among the variants found in these two books,
and labels the text so produced, “Roman.” ‘The title
is, however, even more “ question-begging” than Hort’s
“Western” or “Syrian,” since its localisation is more
precise. Yet there is, in point of fact, no evidence
connecting the 6-text especially with Rome, and much
that connects it with places far removed from Rome.
It is found in the Syriac-speaking countries of the East,
in the Greek-speaking Church of Egypt, in the Latin-
speaking Church of Africa, in the Latin-speaking but
Greek-descended Church of Gaul. From Italy and
Rome our evidence, with the possible exception of
Marcion’s, is all of later date. This type of text may
have been current there in the first and second centuries,
but we have no proof of it. What we do know is that
it was current at the earliest date to which our know-
ledge extends in nearly all the other parts of the world
to which the Gospel had been carried,—a phenomenon
for which it is difficult to account on the theory that
this text was of purely Roman origin, and that rival
texts of equal authenticity had already been given to
the East by the author himself.

What, then, is the alternative explanation which
will fit the facts of the case, if Blass’ ingenious hypo-
thesis be discarded? It is that to which allusion has

VIII THE TEXTUAL PROBLEM 355

already been made, and which Blass himself applies
in the case of the Gospels of Matthew and John, the
hypothesis of free handling of the text by scribes and
teachers in the early days of Christianity. It must be
remembered that the circumstances of the tradition of
the New Testament text (and especially in the case of
the Gospels) for more than 200 years were wholly
unlike those of any other literary work. We have no
creat libraries enshrining standard copies of the precious
volumes, no recognised book-trade multiplying carefully
written transcripts of them, no scholars keeping a
Gitical eye on the purity of the text. “Instead of all
this we have roughly written copies circulating from
hand to hand among congregations whose sole care
was for the substance, not for the precise wording, of
the Gospel narrative; we have the danger of destruc-
tion impending over them, if they were brought too
prominently before the eye of the civil power; we
have periods of persecution, during which active search
was made for the sacred books of the prohibited sect.
Circulating in this irregular fashion, and for the most
part among populations with no high standards of
literary tradition, it is not surprising that the text was
often treated in a way to which we are not accustomed
in dealing with the ordinary works of literature which
have descended to us from the past. It would not
seem unnatural, still less wrong, to insert additional
incidents, believed to be authentic, in the narrative of
our Lord’s life; and verbal changes, whether of pro-
nouns or proper names, or of one synonym for another,
would be matters of indifference.1 To these causes of

1 A vivid description of the corruptions which may overtake the manu-
scripts even of works copied under much less unsettled conditions is given
by the editor of St. Augustine’s De Czvtate Dez, J. L. Vives (in the preface
to the edition of 1555; the passage is not in his first edition of 1522):
‘* Mira dictu res quanta in codicibus varietas, ut unusquisque describentium
putaret sibi licere verba arbitratu suo ponere, modo constaret sensus, quasi

356 CRITICISM OF NEW TESTAMENT . cu.

variation may be added the plausible, though not yet
fully developed, suggestion of re-translation from the
Syriac as a possible factor in the production of this
type of text.’

So through the outlying tracts of the Roman world,
in Syria and Asia Minor, in Africa and Gaul, the books
of the New Testament circulated through the first two
centuries or more of their existence. There must have
been much truth in Jerome’s remark, “tot sunt ex-
emplaria paene quot codices.” What we have called
the 6-text, indeed, is not so much a text as a congeries
of various readings, not descending from any one
archetype, but possessing an infinitely complicated and
intricate parentage. No one manuscript can be taken
as even approximately representing the 6-text, if by
“text” we mean a form of the Gospel which once
existed in a single manuscript. There are a multitude
of readings in D (such as those referred to just now,
which incorporate passages from one Gospel into
another, or the pseudo-Homeric insertion in Luke
Xxlil. 53) which no editor can possibly regard as
authentic. No two manuscripts of the Old Latin

interpretatio esset, non exscriptio. Ita in hoc libro legas ardztror, in illo
puto; in hoc significatum, in illo fguratum ; hic praesens, alibz iste ; hic
habet evgo, ille zgztur; hic acternzs, ille zmmortalis; hic flextsse, ille
deflextsse, alius z7flextsse. Sed haec fortasse tolerabilia. Quid illa detracta,
quid addita, inversa? Jam quoties erratum, quod versionem Ixx. inter-
pretum, quam ubique Augustinus adducit, voluerunt ad hanc nostram
detorquere, et ex duabus male cohaerentibus unam facere.”” It will be
observed that three of the most important forms of corruption of the
Biblical text are here described: (1) wanton handling of the text, (2)
alteration of quotations from their true form into one more familiar to the
scribe, (3) conflation.

1 See above, p. 94 ff. Mr. Hoskier has recently carried this idea
further, by attributing much of the intermingling of different strains of
text to the existence at a very early period of Graeco-Latin-Syriac-Coptic
MSS. _ It would be difficult enough for bi-lingual MSS. (such as the Graeco-
Latin and Graeco-Coptic MSS. which we know to have existed in the
fifth and sixth centuries) to have been produced in the second or third
centuries; and a quadri-lingual MS. passes belief. Nor do the facts
adduced seem to require so improbable an assumption.

VII THE TEXTUAL PROBLEM 357

agree with any closeness with one another; the
differences between the two extant codices of the Old
Syriac are flagrant ; and the Old Latin and Old Syriac
texts differ somewhat markedly from one another.
The readings which we group under the head of the
6-text often bear, it is true, the signs of deliberate
revision, but never of a coherent and continuous revision.
They are the work of many individuals, in many places
and at many times, not of a single editor, still less of
the original author.

But it may fairly be asked, how, on this hypothesis,
is the special predominance of readings of the 6-type
in the two books of St. Luke to be accounted for. It
might be possible to frame various hypotheses not
more improbable than that which Blass applies to the
Gospel of St. Mark,—-possible explanations on which
πο veriication can: be) brought to: bear...) But) the
simplest theory perhaps is to suppose that these two
books were most exposed to free treatment because
they circulated most among the Gentile converts to
the Christian faith, The Acts of the Apostles, written
by the companion of the great Apostle of the Gentiles,
would naturally be a favourite book among the peoples
whom that Apostle had been the first to bring into the
Church ; and when they desired an authentic narrative
of the life of Christ, they would naturally turn to that
which came to them with the authority of this same
teacher. Circulating in Palestine and Asia Minor and
Greece, the Acts might easily receive those touches
of local detail which most suggest the author’s own
hand ;’ while in the case of the Gospel the variants
rather suggest free handling in the way of insertion or

1 This, it may be observed, is the explanation offered by Professor
Ramsay before the promulgation of Blass’ theory. The truth of the local
touches impressed him, but he was able to account for them in this way,
without attributing them to Luke himself.

358+ CRITICISM (OF NEW TESTAMENT © ex:

omission or verbal change than the application of local
knowledge. Other explanations might be possible, if
we knew more of the circumstances of the production
and early circulation of the Christian Scriptures ; but
in the present state of our knowledge this seems best
to account for the facts as they now lie before us,
and to contain in itself nothing either impossible or
improbable.

If a clearer understanding of the nature of the
6-text, and therewith of the early history of the New
Testament Scriptures, is to be achieved, it will hardly
be by heroic measures such as those we have just
been considering, but rather through a patient and
detailed examination of the various authorities in
which it is contained. Each of them possesses a dis-
tinct individuality, which requires careful investigation
before its relations with the others can be determined.
Most of them have already been described at some
length in the preceding chapters. For the Gospels we
have pre-eminently D, though it must not be assumed
that everything which appears in D is part of the
earliest form of the 6-text. Rather it must be regarded
as embodying the principles to which the 6-text is due,
carried to their farthest extent. No other Greek MS.
of the Gospels belongs wholly to this type, though
readings characteristic of it appear in RACL#, and
perhaps in other MSS. Of the minuscules 473 (Hort’s
81) is the most valuable representative of this class;
and with it may be mentioned 235, 431, 604, 1071,
and Evil. 250. The Ferrar group also (13, 60, 124,
4326) 248) 543,07 13,988; 826.829) sea pam
though agreeing in the main with the ordinary a-text,
yet in its characteristic features is allied with the 6-text ;
and Professor Rendel Harris has sought to establish
a Syro-Arabic origin for it, which must be considered in

VIII THE ,TEXTUAL PROBLEM 359

connexion with Dr. Chase’s theory of a Syriac origin for
the characteristic features of the 6-text in general. But
although all these MSS. may be cited as witnesses to
the 6-text in greater or less degree, the real strength of
its support comes from the Versions (especially the Old
Syriac and the Old Latin) and the early Fathers. The
Syriac and Latin versions, though often agreeing, appear
to represent somewhat different stages in the develop-
ment of the text. Prof. Burkitt has shown’ that the
longer additions to the Gospel text are especially found
in the African form of the Old Latin (represented
mainly by e and Cyprian), while the smaller additions
are especially characteristic of the European Latin (ad
and their colleagues); and the Old Syriac authorities
have some special interpolations peculiar to themselves.
Of the two Old Syriac MSS., the Sinaitic undoubtedly
represents the earlier and purer form of the version.
Besides these versions, the Harkleian Syriac and the
Sahidic and Armenian must be mentioned among the
intermittent supporters of the 6-text. Of the Fathers,
it has been shown already that all the earliest among
them exhibit texts more or less of the 6-type. Most
notable are Tatian, Aphraates, and to a less degree
Ephraem, among the Syriac writers; Justin, Marcion,
Irenaeus, and Tertullian among the earliest represen-
tatives of the West; Cyprian and Tyconius in Africa
at a somewhat later period; Clement and sometimes
even Origen in Egypt. The special importance of these
writers has been considered in an earlier chapter.

In the Acts D stands out still more prominently as
the representative of the 6-text, through the disappear-
ance of the Old Syriac. It is joined, however, by
the other Graeco-Latin MS. E,, and the minuscules

1 The Old Latin and the [tala (Cambridge Texts and Studies, iv. 5),
especially pp. 46-53.

360. CRITICISM OF NEW TESTAMENT "ca.

69; 431, 614,376 @lort’s 44), 1518) | The Old Latin
attestation is weaker than in the Gospels (see above,
p. 206); while the Old Syriac fails entirely. No MS.
of it is in existence, and it is only by the indirect
evidence which we possess, through the Syriac Fathers,
of the existence of an Old Syriac version of the Pauline
Epistles, that we are enabled to assume that there must
also have been an Old Syriac Acts. From the Fathers
the testimony to the 6-text is substantially the same as
in the Gospels, but their quotations are less numerous.

For the Catholic Epistles the evidence is still weaker,
through the loss of this part of D, and the absence of
these books from E,. The importance of the 6-text is,
however, much less after we leave the Gospels and Acts.

In the Pauline Epistles’ also the 6-text is less
strongly marked, but it exists quite definitely in the
Graeco-Latin group of uncials D,E,F,G,, all of which
have been traced by Corssen to a common ancestor,
written colometrically, probably in Italy. The Gothic
version is also connected with this group. The Old
Latin is represented by the MSS. enumerated above
(p. 209); and the existence of an Old Syriac version of
these books may be inferred, and some of its readings
gathered, from the commentary of Ephraem, of which
an Armenian translation exists, which was made avail-
able for general use by the publication of a Latin version
of it in 1893.2. In addition to these normal witnesses
to the $-text, B itself, as has been observed above, has
a strong tinge of this type in the Pauline Epistles.

For the Apocalypse there is no uncial representative
of the 6-text, which is practically represented only by
the Old Latin, here reinforced by the commentary of
Primasius. Little study, however, appears to have

1 See Sanday and Headlam, Romans, pp. 1xix-lxxiv.
2 See Armitage Robinson, Zuthalzana, p. 83.

VIII THE: TEXTUAL PROBLEM 361

been made of the characteristics of the 6-text in this
book.

In conclusion, it may be permissible to repeat that
the great diversity among the authorities of the 6-type
is a strong argument against such a theory of its origin
as that propounded by Blass. It is a type rather than
a text, produced in many places by the operation of the
same causes, and with intricate inter-relations existing
among its surviving representatives, which it will need
the patient labour of generations to unravel. The
scholars who undertake the task, however, will be
sustained by the consciousness that so far as they may
succeed in elucidating the genesis of the 6-text, they
will have discovered the key to the whole textual
history of the New Testament.

We come back then, finally, to a position substan-
tially the same as that of Hort, though with some modi-
fications. The early history of the New Testament
text presents itself to us as an irregular diffusion of the
various books among the individuals and communities
which embraced Christianity, with few safeguards against
alteration, whether deliberate or unintentional. To this
stage, which follows very soon on the production of
the original autographs, belong the various readings,
early in their attestation yet comparatively rarely con-
vincing in themselves, which we call the 6-text, and
which Hort terms “ Western,” and Blass (in the case of
the two books of St. Luke) “Roman.” In Egypt alone
(or principally) a higher standard of textual fidelity
prevailed, and in the literary atmosphere of Alexandria
and the other great towns a comparatively pure text
was preserved. This has come down to us (possibly
by way of Origen and his pupils) in the Codex
Vaticanus and its allies, and is what we have called the
8-text, and what Hort calls “ Neutral.” Another text,

362 CRITICISM OF NEW TESTAMENT cu.

also found in Egyptian authorities, and differing from
the last only in minor details, is that which we call the
y-text, and Hort “ Alexandrian.” Finally there is the
text which, originating in the neighbourhood of Antioch
about the end of the third century, drew together many Ὁ
of the various readings then in existence, and with many
minor editorial modifications developed into a form
which was generally adopted as satisfactory throughout
the Eastern Church. This is the a-text of our nomen-
clature, Hort’s “Syrian”; the text which monopolised
our printed editions until the nineteenth century, but
which is now abandoned by all but a few scholars, though
it is enshrined in the affections of the English people
through its incorporation in our Authorised Version.
Nevertheless, in coming back to Hort’s position, and
believing his analysis of the textual problem to be sub-
stantially sound, we do not necessarily go so far as he
does in the rejection of all evidence which lies outside
the 8-group. The course of investigation, aided by
evidence which has come to light since Hort wrote, has
tended to emphasise the early and widespread character
of the 6-text, and to render it probable that, among
much that is supposititious, there is also something that
is original, and yet is not preserved in any other form
of text. However highly we rate the accuracy and
judgment of the Alexandrian scribes and scholars to
whom the @-text is due, it is most improbable a prio
that they should always be right, and the scribes of the
6-text always wrong, when they differ. One special
class of readings in which, as Hort pointed out, a przorz
probability is on the side of the 6-text is in the case of
its omissions, such as those which occur in the final
chapters of St. Luke. Addition to a text is always
much easier to account for than omission, except when
the omission can be shown to be either purely accidental

VIII THE TEXTUAL PROBLEM 363

or due to doctrinal considerations ; and neither of these
explanations suits the cases in question. The passages
which the 8-text contains over and above what is given
in the 6-text may be genuine incidents and sayings,
and two of them at least, the Bloody Sweat and the
angel in the garden of Gethsemane, and the word of
forgiveness from the Cross, are such as one would not
readily abandon ; but it must remain very questionable
whether they formed part of the original Gospel. The
same may be said, with various degrees of probability,
of many of the passages which the 6-text alone con-
tains. But it is not safe to condemn a reading off-hand
because the authorities for it are of the d-type. If they
are right, as against the @-text, in omitting the passages
above referred to, it is probable that they are some-
times right in their other divergences. The several
passages must be considered on their merits, and the
ordinary canons of textual criticism, which enable one
to judge which of two rival readings is original, must
be applied to them. The presumption, no doubt, is on
the side of the @-text, since it has the higher character
for accuracy on the whole; but we must have an open
mind to consider the claims, in special instances, of its
rival. It would be simpler, no doubt, to be able to rule
out all 6-readings, as we rule out all recognisable
a-readings ; but the easiest way is not always the one
which leads to truth, and the tendency of recent criti-
cism has certainly been to rehabilitate, to some extent,
the 6-text, and to demand a more respectful considera-
tion of it in the future.

§ 5. Von Soden’s Textual Theory

The course of textual criticism, since the publication
of the first edition of the present work in 1901, has

364 CRITICISM OF NEW TESTAMENT © ex.

not disturbed the main conclusions then stated. On
the contrary, while the advocacy of the a-text has
practically come to an end, the advocates of the 6-text
cannot be said to have gained ground. Little is heard
now of Blass’ theory, and it is rather the differences
than the agreements of the several witnesses to the
6-text that have attracted attention. The Latin versions
have been the subject of much careful work, but no
great results in the way of constructive theory have been
reached. Perhaps the most notable single contribution
to the textual history of the New Testament has
been Prof. Burkitt’s demonstration (for it is little less)
that the Peshitto may be assigned with considerable
confidence to Bishop Rabbula, in the first quarter of
the fifth century ; a fact which clears the way for a
reasonable view of the history of the Syriac versions in
general.

Meanwhile the great edition of the Greek New
Testament, planned by von Soden and executed with
the help of a band of assistants, has been slowly coming
to the light. The prolegomena, in 2203 pages, are
complete; the text and critical apparatus (originally
promised for 1903!) are still in the press. The prole-
gomena, in addition to the new numeration of the manu-
scripts, described above (pp. 52-55), contain a very
elaborate classification of their textual evidence, based
upon a much more extensive examination of them than
has ever been undertaken previously. The exact bear-
ings of this classification in von Soden’s mind cannot be
fully discerned until the text and apparatus are pub-
lished ; but its general character may be briefly indicated.

1 Die Schriften des Neuen Testaments in threr dltesten erretchbaren
Textgestalt hergestellt, auf Grund threr Textgeschichte, von Dr. Theol.
Hermann Freiherr von Soden, Band i. (in four parts), θυ], 1902-1910.
For a good analysis of it, see K. Lake, Professor Hl. von Soden’s Treatment
of the Text of the Gospels, Edinburgh, 1908 (reprinted from the evzew of
Theology and Philosophy).

VIII THE TEXTUAL PROBLEM 365

In all the mass of authorities for the text of the
New Testament, von Soden discerns three main families.
The family which ultimately became dominant, and
which includes by far the greater number of extant
MSS., is designated by the letter Καὶ (=Kown). Over
against this we distinguish a much earlier family, de-
noted by the letter H, because von Soden (following
Bousset) regards the Alexandrian scholar Hesychius
as the editor of this recension, of which the Codex
Vaticanus is the principal representative. And beside
this, a study of these earlier witnesses leads to the
discovery of a third family, nowhere preserved in any
single MS. so plainly as H is in the Vaticanus, but
pervading a number of authorities, such as the Old
Latin, Old Syriac, the) Ferrar group, D, and: 1 This
family, which von Soden claims as his own discovery,
is designated I (Ἱερουσαλήμ, its home being assigned
to Jerusalem).

All these families are examined in detail, and sub-
divided into groups, of which the characteristic readings
are set out in full. Thus Κα has as its earliest form
Κ΄, followed by Ki and K** ; then K* (the dominant text
from the tenth or eleventh century onwards), and K*
(a twelfth century revision for liturgical use). Besides
these ‘there: is the group K*, headed ὃν. the Codex
Alexandrinus, which however is influenced by H; it
is a large group with a good many subdivisions. The
second family, H, is represented by about 50 witnesses,
mostly imperfect, headed by B and δὶ (Soden’s 61 and
62, from which he deduces a common ancestor, 8”).
B (6') comes nearer to H than any other MS., but is
influenced by I and K*. I has a rather bewildering
number of subdivisions, showing the influence of the
parent type in varying degrees; these are H’ (found
predominantly in the cursive MS. 1), J (=the Ferrar

366 CRITICISM OF NEW TESTAMENT ὅπ

group), ® (with varieties indicated by ©*, D>, H*, OH"),
B, K* (which thus appears under both K and I, being
regarded as a revision of an I text in the direction of
K', with reference to H), I’, II (the purple MSS.),
O, [°, and a number of small groups in which the
original traces of the I type are more and more
swamped by the predominance of K. For the MSS.
constituting all these groups and sub-groups, the student
must be referred to von Soden’s own work. JI? is the
best representative of the I type, and includes the
manuscripts D, 399, 565, étc.; D is described asa
mixture of I and K’, showing much careless copying
or dictation, but also many deliberate alterations, due
to various ancestors. The influence of parallel passages
in other books is very strong in it, as it is in the Old
Latin, by which D is strongly influenced.

Behind these three families, I, H, and K, lies the
common basis I-H-K, which must be arrived at by a
comparison of the three branches. Of these K is much
the farthest removed from the original, and shows much
verbal and stylistic alteration. It originated in Syria
in the fourth century, and is perhaps due to Lucian,
whom we know to have produced an edition of the
Septuagint. H (which is substantially the text of
Athanasius, Didymus, and Cyril of Alexandria) obvi-
ously is an Egyptian text, and presumably due to
Hesychius. It represents I-H-K much more faithfully
than K, and rarely departs from it except on grounds
of verbal style. I in general preserves the common
original most faithfully of all. Though now not pre-
served anywhere in its integrity, it must have been
authoritative for centuries, since its influence is so wide-
spread. It is the text of Cyril of Jerusalem and
Eusebius of Caesarea, to whom it is probably due.

The date of I-H-K goes back before Origen, who

Vill THE TEXTUAL PROBLEM 367

used it. It might be a tempting hypothesis to claim
him as its author, but this is not possible, since he often
varies from it; though if this argument is to be valid,
it would be necessary to examine the chronology of
Origen’s works, in order to see whether the divergence
continues to the end of his life. The Egyptian Fathers
after Origen use I-H-K, and this was the Greek text
used by Jerome in the preparation of the Vulgate.

If we try to carry our investigation back yet further,
we find in such evidence as is available (the Old Latin,
the Old Syriac, the early Fathers) many variations
from I-H-K, often palpably inferior, yet supported by
more than one witness. What is the cause of this?
Here we reach the most striking point in von Soden’s
textual theory. He attributes all the vagaries which
disfigure the text and complicate its history to the
influence of Tatian’s Diatessaron. This he regards as
having been originally written in Greek, before Tatian
left the orthodox Church; and he finds its influence at
work in the Old Latin, the Old Syriac, and sporadically
in the early Fathers; also in K. Among MSS. of the
Gospels, D has most of it, and there is much in x, but
hardly any in B.

If we can eliminate the variations due to Tatian, we
can fix the text of the New Testament as it existed
about A.D. 140, and the probability is strong that it
had not been seriously modified before that date.

Such is, in brief outline, von Soden’s theory. It is
obvious that it cannot be fully judged until the text
which he has based on it is published; but the main
points of interest and inquiry suggest themselves readily
in advance. Of the three families of text which he
discriminates, K is Hort’s “Syrian” text, and H his
“ Neutral.” Here von Soden’s principal service is that
he has examined the Syrian or a-text much more

30603 CRITICISM OF NEW TESTAMENT cu,

thoroughly than any other critic, and has classified it
into families. A statement of the typical readings of
these families in tabular form, not in solid blocks of
very small type, would be a considerable service to
future collators of new MSS. The third family, I,
includes most of the authorities which are generally
assigned to the “ Western” or 6-text; but von Soden
appears to claim that he can purge away the many
errors (principally due to Tatian) with which they teem,
and disengage a form of text which is superior even to
H, and which goes far towards giving us back the
fundamental text I-H-K. This is a claim which can
only be criticised when the text is published, and on
his success in this respect will depend the validity of
any claim that von Soden has made a material advance
on the position of Westcott and Hort.

The identification of Tatian as the great element of
disturbance in the text of the New Testament (analo-
gous in some degree to the influence of the Hexapla
on the Old) is a proposition which needs fuller proof
than von Soden has as yet provided. It is not without
attractiveness. To those who hold that the 6-text is
full of non-authentic divergences, it provides an alter-
native explanation to that of the carelessness, and lack
of opportunity of correction, of early scribes. Further,
Tatian’s work was undoubtedly influential, especially in
the Syriac-speaking countries, while it was also known
in Italy; hence the appearance of its characteristics in
both Eastern and Western versions and Fathers can
be readily accounted for. But fuller proof of the hypo-
thesis is needed ; and von Soden’s present treatment of
it is too superficial to be convincing. In particular,
the discussion of the Versions is very slight in propor-
tion to their importance.

It is, however, satisfactory to find that (so far as

vit THE TEXTUAL PROBLEM 369

can be judged before the appearance of the text) the
general result of von Soden’s very exhaustive inquiry
has been to confirm the main features of the textual
theory laid down by Westcott and Hort. Differences
of judgment must always remain, and the acceptance
or rejection of individual readings of the 6-text or the
I text must always depend on the personal equation of
the critic; but it does not appear that von Soden has
made any discovery which threatens the pre-eminence
of the text which found its home in Egypt.

Here, then, the subject must be left for the present,
—a subject still full of life, as it is perennially full of
interest ; a subject to which scholars may still devote
their best endeavours, with the expectation of sub-
stantial fruit; a subject which may still be illuminated
by fresh discoveries of materials, or by the promulgation
of fresh theories ; but one of which the foundations
now stand fast on a firm basis of ascertained historical
fact, and on which the superstructure of religion may
be built with full hope and confidence that it rests on
an authentic text.

2B

LNDE Xl

SUBJECTS AND PERSONS

Abbot, Ezra, 149, 245

Abbott, T. K., 116, 131, 201, 205,
232

Abbreviations in papyri, possible
cause of variants, 96

Acacius and Euzoius, 48

Accents, in papyri, 24

Acts, complete uncial MSS. of, 57;
section-division of, 81, 82, 226;
Western or 6-text in, 343-346

Acts of Paul, 98

Adier; J. αἱ C., £67

Africa, suggested as home of earliest
Latin version, 198

Akhmimic version, 181, 194

Alcuin, revised edition of Vulgate by,
218, 231, 233, 235

Aldus, edition of N.T. by, 270

‘« Alexandrian’’ type of text, 299.
See also y-type

Alexoudi, Anthymius, 120

Alter, F. K., publication of MSS. by,
136, 283

Ambrose, St., 260

Amélineau, E., Coptic MSS. pub-
lished by, 186, 188

Ammonius of Alexandria, Gospel
harmony by, 69

Anthony, St., 178

Antioch, suggested as home of

earliest Latin version, 199 note
Aphraates, N.T. quotations in, 258
Apocalypse, uncial MSS. of, 57, 122 ;
originally absent from Coptic ver-
sions, 185, I91
Apocalypse of Peter, 98

371

Apostoli (or Praxapostoli), 128

Arabic versions, 195 f.

Aristion, the elder, named as author
of Mark xvi. 9-20, 173

Armenian version, 171 ff., 334, 359;
suggested connexion of, with Eu-
thalius, 106

Athanasius, St., on canon of N.T.,
81 note; quotations in, 260

a-type of N.T. text, 298 ff, 315-
327; authorities for, 326

Augustine, St., on Old Latin Bible,
210, 213; quotations in, 261, 262

Autographs of N.T. books, appear-
ance of, 25 ff.

Baber, ΗΣ Εἰ 75

Balestri, edition of Sahidic Μ55., 186

Balg, G. H., edition of Gothic version
by, 241

Barnabas, Epistle of, 62, 63, 208,

247

Barnard, P. M., on N.T. text of
Clement of Alexandria, 251, 339

Barrett, J.,. 116

Bartolocci, G., 78

Bashmuric version, 180

Basil, St., of Caesarea, 259

Batiffol, Mgr. P., 120, 122, 204, 205

Bede, use of E, by, 101

Belsheim, J., edits Evan. 565, 138;
edits Old Latin texts, 200-208

Bengel, J. A., 14, 115; edition of
Nii by; 278

Bentley, R., 86, 103, 104, 211;
edition of N.T. prepared by, 222,

372

276-278; MSS. collated by, 227,
228

Berger, S., on Latin versions, 207,
208, 218 ; on Vulgate MSS., 229 ff.

Bernard, J. H., on Old Syriac, 159
note

Beza, T., owner of Codd. D,D,, 89,
97; editions of N.T. by, 271

Bianchini, G. M., editions of Old
Latin MSS. by, 200, 201, 203;
of Vulgate MSS., 230, 232, 233

Birch, A., collation of MSS. by, 78,
114; edition of N.T. by, 283

Blass, F., his theory of the 6-text,
33) 341-354

Boerner, C. F., 104

Bohairic version, 180 ff., 332

Books, circulation of, in early days of
Christianity, 35-37, 355

Bousset, 65 note, 331 note, 365

Brightman, F. E., on Cod. Bezae,
92; on the term Lvangeliarium,
128 note

B-type of N.T. text, 59, 298 ff., 327-
334; authorities for, 332-334

Buchanan, E. S., on Latin MSS.,
200, 201, 207, 237

Budge, E. A. W., 186 note

Burgon, J. B., index (Ὁ patristic
quotations, 243 note, 264; works
on textual criticism, 307; his
textual theory examined, 315-326

Burkitt; F..C.,00 A, γα; on Ὁ, οἱ;
on Diatessaron, 149; on Sinaitic
Syriac, 152 ff.; on N.T. quotations
in Ephraem Syrus, 162; on origin
of Peshitto, 163; on Palestinian
Syriac, 169; on Arabic versions,
195; on Codex Bobiensis, 203 note ;
on Latin versions, 213 ff. ; on St.
Augustine, 262; on Tyconius, 263

Caesarea, suggested as home of x
and B, 70, 82-85 ; library of, 258 ;
connexion of B-text with, 329

Cambridge Greek Testament, 310

Castiglione, C.O,, 241

Catenas, in N.T. MSS., 115, 118

Ceolfrid, causes Cod. Amiatinus to
be written, 225

Ceriani, A. M., 76, 206, 336

Chapman, J., 226, 227

Chase, ἘΣ H:,.on Cod. Gezae,:.94%
on Syriac origin of 6-text, 359

CRITICISM OF NEW TESTAMENT

Chrysostom, N.T. text used by, 259

Ciasca, A., 149, 186

Classical authors, texts and MSS. of,
compared with those of N.T., 4,
5» 319

Classification, of MSS., 304 ff.

Clement of Alexandria, N.T. text
used by, 250, 251, 333, 337) 339

Clement of Rome, Epistles of, 72,
74, 246

Clement VIII.,
Vulgate by, 221

Clementine Homilies, 248

Codex-form of book, introduction of,
38-40

Codex: for list of codices, see Manu-
script

Colometry, 90, 101

Complete. copies) of. Ν᾿ 7. 46,057.
132

Complutensian Polyglott, 267

Conflate readings, 301, 302

Conjecture, place of, in
criticism of N.T., 16

Constantine, emperor, orders 50 MSS.
for his capital, 47, 83

Conybeare, F. C., on Armenian ver-
sion, 171 ff. ; on Georgian version,
174

Coptic versions, 176 ff.

Coptic writing, origin of, 177

Corinthians, third Epistle to, 172,
174

Gornill, (Ο: B.,-76.4936

Corssen, P., on Graeco-Latin MSS.,
99; on Latin Bible, 236

Cowper, B. Ἡ:, 78

Cozza-Luzi, G., 79, 111 note

Cronin, H. S., edition of Cod. N.,
tog; collation of Evan. 565, 138

Crum, W. E., on Coptic versions,
187, 193, 194

Cureton, W., discovers Old Syriac
version, ΤΟΣ

Cursive writing, 125

Curzon Catena, the, 183

Cyprian, Latin Bible used by, 203,
207, 210, 255

Cyril, of Jerusalem, 259, 366

pope, edition of

textual

Damasus, pope, 216

Dates of uncial MSS., 58, 66-69
Deane, H., 167

De Dieu, L., 167

INDEX I

Delaporte, 187

De Rossi, G. B., 225

Diatessaron, the, of Tatian,
B51; 367

Dillmann, on Ethiopic version, 194

Dionysius Alexandrinus, 256

Divisions of text in Gospels, 68, 80,
119; in Acts and Catholic Epistles,
81, 82, 226; in Pauline Epistles,
81, 263

Dobrowski, J., 231

§-type of text, examples of, 93, 153-
157; examination of, 299 ff., 338-
363; authorities for, 358-360

148-

Egypt, suggested home of x, 70; of
A, 76; of B, 84; of B-text, 329;
of y-text, 334

Eichenfeld, J. von, 207

Elzevirs, editions of N.T. by the,
271

Ephraem of Syria, 87, 148, 149;
N.T., text used by, 159 note, 162,
172, 258

Epiphanius, 260

Epistles, Catholic, complete uncial
MSS. of, 57

Epistles, Pauline,
MSS. of, 57

Erasmus, editio princeps of Greek
N.T. by, 268 ; later editions, 270;
MSS. used by, 78, 130, 131,
269 f.

Errors of transcription, classes of, 7 ff.

Estienne, R. See Stephanus

Ethiopic version, 194

European family of Old Latin version,
51 ff.

Eusebius, of Caesarea, 47, 68, 257,
366 ; canons and sections of, 68
Euthalius, edition of Acts and Epistles

by, 82, 106, 259; MSS. containing
his apparatus, 106, 135, 140, 333
Evangeliaria, 128, 141

complete uncial

Fayyumic version, 180, 193

Fell, J., edition of N.T. by, 274

Ferrar, W. H., 131

Ferrar group of MSS., 131, 228, 358

Fleck, F. F., 203, 226

Freer, C. L., MSS. belonging to,
107, 115

Froben, publisher of first Greek
N.T., 268

3.1.2

Gabelentz and Loebe, edition of
Gothic version by, 241

Gebhardt, O. von, on date of Cod. B,
ὅτ: on. Cod, ὧν 19

Georgian version, 174

Gibson, Mrs., associated in discovery
of Syriac MSS., 152, 168 f.; pub-
lication of Arabic texts by, 196

Giorgi, A. A., 114, 193, 284

Gospels, complete uncial MSS. of,
57, 107; order of, in Western
church, 90

Gothic version, 240, 360

Goussen, H., 187

Grabe, J. E., 73

Graeco-Coptic bilingual MSS., 114,
188, 193, 335) 337

Gregory, C. R., author of Prole-
gomena to Tischendorf's N.T.,
etc., passim

Gregory, of Nazianzus, 259

Gregory, of Nyssa, 259

Gregory Thaumaturgus, 256

Gregory the Illuminator, 172

Grenfell, B. P., 41-44

Griesbach, J. J., editions of «Na
by, 108, 281

y-type of N.T. text, 298 ff., 334-338

Guidi, I., on Ethiopic version, 194

Gwilliam, G. H., 160, 161

Gwynn, J., 56; on Philoxenian ver-
sion, 165

Haase, H. F., 204

Hackspill, L., on Ethiopic version,
194, 195

Hagen, H., 206

Hansell, E. H., rot

Harkleian Syriac version, 164 ff., 359

Harmonistic errors, 10

Harnack, A., 119, 246 note

Harris, J. Rendel, on N.T. auto-
graphs, 34; on Cod. D, 90, 94;
on stichometry, 98; on Cod. A,
117; on the Ferrar group, 132;
on Syriac versions, 152, 168; on
Arabic versions, 196

Hartmut, abbot of St. Gall, 237

Haseloff, A., on Cod. 2, 119

Hearne, T., editor of Cod. Ἐν, του,
283

Hentenius, John, edition of Vulgate
by, 220

Heringa, J., 102

374 CRITICISM OF NEW TESTAMENT

Hermas, Shepherd of, 62, 63, 81,
98, 247

Hesychius, edition of Septuagint, 76;
text of N.T. attributed to, 331, 365

Hilary, of Poitiers, 260

Hippolytus, of Rome, 255

Homoioteleuton, 9

Horner, G., on Bohairic version,
182 ff. ; on Sahidic, 187 ff.

Hort, F. J. A., his introduction to
the N.T., 225, 294-309, etc.

Hoskier, H. C., collation of Cod.
700, 139; on Latin Bible, 234;
on textual history, 356 note

Hug, L., 78, 80

Hunt, A. S., 41-44

Hutton, E. A., 131, 302

Ignatius, Epistles of, 247

Irenaeus, N.T. text used by, e212,
249

Irish family of Vulgate MSS., 238

Itala, supposed variety of Old Latin
version, 210 ff.

James, M. Ἐς; on ‘Coed. 65, 1325

James, St., Epistle of, not recog-
nised in Latin Church before fourth
century, 208

Jerome, St., 48, 49, 216 ff., 261

John, St., Blass’ theory on the 6-text
of the Gospel of, 353

Justin Martyr, quotations in, 247

Karkaphensian Syriac version, 170

Kells, Book of, 232

Kennedy, H. A. A., on Old Latin
version, 199 note

Kenyon, Ἐς Οὔ, τὸ, 25; 193, 39395

Kipling, T., edition of Cod. Bezae
by, 89, 284

Knittel, F. A., MSS. edited by, 112,
209, 241, 284

Krall; \J., 193

Kuster, L., edition of N.T. by, 275

κῶλα, 9O, IOI

Lachmann, K., editions of N.T. by,
204, 222, 229, 286-288

Lactantius, 260

Lagarde, P. de, 168, 182

Lake, K., facsimile of Cod. x, 63;
on MSS. at Mt. Athos, r1o note;
on Codd. 1-118-131-209, 130; on

Ferrar MSS. , 135, 139; on Cod. Ψ,
121; on Cod. 1071, 139

{.angton, Archbishop Stephen, author
of division of Bible into chapters,
219

lLaodiceans, apocryphal epistle to the,
226, 237

Latin versions, 196-240

Laud, Archbishop, owned Cod. E,,
IOI

Law, T. G., 201

Leclerc, Jean, 342

Lectionaries, 128, 141

Lewis, Mrs., discovers Sinaitic Syriac
MS., 152; discovers Palestinian
Syriac MSS., 168, 169

Lightfoot, Bishop, on Western text
of Acts, 342

Lindisfarne Gospels, 234

Logia Jesu, papyrus fragments of,
21, 39

Lucar, Cyril, 72, 75

Lucian, edition of O.T. by, 303,
324, 366; a-text of N.T. attri-
buted to, zd,

Lucifer, of Cagliari, 99, 212, 260

Luke, St., Blass’ theory with regard
to his works, 341 ff. ; Western or
d-text of his Gospel, 346-349

Macarius Magnus, 260

Macdonald, Major, 61

Mai, Cardinal, 78, 203, 204, 241

Manuscripts, Biblical, catalogue of—
[N.B.—Numbers in heavier type
indicate the place at which the
principal description of the MS.
is given. |
Codex Alexandrinus (A), 58, 67,

68, 72-77, 326, 332, 337, 358

— — Amiatinus (A), 225

Argenteus (Gothic), 240

——- Augiensis (F,), 103, 360

—— Augiensis (7), 209

——— Bambergensis (B,), 235

—— Basiliensis (E), 100, 326

—— Beneventanus (*), 227

Beratinus (®), 120, 326

—— Bezae (D), 88-97, 338 ff., 358

—— Bezae (d@), 200

—— Bibliothecae Angelicae (L,),
108, 326, 333

—— Bigotianus (B), 227

—— Bobiensis (2), 203

INDEX I

Codex Bobiensis (s), 207
Bodleianus (x), 208

—— Bodleianus (x), 209
Boernerianus (G,), 104, 360

-—— Boernerianus (yv), 209

Boreeli (F), 102, 326

Borgianus (T), 114, 190,

332, 337

Brixianus (7), 201, 215

Campianus (M), 108, 326

Cavensis (C), 227

—— Claromontanus (D,), 97-100,

360

—— Claromontanus (4), 209

——— Claromontanus (2), 202

Coislinianus 202 (Hs), 105,

3

Colbertinus (c), 200

Corbeiensis (72), 208

Corbeiensis I. (#1), 201

Corbeiensis II. (773), 201

Corporis Christi Cantabrigi-

ensis (X), 233

Cyprius (K), 107, 326

—— Dublinensis (D), 227

Dublinensis (Z), 116, 332

Dunelmensis (A), 228

Egertonensis (E), 228

Ephraemi (C), 67,

326, 332, 333, 337, 358

Epternacensis (P), 228

Forojuliensis (J), 230

Friderico - Augustanus (οἵ

OrT.), ‘Gz

Frisingensis (7), 209

Fuldensis (F), 228

—— Gigas (9), 206

—— Gottvicensis (77), 209

Guelpherbytanus A (P), 112,
332

— Guelpherbytanus B (Q), 113,
332

Guelpherbytanus (gwe), 209

—— Harleianus (Z), 234

—— Harleianus II. (Z,), 236

— Hubertianus (H), 229

Ingolstadiensis (1), 230

—— Juueniani Vallicellianus (1),
235

Karolinus (K), 231

—— Kenanensis (Q), 232

Laudianus (E,), 67, 100-102,

85-88,

Laudianus (e), 206

3.1.5

Codex Lemovicensis (1.94), 235

-—— Lichfeldensis (L), 231

——— Lindisfarnensis (Y), 234

—— Lombardicus (L,), 235

—— Macedonianus (Y), 116

—— Martini-Turonensis (M), 231

—— Mediolanensis (M), 231

— Monacensis (M,), 236

—— Monacensis (X), 115

—- Monacensis (Φ), 205

—— Mosquensis (K,), 107

—— Mosquensis (V), 115, 326

—— Mutinensis (H4), 105, 326

—— Nanianus (U), 114, 326

——- Nitriensis (R), 113, 332

—— Oxoniensis (O), 231

—— Oxoniensis II. (O,), 236

—— Oxoniensis III. (O,), 236

-—— Palatinus (e), 200

Patiriensis (048), 122

Perusinus (P), 232

Petropolitanus (II), 119, 326

Porphyrianus (P,), 112, 326,

327

Purpureus Petropolitanus(N),

49, 109, 326

Regius(L), 108, 332, 337, 358

Regius (R,), 236

Rehdigeranus (/), 204

de Rosas (R,), 236

Rossanensis (2), 119, 326

— Rushworthianus (R), 232

Sangallensis (A), 104, 117,

2

Sangallensis (6), 206

Sangallensis (S,), 236

Sangermanensis (Es), 102,

360

Sangermanensis (e), 209

Sangermanensis (G), 229

Sangermanensis I. (91), 202

Sangermanensis II. (7), 202

Saretianus (7), 203

Sinaiticus (x), 57, 58, 60-72,
332, 337

——— Sinopensis (O), 111

— Stonyhurstensis (5), 232

—— Theodulfianus (©), 230

—— Tischendorfianus III. (A),
127; 332

—— Tischendorfianus IV. (ΓΤ), 116

—— Toletanus (T), 232

—— Ulmensis (U,), 236

—— Usserianus I. (77), 205

370. CRITICISM GF NEW TESTAMENT

Codex Usserianus II. (77), 205
Vallicellianus (V), 233
Vaticanus 1209 (B), 58, 67,
69, 70, 77-85, 327 ff. ; character
of text in 9; 1:,,853

Vaticanus 2066 (046), 122
Vaticanus 354 (S), 114, 326
—— Vercellensis (az), 200

—— Veronensis (4), 200

—— Vindobonensis (z), 203

-—— Willelmi de Hales (W), 233
— Wolfii A (G), 103

— Wolfii B (H), 105

—— Zacynthius (=), 118, 332,

Q, 121

Fragmenta Ambrosiana (s), 206

Bernensia (2), 206

—— Monacensia (¢), 208

Sangallensia (7), 204

Ultratraiectina (U), 233

Fragmentum Mediolanense (53),
206

Vindobonense (V), 206
Palimpsestus Floriacensis (2), 206

Manuscripts, numeration of, in N.T.,
51 ff., 126; approximate number
of, 129; corruption of, 7-15, 355
note

Marcion, edition of Luke and Pauline
Epistles, 248

Margoliouth, G., 169

Mark, St., Blass’ theory on the 6-
text of the Gospel of, 353

Marshall, T., 182

Martianay, J.. 201, 208, 221 note

Martin, Abbé, on the Ferrar group,
131

Maspero, G., 186 note, 193

Massmann, J. F., 241

Matthaei, C. F., MSS. collated by,
To2, 104, 108% edition of N.T.,
282, 284

Matthew, St., Blass’ theory on the
6-text of the Gospel of, 353

Mazarin Bible, 219, 267

Memphitic version, 181, 193. See
also Bohairic

Mesrop, St., 171

Methodius, 257

Meyncke, Dr., 236

Mico, Abbé, 78

Middle Egyptian versions,

192 ff., 335

Mill, J., edition of N.T. by, 73, 274

Miller, E., edition of Scrivener’s
Introduction, 45, 308; edition of
Gospels, 244 note; works on
textual criticism, 308 note;
examination of his textual theory,
315-323

Miniscalchi-Erizzo, Count, 168

Minuscule MSS., notation of, 126;
numbers of, in N.T., 128; com-
plete copies of N.T., 132

Minuscule writing, 124 ff.

Moesinger, G., 148

Morgan, J. P., M.S. belonging to, 234

Miintz, E., and Fabre, P., 77

181,

Nestle, E., 17, 45, 91; edition of

Vulgate N.T., 223; of Greek
N., 312

‘‘Neutral”’ type of text, 299. See
also B-type

Nitria, MSS: from, £13, 15%)" ΤΌΣ,
168

Northumbrian family of Vulgate
MSS., 237

Numeration of MSS., systems of,

50-56

Old Latin version, 197-216, 359

Old Syriac version, 151-158, 359

Omont, H., 106, 112

Opisthograph books, 23

Origen, Hexapla edition of Septua-
gint. by; 65, Ὁ. ΤΟΣ; 299% (Nid.
text used by, 251-254, 320,''330,
333) 337, 340, 367

Oxford Greek Testament, with the
Revisers’ readings, 310, 311; with
apparatus by Souter, 312

Pachomius, St., 178

Palaeography, periods of, 19

Palestinian Syriac version, 168 ff.,
284

Palimpsest MSS. of N.T., 85, 112,
L13; 126, ΤΙ, 122, 782,599 590,
207, 209, 241; meaning of term,
85

Palomares, C., 233

Pamphilus, library of, at Caesarea,
65, 253, 258; references to MS.
of, 65,,. 82; τοῦ, 1324, 1590) 535

INDEX I

Paper, invention of, 19, 125

Papyrus, how manufactured, 22 ff. ;
method of writing on, 23 ff.; liter-
ary and non-literary hands on,
23, 24; papyrus codices, 39, 47;
Biblical fragments on papyrus, 20,
41-44

Patristic quotations, Miller’s analysis
of, 321-323

Paul, Acts of, 98

Pauline Epistles, divisions of text in,
81, 106, 263

Persian versions, 175

Peshitto Syriac version, 158 ff., 195 ;
relation to a-type of text, 164, 304,
324

Peter, Apocalypse of, 98

Philoxenian Syriac version, 164 ff.,
196

Pococke, R., 167

Polycarp, Epistle of, 247

Praxapostoli, 128

Primasius, commentary on Apoca-
lypse by, 210, 212, 263

Priscillian, 212, 263

Punctuation, in papyri, 24

Purple vellum, MSS. on, 109, 111,
110, 120, 200, 201, 203, 234,
240

Quentin, H., on history of Cod.
Bezae, 89

Rabbula, Bishop, probable editor of
Peshitto, 163

Rahlfs, A.,i on date ofCod. -B,
81

Ramsay, W. M., on 6-text of Acts,
33, 357 note

Ranke, E., 204, 229

Rettig, H. C. M., 117

Révised Version of N.T., 306

Ridley, G., 167

Robinson, Dean
Euthalius, 82, 85, 107; on Old
Syriac version, 159 note; on
Armenian version, 171

Robinson, Forbes, on Coptic versions,
178, I91

Rolls, length of, in papyrus MSS.,
34 ff.

Roman Church, Greek character of,
in first two centuries, 197

Armitage, on

S77

Rome, suggested as place of origin
of B, 82; suggested association of
6-text with, 341, 354

Ronsch, H., 209, 254

Rulotta, Abbé, 78

Saba,
195

Sabatier, P., on Latin Bible, 200,
207, 206,°231

Sahak, St., 175

Sahidic version, 180, 185 ff., 332,
359; order of Pauline Epistles
in, 84

Salmon, G., 33, 245; 329) 334

Sanday, W., 92, 203, 207; on Old
Latin version, 211; appendices
to Oxford Greek Testament, 311

Sardinia, home of E,, 1o1; possible
home of D, 92

Scheibel, J. E., 204

Scheil, V., 42

Schepss, G., 263

Schmidtke, A., 81, 138

Scholz, J: M.. A., 78, Τοῦ; 209;
127; catalogue of MSS. by, 127,
284

Schwartze, M. G., 182

Serivener, F. H. A., 89, 17, δ
passim ; works on textual criticism,
293

Section - divisions in Gospels, 68,
80, 81: in Acts, 81, 82, 226; in
Epistles, 81, 106

Seidel, A. E., 103

Semler, J. S., classification of MSS.
by, 280

Sinai, Mt., MSS. from, 60, 152, 168,
196

Sinaitic Syriac MS., 152 ff.

Sixtus V., edition of Vulgate by,
220

Skeat, W. W., 232, 234

Skins, used for writing in Palestine,
25

Smith, W. B., on Cod. Ge, 105

Soden, H. von, system of numeration
of MSS., 52-55; on Diatessaron,
149, 151; his textual theory, 363-
369

Solomon, Psalms of, 74

Sophocles, MSS. of, compared with
those of N.T., 319

Souter, A., edition of N.T., 57, 312;

Mar, monastery of, 107,

378 CRITICISM OF NEW TESTAMENT

on Latin text used by Jerome,
215; on Latin text of D, 99

Spanish family of Vulgate MSS.,
238

Speculum, pseudo-Augustinian, 204

Spiro; Dr. ἘΠ 77

Stephanus, R., 89, 108; editions of
N.T. by, 219, 270

Stichometry, 98 note

Streane, A. W., 234, 292

Streitberg, W., edition of Gothic
version, 241

Stunica, L. de, editor of Com-
plutensian N.T., 268

Syriac language, 147

Syriac Versions, 147-171

‘“Syrian”’. type of text, 299, 303.
See also a-type

Tatian, Diatessaron of, 148 ff., 367 f.;
WN... text of, 278

Tattam, Archdeacon, secures MSS.
from Nitria, 151

Teaching of the Twelve Apostles,
247

Tertullian,
254

Text of N.T., families of, 58-70,
298

Textual criticism of N.T. compared
with other books, 3 ff. ; methods
of, 7 ff. ; ancient textual criticism,
265

Textus Receptus, 59; origin of the
term, 272 note

Thebaic version. See Sahidic

Thecla ** the martyr,” 75, 76

Theodore, of Mopsuestia, 259

Theodore of Tarsus, MSS. brought
to England by, ΤΟΙ, 225, 234

Theodulf, revised edition of Vulgate
by, 218, 220, 230, 238

Thompson, Sir E, Maunde, 19, 25,
45» 47, 50, 73

Tischendorf, L. C. F., 60 and passim ;
texts and collations published by,
288-290

Vodd, 7 Ἢ 20r

‘«Traditional’’ text, Burgon and
Miller’s theory of, 315 ff.

Tregelles, 5. P., 78, 100-119 passzm,
226, collations and editions by,
291-293

{πὶ} RK. 185

N.T. quotations in,

Turner, GC, Hi, 263, 220, 2:0, 256,
314
Tyconius, 210, 212, 262

Ulfilas, author of Gothic version, 240

Uncial writing, meaning of, 49;
history of, 49 ff.

Uppstrém, A., 241

Uspensky, Bishop Porphyry, 43, 61,
63, 137

Vallarsi, D., 221 note

Variorum Bible, 311

Vellum, introduction of, 46

Vercellone, C., 77, 79, 221 note

Versions, use of, 146

Victor of Capua, edits Diatessaron,
149 note, 228

Villa, F., 231

Vulgate, the, 216-240, 332

Walker, J., assistant of Bentley, 222,
277, 278; collations of Vulgate
MSS. by, 227 ff.

Walton, B., Polyglott Bible of, 73,
274

Weihrich, F., 204

Weiss, B., examination of MSS. of
ΝΥ. by; 308-310; ‘an’ Cod. B,
328, 329; on Cod, D, 90; 94, 95,
359

Wells, E., edition of N.T. by, 276

Wessely, C., 187, 193

Westcott, Bishop B. F., on Latin
MSS., 208, 209, 237; edition of
N.T., with Hort, 294-307

‘‘ Western ” type of text, 209 ff. See
also 6-type

Wetstein, J. J., system of numeration
of MSS., 51; collations by, 86,
97; 100, 103, 108, 100; edition
of N.T. by, 279

Weymouth, R. F., edition of N.T.
by, 312

Wheelocke, A., 175

White, H. J., ror; on Old Latin
Versions, 201, 203-207; on Six-
tine Vulgate, 220; on Vulgate
MSS., 224 ff. ; edition of Vulgate
NAL 255

White, J., 167

Widmanstadt, A., 159

Wilkins, D., 182

INDEX I

Woide, C. G., 73, 185, 284

Wolf, J: C., 103, 105

Wordsworth, Bishop J., Old Latin
texts edited by, 201, 202, 203, 206,
208; edition of Vulgate, with H. J.
White, 223 ff.

Ximenes, Cardinal, promoter of Com-
plutensian Polyglott, 219, 267

379

Young, Patrick, 73, 241

Youngman, G. M., Vulgate MSS.
collated by, 228, 229, 231-
233

Zacagni, 107, 135
Ziegler, L., 208
Zoega, G., 186, 193

PASSAGES OF WHICH THE

Matt.

”

᾽»

i. 8

1.26 132 note,

ΣΨΠ 2,3
xvii. 21
ΧΡ]. 11
xix. 16
xix; 17
xX. 22, 25
xx. 28.
Xxl. 44
Xxlil. 38
xxiv. 36
XXVii. 16
XXVil. 49

71,
71,

93, 120,

1

Vi: Go = ‘
ix. 44, 46, 49
xiii. II ϊ
xv. 28. y
XVi. 9-20 72, 108,
138, 155, 173, 184,

71,

Bae 1
iv. 18
VM. GeLr:
v. 26

Vv. 39.
1. Ὁ

INDEX

II

VARIOUS READINGS ARE

NOTICED
PAGE PAGE
«| X55 | Luke vi. Ὁ 93, 156, 348
136, 154 ΙΧ ΕΣ . 156; 348
71. 1ῈῈ jee A 2.2 156, 348
71, T55 ” ΣΙ: 2-4. 71, 156, 348
ΠῚ; 255 Beye <) Cane: aaa 348
71, 255 »» XIX. 31-35 : . 348
oy oes jy ah. 17-20 93, 156, 349
70, 255 fo RL, 8: Fi, TO 272)
155, 323 Ἢ 174, 184, 349
155, 323 1, ΧΧΠῚ. 34 71, 156, 349
Sg tc "ἌΧΗ 245 : ee
ΠΕ 4 = >» KI. 248 τ VESG
LSS 1» XXIL. 53 349, 356
55; 352 , ee Oe . | 346
| aye iin) RIV, Ge: 156, 349
nee , URXIV, ΤῸ ; 93, 156
71, 155 » Χχίν. 36 93, 156, 349
155, 246 1» XXIV. 40 93, 156, 323, 349
: 71 ,», RY, 45, 27 ἘΠ τ
»» XXIV. 51 71, 93, 154, 349
71 53) SERS 157, 302

ΣΙ

r55, 302 | John i. 28 71
302 note fig ΣΕ
ΠΟΥ sy hile 5 ἼΣ, ΤΠ
ΤΩΙ, 136 A. (ith. 20; (en ; 9
204, 323 » Ive 9 157, 352
» Ve 8, 4 157, 185
71; 756 53 wi, 60 72, 157
5 eee ») Vil. SQ-Vill. ΤΊ 72, 74, 132,
347 139, 157, 174, 185, 349
348 sy Wil. 59.4 : ; ΕΠ
348 1, XL. 39 157
71 jy) OMe EG 9

380

XViil. 13-24
ΧΧῚ. a5.

i. 26

ae 14

ii. 29
111, IO
iii. 26
iv. 37
v. 29

vi. 8

vi. 15
Will. 8s
viii. 24 .
X. 25
xi, 2
Kil. TO).
Sil, I4 .
xiv. 2
Xiv. 5
xiv. 20.
FV. Οἱ.
XV. 29
Vil? .
sV1,.25.
Zvi. 95,-
XVI 39)

INDEX II 381

PAGE PAGE
. 3g ἈΓΙ5 xvi. 25. : , . 544
72, 323 ΧΙ 2 - ‘ : ποῦ
»» XVill, 25 : » 76 note
96 og. EVI. 27 Σ : +) 315
95 a xix. Bite : : - 345
95 SRS Sb tao ae ; F ESAS
95 ae ee δ ς ; : Mie
96 ΠΣ 17 ς : ; 3 95
96 ἘΠ ΈΧΧΙΣ eee : : . ‘Os
« 34g os xxi, τό. : Ξ . 345
302 ja) ER. 24, ; : 3. 545
95 5) SKIL. 26:30 : : > Sen
102 ae eelVa 24, 25 "- ‘ . 946
343 jy MEVIL) E : : ᾿ 5310
343 ἢ SELL, 5 : : 520
344 iS) ΧΧυ BIR τοι. : - 346
244 Jk XXVill,) τὸ : : 4 | 346
96 »» SXVIL τὸ : ὃ . 346
344 A eevee : Ε . 346
344
ΟΣ ἈΠ ον, ον τ...
95
see τ Johnv. 7,8 133; 138, 204, 208,
95 270 note
344
a44 \ Rev. xxii, 16-25: : . wgo
THE END

Printed by R. & R, Cuark, LIMITED, Edinburgh.

HANDBOOK

TO THE

TEA TUAL CRIPICISM ΟΕ [Ee
NEW TESTAMENT

BY

ork FPREDERIC G KENYON, ΚΟ Β ἘΡΑ:

SOME PRESS OPINIONS OF THE FIRST EDITION

JOURNAL OF THEOLOGICAL STUDIES.—“ It will at
once take its place as the best of its kind by an Englishman. It
is beautifully written and eminently fair.”

CRITICAL REVIEW.—“ From Dr. Kenyon it is natural to
expect the best style of work in his own special department. We
get that in this handbook, which is not less handsome in form
than rich and reliable in its matter. It makes an important and
seasonable addition to the number of books to which the English
student of the textual criticism of the New Testament can turn
with confidence. . . . It gives all that the student requires, and
does that in an admirably clear and telling way. All that is of
real value in the description, valuation, and history of the manu-
scripts and versions is placed at our disposal. . . . Will be

greatly valued by all interested in this fundamental department of
New Testament study.”

CHURCH QUARTERLY REVIEW.—“ This book will
immediately take its place as the best of its kind by an English
writer.”

ATHENA UM.—“ The handbook deserves high praise. . . .
It states clearly and concisely what ought to be set down in such
a book.”

OXFORD MAGAZINE.—“ There is no better book on the
subject of the same length in English.”

MACMILLAN AND CO., Ltp., LONDON.

GUARDIAN.—“ Those who are acquainted with what Dr.
Kenyon has already produced in this field will know that they can
count upon finding two things in this volume—adequate know-
ledge and clear statement of results. They will also find fair and
well-balanced statement of results. . . . It is to be hoped that this
handbook will be widely read. It contains a clear account of the
evidence which exists, and of the interpretation put upon it thus
far, and it enables the student to understand, and perhaps pro-
mote, further developments.”

Ready Autumn 1912.

8vo.

THE NEW TESTAMENT DOCUMENTS
THEIR ORIGIN AND EARLY HISTORY

BEING THE CROALL LECTURES FOR 1911

BY

GEORGE MILLIGAN, D.D.

PROFESSOR OF DIVINITY AND BIBLICAL CRITICISM IN THE
UNIVERSITY OF GLASGOW

The object of these Lectures is to trace the history of the
documents as far as is now possible from the time of the Apostolic
autographs to the formation of the Canon about A.D. 400. The
first lecture deals with the outward appearance of the original
manuscripts ; the second is devoted to the language of the Greek
New Testament ; the third and fourth deal with the literary form
and structure of the various books ; the fifth traces the circulation
of the New Testament writings from the papyrus roll stage ; and
in the concluding lecture an account is given. of the gradual
collection of these scattered writings into what we now know as
the New Testament. The volume is illustrated by facsimiles of
the most important papyrus and other texts,

MACMILLAN AND CO., Ltp., LONDON.

:
t

.