DIVINITY SCHOOL
TROWBRIDGE LIBRARY

GIFT OF
Estate of the
Reverend Raymond Gilmore Clapjj

THE
SYRO-LATIN TEXT

OF

THE GOSPELS.

THE
SYRO-LATIN TEXT

OF

THE GOSPELS

BY

FREDERIC HENRY CHASE, D.D.
LECTURER IN THEOLOGY AT CHRIST'S COLLEGE,
PRINCIPAL OF THE CLERGY TRAINING SCHOOL, CAMBRIDGE,
AND EXAMINING CHAPLAIN TO THE LORD ARCHBISHOP OF YORK.

Honion :
MACMILLAN AND CO.
AND NEW YORK,
189s
[A/l Rights reserved.'^

C38fes

PRINTED BY J. & C. F. CLAY,
AT THE UNIVERSITY PRESS.

o Q)i- o
J
(} -d

Cainbrilige : ^

-r) I
I

l I

J) o

TO THE
RIGHT REVEREND FREDERIC WALLIS D,D.
BISHOP OF WELLINGTON NEW ZEALAND .
FELLOW OF GONVILLE AND CAIUS COLLEGE
THIS ESSAY IS DEDICATED
WITH GRATEFUL AND TRUE AFFECTION
BY ONE
OF HIS MANY FRIENDS AND FELLOW-WORKERS
IN CAMBRIDGE

PREFACE,
THE present volume is the sequel of an Essay which I
published two years ago on the Old Syriac Element in
the text of Codex Bezae. The latter, primarily an offshoot of
a larger work on the Acts on which I am engaged, dealt with
the Bezan text of that Book. Several critics, whose opinion
I respect, urged against my conclusions the not unnatural
objection, which I had fully anticipated in the Preface, that I
could produce no direct evidence for an old Syriac text of
the Acts, Convinced that assimilation to Old Syriac texts
was a predominant factor in the genesis of the Bezan and of
cognate texts, I felt that it was almost a matter of honour to
extend the investigation to the Gospels, where ample evidence
for Old Syriac readings is supplied by the Sinaitic and Cure-
tonian MSS., by the Arabic Tatian, by Ephrem's Commentary
on the Diatessaron, and by Aphraat's quotations.
The rough draft of this Essay was drawn up before the
publication of the Sinaitic Palimpsest in October 1894. Since
that time the whole has been re-written. One note however —
that on Luke ii. 5 (see p. 28 f.) — I have ventured to leave
exactly as it stood before I saw the Sinaitic text, appending
a statement of the evidence derived from that MS., because,
as a concrete example, it seems to me to indicate how far
results obtained by a critical process are likely to be right.
I may be allowed to add that again and again I have found
my conclusions confirmed by the Sinaitic text.
An apology is perhaps required for the title of this
volume. The term, the ' Western ' text, is generally allowed

Vlii PREFACE.
to be misleading. ' The time is, we hope, not far distant,' if I
may quote and adopt as my own some words of the Rev.
H. Lucas, S.J. {Dublin Review, July 1894, p. 52), 'when the
term " Western " will, for the future, give place to the term
" Syro-Latin," the only one which truly represents, in our
opinion, the facts of the case.'
In my references to Tatian's Diatessaron T have generally
used the convenient volume of Mr J. Hamlyn Hill, which
bears the somewhat quaint title ' The Earliest Life of Christ
ever compiled from the Four Gospels! In the parts of this
book which have been of service to me Mr Hill has secured
the cooperation of other scholars. Mr G. Buchanan Gray,
B.A., of Oxford, collated with the Arabic text Mr Hill's
English rendering of Ciasca's Latin, while Professor Armitage
Robinson is responsible for the English translation of the
evangelical quotations in the Armenian version of Ephrem's
Commentary. In regard to the Latin texts of the Gospels, I
have made continual use of the Oxford edition of the
Vulgate, for which scholars owe a great debt to the Bishop
of Salisbury and Mr H. J. White. The volume however
containing St John's Gospel did not appear until the pages of
Chapter I., which deal with that Gospel, had passed out
of my hands. The third volume of Dr Resch's Aussercanon-
ische Paralleltexte — Paralleltexte zu Lucas — reached me too
late for me to make any use of its rich stores of Patristic
citations. It only remains for me to express my sincere gratitude to
several friends in Cambridge, and especially to a younger
friend, Mr F. Lillingston B!A., late Scholar of Pembroke
College, for their kindness in helping me in the correction of
proof-sheets. I wish also to thank the readers and workmen
of the University Press for the pains which they have be
stowed on the printing of the present volume and of its
predecessor. Cambridge. July, 1895.

TABLE OF CONTENTS.
PAGES
Preliminary Statement of Purpose and Plan. i — 2
I. Select Passages from St Matthew, St John, and
St Luse. 3—75
St Matthew, i. 16 [3f.]; x. 11 — 13 [4f.]; x. 42 [6]; xv. 26, xvi. 16
[7]; xvii. 27 [7f.]; xviii. 2, 20 [8]; xx. 28 [pff-]; xxi. 28ff. [15];
xxii. 34[i5f.]; xxiii. 9 [16]; xxv. 41 [16 ff.]; xxvi. 15 [18].
St John, iv. 42 [19]; vi. 17 [19 f.]; vi. 23 [2of.]; vi. 56 [21 f.]; viii.
53 [23]; xi. 9f. [23 f.]; xi. 14, 28, 35 [24f.]; xii. 32 [25]; xiii. 14
[25 f.]; xxi. 7 [26]; xxi. 13 [27].
St Luke, i. 79 [27]; ii. 5 [28 f.]; ii. 48 [29 f.]; iii. 10, 12, 14 [34f.];
V. 7f- [35f-]; ix- 16 [36 f.]; ^. 5 [37 f.]; xi. 528. [38 ff.]; xiii. 11
[41 f.]; xiii. 17 [42 ff.]; xiii. 24 f. [44]; xiv. 9 [45]; xv. 4 [46]; xv.
29f. [46 f.]; xvi. 31 [52]; xviii. 14 [525.]; xix. 4 [53f-]; xx. 34 [55];
xxii. 12 (Mc. xiv. 15) [56 f.]; xxiii. 36 f. [57f.]; xxiii. 40 ff. [58 ff.];
xxiii. S3 [62 ff.]; xxiv. 32f. [68 ff.]; xxiv. 37 [72 ff.].
Double renderings [30 ff.]; assimilation to the Old Testament
[47 ff.].
Notes: Lc. xxii. 27 [i4n.]; Mc. xii. 14 [i8n.]; Mr Rendel
Harris' note on Acts xv. 29 [70 n.].
2. Harmonistic Influence. 76 — 100
St Matthew, xxi. 18, xxiv. 31 f. [77]; xxvi. 59ff. [78ff.]; xxvii. 28
[81].
St Luke, iii. 23—38 [8if.]; iv. 31 [82 f.]; v. lof. {Ssf.]; v. nf. [84ff.];
vi. 42 [Sgf.]; viii. 35 [goff.]; xi. 2 [92 f.]; xx. 20 [gsf.]; xxi. 7
[94 f.]; xxiii. 45 ff. [95 f-] J xxiv. i. [96 f.].
St Mark, viii. 10 [97 f.]; xiii. 2 [98]; xv. 25 ff. [99 f.].
Notes: Confusion of Syriac suffixes 'your,' 'their' [86 n.]; Mr
Rendel Harris' note on Acts ii. 17 [86 n.]; An indication of an
farly Greek harmony [97 n.].

X CONTENTS. PAGES
3. Proper Names and Forms of Words. ioi — m
Prefixed |^[ioif.]; Iscariot [102 ff.]; termination-forms [104]; Gen-
nesar, Lazar [105 f.]; Caiphas [106]; Matt, xxvii. 46, Mc. xv. 34
[106 f.]; John i. 6 [107]; xi. 54 [108]; Luke xxiv. 13 [109];
Mark v. 41 [109 ff.].
4. Grammatical Points. 112 — 127
(i) The definite article [112 f.]. (2) Prepositions: repetition of a
preposition [113 f.]; the preposition 'upon' [ii4f.]. (3) Con
structions of the verb: resolution of participle in Bezan text [iisf.];
this due to Syriac, not Latin, influence as shewn by (i) examination
of e.g. Matt. ii. 7 — 12 (Syr. Lat.) [116 ff.], (ii) cases of partial
resolution in Bezan text (Gr. Lat.) [118 ff.]. Three other classes of
passages in D: (a) converse of resolution of participle [121];
[i] Syriacised indicative [i2if.]; (c) Syriacised participle (i22f.].
Verbal constructions — final, temporal, circumstantial — in Matt.
xxiii. 15, xxv. 10, xxvii. i, Mc. xiv. 55 [123 f.], in Jn. vi. 6r,
Lc. xxi. 36, Mc. v. 17 [124], in Mc.. vi. 48 [126]; insertion and
omission of ' /le (ihey) began' in Syriac texts and in D [124 ff.].
Summary of Facts and Conclusions. 128 — 142
Summary of facts [128 ff.]. Conclusions : (i) Date of Syro-Latin text
of Gospels [132 ff.]; (2) Its genesis (i) Bilingual texts [134 f.]; (ii) Bi
lingual scribes [135 f ] ; (iii) Interpolations [136 ff.]; (3) Its birthplace:
conditions of problem [138]; these satisfied by Antioch [138];
(i) its vigorous life [138 f.]; (ii) a bilingual city [139 f.] ; (iii) a
trading centre [140 ff.].
Notes : Dr Hort 's theory of ' Western non-interpolations ' [ 1 30 n. ] ;
Dr Blass' theory as to the Bezan text of Acts [133 n.]; Relation of
Old Latin texts to the birthplace of the Syro-Latin (Western)
text [141 n.].

Index I. MSB., Versions, Patristic Writings. 143 — 146
Index II. General. 147^ 148

THE SYRO-LATIN TEXT OF THE
GOSPELS.
It seems advisable briefly to state the purpose and the
plan of this essay.
The purpose of the investigation which follows is to gather
and review evidence which supports the theory that assimila
tion to Old Syriac texts was a predominant factor in the
formation of the Greek and Latin (so called) ' Western ' texts
of the Gospels.
The plan which I shall follow is this. I shall take Codex
Bezae as the spokesman of the ' Western,' or, to use a more
accurate phrase, the ' Syro-Latin ' authorities. I shall, that is,
take the text of Codex Bezae as the chief subject of the in
vestigation, in connexion with it adducing and examining the
readings of kindred- authorities, especially those of the Old
Latin MSS.
The discussion will, for convenience sake, be conducted
under four heads :
(i) Select Passages, These I have taken from the first
three Gospels as they stand in Codex Bezae, i,e., St Matthew,
St John, St Luke. I have confined myself (as far as the
selected passages are concerned) to these Gospels, because,
while of St Mark's Gospel the Curetonian has only a few
verses, at least in large sections of the other Gospels we
have now the two Old Syriac texts, the Sinaitic and the
Curetonian. (2) Harmonistic influence. The prin^ple of assimilation
played an important part in^the genesis of the ' Syro-Latin '
text of the New Testament generally. Naturally this is
C. I

2 THE SYRO-LATIN TEXT OF THE GOSPELS.
especially true of the text of the Gospels. For in the case
of the Gospels the tendency to assimilate kindred passages
took definite form in at least one well known Harmony.
It does not however fall within the scope of my work to deal
directly with the complicated questions which gather round
Tatian's Diatessaron.
(3) Proper Names and forms of words.
(4) Grammatical points,
A reading from the text of Codex Bezae, the first time it
is quoted, is printed in small uncials. A fine line under
neath any of its words denotes divergence from the normal
text : a thick line indicates an interpolation : the sign ^
marks an omission. I have used the term ' the true text ' to
denote the common form of the Greek text, as distinguished
from the eccentric ' Syro-Latin ' text. As the true text in
this sense I have printed that given in Dr Westcott's and
Dr Hort's edition of the New Testament. The term ' the
Bezan scribe' I have used to denote the scribe who in any
particular passage altered ' the true text ', and produced ' the
Bezan text ' at this place.
To this brief general statement I add two remarks.
The evidence in support of my main thesis, afforded by
the consideration of the phenomena of the ' Syro-Latin ' text,
varies^ infinitely in point of cogency, sometimes amounting, as
it appears to me, to that kind of demonstration which alone
is possible in critical and literary investigations, sometimes
hardly, if at all, rising above simple illustration.
Again, the strength of such evifdence lies in its cumulative
character. It is always possible, in criticising such a theory
as mine, to allege some cause, other than the one suggested, as
having produced this or that particular reading. If however
a single theory supplies a natural explanation of a series of
readings differing from each other in kind, though in the
several cases other explanations of various sorts are not im
possible, the legitimate conclusion is that that theory must
be taken (at least provisionally) as true.

SELECT PASSAGES FROM ST MATTHEW,
ST JOHN, AND ST LUKE.
Matt. i. 1 6. iacob autem genuit ioseph
cui desponsata' uirgo maria

peperit xpm ihm.
The Bezan Greek is wanting at this point, but there is no
reason to doubt that it corresponded to the Latin.
The true text is 'la/cw/S Se iyevvrjcrev rov 'Imcrrjif) rbv avSpa
M.api,a<;, i^ ^9 iyevvrjOrj ^Irjaovif 6 \ey6/Mevo^ XptffTo?.
The phrase cia desponsata uirgo Maria is obviously sug
gested by V. \% (fivt)aTev6ei<Trj<; Tfj<; iM7jTpo<; avrov Mapia<; t«3
'Ia)a-7j<f)), compare Lc. i. 27 (tt/oo? irapdevov eiivqarevfjLevrjv avBpl
K.T.X.). There are strong reasons for holding that this case of
context assimilation arose in an Old Syriac version and
passed thence into other texts, (i) The Sinaitic and the
Curetonian texts, as it is well known, seriously differ in this
passage. The problems suggested by this difference lie out
side the present enquiry. But the very seriousness of their
divergence emphasises their agreement in the words under
discussion, I give the two texts side by side :
Sin. Cur.
Jacob begat Joseph : Jacob begat Joseph,
Joseph, to whom betrothed was him to whom betrothed was
Mary the- Virgin, Mary the-Virgin,
begat Jesus, she who-bare Jesus
who-called (was) the-Messiah. the-Messiah.

4 THE SYRO-LATIN TEXT OF THE GOSPELS.
The agreement (amid such difference) in the words under
lined seems to stamp these as relics of a primitive Syriac text.
(2) The fact that in v. 20 (TrapaXa^etv M. rrjv ryvvaiKa aov)
the Curetonian reading 'to-take Mary thy-betrothed' — a
reading, so far as I know, found in no other authority — is
assimilated to v. 18, confirms the impression that this type
of phrase was characteristic of the early Syriac texts of this
Gospel. (3) The use of the active verh— peperit Christum
Jesum, — in place of the passive in the true text, is a very
natural corollary of the reading in the earlier part of the
verse. The change of a passive clause into a corresponding
active clause is very common in the Old Syriac texts of the
New Testament (see below, p, 17).
This reading is found in the Ferrar-group' — a> fivrja-Tevdetaa
•7rap9ivo<! M^apta/ji, eyevvrjcrev ^Iijerovv rbv Xeyofievov H-piarov:
in the Old Latin a g' k q (cui desponsata uirgo (om. q) maria
genuit ihm), b c (cui desponsata erat uirgo maria : uirgo autem
maria genuit ihm), and in the Armenian version.
Matt. X, II — 13.
II. H noAic yv 6IC HN AN eiceAeHTe eic aythn
el6T<\CAT6 
12. eicepxoMeNOi Ae eic thn oiKeiAN
ACnACACeAl AYTHN AerONT6C
eipHNH TU> OIKO) TOyTCO
13. A €AN MeN H H OIKeiA AIIA" eCT6 H 6ipHNH
YMCON en AYTHN.
The true text is : 1 1 eh fjv K av iroKiv rj Kcofir/v elaeXd'qTe,
i^erdaare...!^ koI iav...iX0dTa> rj elpijvr/ vfi&v eir aiiTrjv.
The Curetonian is wanting at this point. The Sinaitic is
1 The cursives 13, 69, 124, 346 form the so-called Ferrar-group. It seems cer
tain that these MSS. are derived from a common lost original, an original which
Ferrar and Abbott (A Collation of Four Important MSS. , Dublin, 1877) approxi
mately restore. Mr Rendel Harris (Ore the Origin of the Ferrar-group, 1893) con
cludes his discussion of the superscriptions and of certain readings found in these
MSS. thus : ' I think we may take it to be demonstrated that there is a decided
streak of Syriac in the Ferrar-text' (p. 19).

SELECT PASSAGES FROM ST MATTHEW, S
as follows : ' Into-whatsoever city entering (are) ye (r>«li*«^A
^.^ouK* ^Av..! rtlsiiA), be asking who (is) worthy and-there
be (ooco)i till,.. And-when entering (are) ye it, (even) the-
\\o\x%Q, give peace to that house, and-if worthy (is) that house,
your-peace shall-be (K'ocm) upon-it.'
The points of the passage are these: (i) In the last line
but one D has e<TTai, for eKQdrm. So far as I know, the
Sinaitic is the only other authority which has this reading.
It is one which would naturally arise in translation. (2) In
the first line note (a) D, like Sin., omits 77 Kmfiijv; so 1-118-
209 604 a b ff^ h k ; (b) the order of words in D has the appear
ance of being due to rough retranslation, the relative, as in
the Syriac, being brought near the verb ; (c) the words ek ^v
...ets avTJ]v are the reproduction of a Syriac idiom. The
Peshitta has caA....'i...r<'.'wr<ll (into-whatsoever,., into-it),
and so has the Curetonian in the parallel passage Lc. x. 5,
8, ID (so Sin. in v. 5, wanting in v. 10). Thus the Bezan
Greek in this line reveals clear signs of retranslation from a
Syriac text differing only from the Sinaitic by the addi
tion of the word enX (into-it). The only MS., it appears,
which coincides with D in this line is 28. (3) The addition
in z/. 12 \6yovTe<!...TovTa from the parallel passage (Lc. x. 6)
is found in a large number of authorities including N*L(/)
1-209 Old and Vulg. Lat. MSS., the Armenian. It might well
arise independently in different texts. It may be noticed
however that it would be suggested by the Syriac rendering
(give peace) of dcrTrda-aa-de, and that Ephrem's quotation shews
that it had a place in the Diatessaron (Hill, p. 344).
1 This 'be' (so Pesh,) represents the /idvare of the Greek. Either it is an
instance of the use of 'to be ' in the Syriac to represent a more definite verb in
the Greek; compare Matt. xxi. 17 ' that-Ke-might-ie (j^Oenl.l)' {=rii\ifferi)
in Cur., and below, 'your-peace shall-be' ( = A9iiTai); or it is a corruption
of a primitive reading aoa (abide) ; compare Lc. xix. 5, where Sin. Cur. Pesh,
have r<'O0Br^ [I-should-be) to represent /neivoi ; compare my Old Syriac
Element, p. 9.

6 THE SYRO-LATIN TEXT OF THE GOSPELS,
Matt, X. 42. KAi oc AN noreicH ¦ eNA tcon sAaxictcon toytcon
noTHpiON Y^ATOc YYXPOV A eic onoma maGhtoy
AMHN AerCO YMIN' OY MH AnoAHTAI 0 MICGOC AYTOY-
The true text has koX 09 av TroTla-y eva r&v fiiKp&v
TOVTCov iroTrjpiov ^Irv^^pov /xovov elf bvofia...ov firj airoXea'Tj
Tov fiiaObv aiiTOv.
The points of the passage are these : (i) The Sinaitic
and Curetonian add ' of water ' ; so most of the Old Latin
MSS, and the Latin Vulgate. (2) The Sinaitic and the
Curetonian omit the word ' only ' : it is retained in the Old
Latin MSS., and in the Memphitic. The addition of
'of water' and the omis.sion of 'only' bring the clause (so
far) into conformity with the parallel in St Mark (ix. 41).
(3) ot; yJr] dtroX'qTab 6 fjnaObt; avTov. The Sinaitic Syriac has
' Verily I -say to-you th.a.t-not shall-there-perish his-reward.'
In Mark, where Sin. has 'has lost (saor^),' the Peshitta
(alone of all authorities) has the same phrase which Sin. has
in Matt. On somewhat similar renderings in the Syriac see
p. 17. This reading is also found in the Memphitic, in most
Old Latin MSS., and in Cyprian. The difference between the
two phrases in Syriac is very slight, for it consists in the
simple interchange of xsr^A (shall-perish) and ."tsoj (shall-
lose). The two phrases in the Latin are ' non perdet mer-
cedem suam ' (e.g. f vg), ' non peribit merces eius ' (e.g. g^ k q),
(4) r&v eXa^ia-rcov rovrcov. The Bezan Latin, with the Old
Latin and Vulgate MSS. generally, has minimis. It is of
course possible that the Bezan Greek is here assimilated to
the Bezan Latin, But it is at least worthy of note that the
Syriac phrase here is that used (Sin, Cur. Pesh.) to render
TovT(ov T&v eXa'xlaTmv in Matt. v. 19. It is instructive to
compare Matt. xiii. 48, where the Sinaitic and the Curetonian
read 'They-chose the-fishes vihich-good (were) (as) good.'
Here it would appear that the repetition 'good good' was
misunderstood and taken as a superlative : hence D ta kaA-
AicTA, d meliora. Old Latin MSS. generally optiinos, optima.

SELECT PASSAGES FROM ST MATTHEW. 7
Matt, XV. 26. OYK eSecTiN AaBbin ton AproN toon t6kncon.
The true text has ovk ea-nv koXov k.t.X. The Sinaitic and
the Curetonian have r<lXo f^A ('not necessary'). The word
r^o, though commonly the equivalent of Set, is used in the
Peshitta and the Curetonian of Matt, xx. 4 to render BUaiov ;
in the Pesh. to render d^iov (2 Thess. i. 3), xaO^Kovra (Rom.
i. 28). In Rom, ii, 18 rx'AvAo (fern, plur.) appears as the
equivalent of rd Bia<j)epovTa. Hence this Syriac word «<ilo
would be a very natural rendering of KaXov, and of this Syriac
word the Bezan e^ea-riv a natural retranslation.
It is of course quite possible that the reading may have
originated in a very early Greek copy of the Gospel, in which
the word kuXov was accidentally omitted and hence the reading
OVK ea-Tiv Xa^elv produced. Compare Tert. adv. Marc. iv. 7
(non est auferre), Eus. in Psal. xxi. (Migne P. G. xxiii, 209).
This 'ia-Tiv must then have suggested the emendation e^ea-rtv.
The Bezan reading seems to be implied in Clem. Horn. ii.
19 0 Se...ei7rev' ovk e^ecrnv laxrdai Ta edvr], ioiKora kvctIv. It is
found in Origen, the Old Latin MSS. abcff^-^ gM, and Latin
Fathers, Matt, xvi, 16. CY ei 0 xpc 0 yioc toy 6y to ccozontoc.
In place of aw^ovro<; the true text has ^a>vTo<;.
We have here a reading which at once betrays its Syriac
origin. The Syriac versions (Cur. Pesh. ; Sin. wanting) have
here r<UM K'eolr^.i (of-God living). In Syriac the verb
to live is the regular equivalent of aco^eadai, and the Aphel of
the same verb (to make to live) the regular equivalent of am^eiv.
Hence the word living in Syriac would at once suggest the
ideas of being saved, saving. The change implied in the
Bezan reading from rtiUM (living) to rtLuCsa (making-to-live,
i.e. saving) is small. Compare the note below on Lc, iii. 10.
Matt, xvii, 27, eypHcerc eKei ctathpa.
The added word e«6t is found in the Old Syriac (Sin. Cur.)
— ' And-thou-shalt-find there a-stater.' The addition of the

8 THE SYRO-LATIN TEXT OF THE GOSPELS.
word there is quite in harmony with the additions which the
Old Syriac frequently makes to define time and place (see
Baethgen, Evangelienfragmente, p. 22). Compare e.g. Matt. ii.
23 -and-he-came thither' (Sin. Cur.); iv. 20' and they imme
diately left the nets there' (Tatian, Hill, p. 62); xix. 3 'and-
there-came-near to-Him there the-Pharisees ' (Sin. not Cur.);
Lc. xiv. 8 'lest there-shall-be invited there' (Sin, Cur. Pesh.);
xxiv, 23 'angels we-saw there' (Sin. Cur. Pesh.). In these
passages, so far as I know, the Syriac authorities stand alone.
The addition appears in different forms in Latin MSS. —
(«) a b c g' n L Q R ibi ; (b) d illic ; (c) f in eo ; (d) e in ilium.
Matt, xviii. 2. kai npocKAAecAMeNOC 0 ihc haiAion gn.
The Old Syriac (Sin. Cur.) has here : ' There-called Jesus
oite boy.' The addition of the word one is characteristic of
the Syriac texts. Thus in St Matthew we find ii. 23 ' in-a-
city one' (Cur., not Sin.); viii. 2 ' and -behold one man a-leper'
(Cur. Pesh. (' one le^per'); Sin. wanting); xv. 22 'and-behold
one woman' (Cur., not Sin.); xxi. 2 'one ass' (Cur.; Sin. want
ing) — passages where, so far as I know, the only authority for
the insertion is the Syriac. In our present passage the only
authority for one besides the Old Syriac and D is that con
stant ally of the latter, the Old Latin e.
Matt, xviii. 20.
OYK eiciN r-i^p Ayo h jpeic cYNH.rMeNoi
eic TO 6MON ONOMA
nAp oic OYK eiMei eN Mecco aytojn.
The true text is ov ydp ela'iv...ovofia, eKel elfil iv fieo'a
avTwv. This reading would obviously most easily arise in a version,
where the initial ov was taken as a negative, a negative being
inserted in the subsequent clause to make sense. This is
exactly what has taken place in the Sinaitic Syriac, which
reads here ' For there-are-;?^/ (T-a.^ iul) two or three who
(-are)-assembled in-my-name, in whose midst I (am) 7tot
(^_omh\XLS r^T^ Kil.t).' Of this Syriac reading the Bezan

SELECT PASSAGES FROM ST MATTHEW. 9
reading is a somewhat literal though awkward translation —
Trap' ol<;...iv fiiam avrSv (d aput quos non ero in medio eorum
— a close rendering of the Greek).
The Old Latin g' has a conflate reading. After giving the
true text it adds, after in medio eorum, the words non enini
sunt congregati in nomine meo inter quos ego non sum. The
character of this Latin reading shews that it is not the source
of the corruption.
Matt. XX. 28.
I YM6IC A6 ZHT6IT6- CK MeiKpOY AYSHCAI
KAI CK M6I20N0C GAATTON 6INAI
3 eicepxoMGNoi Ae kai rrApAKAHSeNTec
AeinNHCAr mh ANAKAeiNecGAi
5 eiC TOYC 626X0NTAC TOnOYC
mh noxe eNAoSoTepoc coy eneAGn
7 KAI npoceAOcoN 0 AeinNOKAHToop einn coi
en katoo x<Jipei" kai kataicxynQhch
9 eAN Ae ANATrecHC- eic ton httona tohon
kai eneAGH coy httcon
II epei coi 0 AeinNOKAHTCop- CYNAre en anco
KAI eCTAI COI TOYTO XPHCIMON.
The above passage is an interpolation in the text. The
only other Greek authority which contains this paragraph
is ^ (Codex Purpureus). This text (except in small matters of
spelling, e.g. ^rjTire) differs from that of D in the following
points alone : line 2 iXdrrmv ; 11. 4, 5 /j-tj eh tov? e^ej(,ovra<;
TOTTous dvaKXivecrde; 1. lO om. Kai; 1. Ii dye; 1. 12 -x^prjcrifiw-
repov. The Sinaitic is wanting at this point. The Cure
tonian text contains the following interpolation at the same
point in St Matthew's Gospel^ :
^ Cureton (Gospels, Preface, p. xxxvi) writes thus : 'This same passage is also
read in the margin of the Philoxenian version in the Vatican, and is cited by
Adler in full : and I have found it in the margin of a copy of the Peshito of the
Nitrian manuscripts. No. 14,456 in the British Museum. As it stands in these
copies, it is plain that it has been translated immediately from the Greek and not
been taken from another copy of this Syriac text, from which, indeed, it varies in
language considerably, as it will be seen by comparing them.'

10 THE SYRO-LATIN TEXT OF THE GOSPELS.
ye-may-be-great littleness that-from seek-ye but ye
ye-may-be-little greatness from and-not
.r^iusazMrc' A\i-i \ .^_o^r^ >ii*a.it5q.i r<lsa 3
a-supper to-the-house-of are-ye bidden when
.r^^i^uss r^iv&o.V3 ^ivrqAiflfljq «^_o^c\cn rdA 4
honoured in-the-place reclining be-ye not
more-than-thou who-honoured (is) he there-may-come that-not
below draw-near the-supper the-lord-of to-thee and-there-say
.r^L^jJSiJ^o .1 lis ¦n ^ca.siio 7
those-reclining in-the-eyes-of and-thou-be-ashamed
.pc'A^i^a pc'^o.Ta %A^h\Mh\ ^.i ..^ ^
mean in-the-place thou-shalt-recline but if
than-thou who-mean (meaner) he and-there-come
draw-near the-supper the-lord-of to-thee and-there-shall-say
and-recline and-go-up
in-the-eyes-of honourable glory to-thee and-there-shall-be
.r^ %i*iq no
those-reclining
When we examine the Greek of D and ^ we find indi
cations that we have not before us the original form of the
gloss, (i) The awkwardness of the Greek points to retrans-

SELECT PASSAGES FROM ST MATTHEW. II
lation r see especially the first two lines, (2) The gloss
evidently takes its rise from words in the context (v. 26),
which are as follows : 8? av diXrj iv v/niv fieya^ yeveadai
ea-rai (v. I. ecrTco) vfjiwv SidKovof. But the phraseology of the
gloss itself bears no resemblance to that of the context. (3)
The main body of the gloss is obviously based on Lc. xiv.
8 — 10 OTav KXr]d-§f vTTo Tivo<i eh ydfiovf, firj KaTaKXi,6fj<; eh
TTjv irpeoTOKXicriav, firj -rroTe evTifiOTepof crov y KeKXrjfievoi; vir
avTov, Kai, eXdcov 0 ere koX avTov /caXetras ipel aoi A09 tovto)
TOTTOV, Kai TOTe dp^rj fieTa alcr^xyvrj^ Tbv 'ea')(aT0v tottov Kar-
e')(^ei,v. aXX brav KXri6jj<! TTOpevdeh dvdirecre eh Tbv 'ea')(aT0v
tottov, Iva OTav 'eXOr) 6 KeKXrjKWf (re ipel aot ^iXe, Trpoa-avd^rjdi
avcoTepov ' Tore ea-Tai crot So^a ivcoinov ttuvtiov tmv crvvava-
Ket/jLeveov (yoi. The verbal links between this part of the
gloss and its original, it will be seen, are but few.
From the Greek we turn to the Syriac. (i) The
contrast between the halting and awkward character of the
former and the simple and forcible nature of the other is
striking. Notice, for example, the first two lines — how the
presence of the negative in the second clears up what in the
Greek is obscure. We see at once how parallel these two
sentences are to the two parts of the Lord's saying (Lc. xiv.
1 1, xviii. 14) 'Everyone that exalteth himself shall be humbled,
and he that humbleth himself shall be exalted.' (2) Two
key-words of the first two lines are taken from the context
(v. 26) : ' Whoso (is) wishing among-you that-he-should-be
great (r£.s\)' ; hence .,_a-air^A», rCA^O-ssi'. (3) The
words ' littleness ', ' ye-may-be-little ' seem to be derived from
the parallel passage Lc. xxii. 26 (o fiei^cov iv vfiiv yiveaOm
w? 6 vedoTepo'i), where the Old Syriac (Sin. Cur.) has : 'but who
soever ^/ra^ among-you shall-be as the-little-one (r<'TO:^t).'
(4) The language of the main body of the gloss is modelled
' It will be noticed that, while nothing in the context in the Greek suggests
auf^ffot, the Syriac verb 'to be great' is the regular equivalent oiav^avuv (Matt. vi.
28, xiii. 32, Mc. iv. 8, Lc. i. 80, ii. 40, xii. 27, xiii. 19, Jn. iii. 30), and therefore,
if the Syriac form of the gloss is the original, the Syriac verb would naturally sug
gest this Greek verb here; see p. 14 n.

12 THE SYRO-LATIN TEXT OF THE GOSPELS.
on that of Lc. xiv. 8 — lo, which runs thus in the Old Syriac
(Sin. Cur,) :
[Ati<'A\] rid re'^o^ucsiA Avjp^ 'p3W5a,i .*h\Jsat^
[Cur. shalt-thou-go] not to-a-feast (art) thou bidden when
lest honourable in-the-place [om. Cur.]for-thee shalt-thou-recline
.VV-1-S9 i o 1 .tq.t ^ ^h\ ^.SQUSQ r<'ocn_i
more-than-thou who-honourable (is) he there [om. Cur.'] bidden there-be
.acD vvl KsiriJ^cs ^\ orAo vA.l ^ om i^^rtilso
give to-thee and-say bade and-him that-thee he and-there-come
^i< ^c(x.a njii ^:ucno rOcoA [rCii&o.i] T^\h\T<
thou ashamed while and-then to-this-man [Cur. place] place
Aojso.Hf^.l rdsa r^fX' rt'AuTMr^ r^^o.is vCaduio^
thou-art-bidden when but last in-the-place shalt-recline
that-when last in-the-place [om. Cur.lfor-thee recline go
.1, •» ui vyl "t-tardJ vi^JJMt.i [f-sa] oen K'i^r^.i
my-friend to-thee he-may-say who-bade-thee [Cur.] he there-has-come
in-the-eyes-of glory to-thee and-there-shall-be above ascend
.[«,__acaA_^] r^.\t.SiSo
[Cur. all-of-them] those-reclining
The phrase ' lo7'd of tJie supper ' is drawn from the imme
diate context of the passage just quoted (Lc. xiv. 12) 'And-
He-said also to-the-lord-of the-supper' (Sin. Cur.), the word
supper heing used in vv. 12, 13, 16, 24^. (5) There are
1 Cur. has the Ethpaal part. JSa.l VSi . In the next line Cur. has rsA.I (lest).
^ The compound Greek word in D — 6 SairvoKK-riTUip — seems intended to repre
sent the Syriac compound expression ' the lord of the supper.'

SELECT PASSAGES FROM ST MATTHEW. 1 3
certain correspondences in the Syriac which seem to point to
this as the original form. The ' draw near ' of 1. 6 answers to
the 'draw near' of 1. lO; 'in the eyes of those reclining' of
1. 7 to the same words in 1. 1 1 ; the ' mean ' of 1. 8, applied to
the seat at table, prepares the way for the ' mean ' of 1. g
applied to the guest, just as the ' honoured ' of 1. 4 corresponds
to the ' honoured ' of 1. 5. This last word indeed may be said
to be a key-word, for it has a place in the final promise of
commendation. 'There shall be to thee honourable (or
abundant) glory.' Further, the word r<'A»'iAu20 in the last
line is a link between the Bezan and the Curetonian texts.
For this Syriac root connotes utility, profit ; thus, words from
this root are used in rendering %/3?j(rt/ioi' in 2 Tim. ii. 14,
m^eXifia in Tit. iii. 8, axfteXia in Rom. iii. i. To sum up : a
review of the Syriac form of the gloss shews (i) that it runs
smoothly and naturally ; (ii) that it itself reveals its origin :
it springs out of the language in the context, and it is com
posed of phrases derived from certain passages of the Gospels
which would be naturally brought together (Lc. xiv. 8 ff., xxii.
26). The gloss is found also in many Old Latin MSS., viz.
a b c e ff ^-^ h n ; the first part (v/ieh — elvai) is also preserved
in m g', the second part (elaep^b/J-evoi — the end) in g^. It is
also given (see Bp Wordsworth in loco) in some MSS. of the
Vulgate. The form of the gloss in the Old Latin Cod.
Vercellensis (a), the chief variants being noted, is as follows
(see Tischendorf in loco) —
' uos autem (e enim) quaeritis de (m in) pusillo (m modicis,
emm medico) crescere (m extolli),
et de maiore (and magnis, m maximis, emm maximo, bg'
theo minore) minores (c minor, m emm minui, e minorari, b g'
and theo maiores) esse (e am., ff' g' fieri).
Intrantes (m g^ emm cum autem introieritis) autem et (e ff'
om.) rogati (and om. et rogati, m g° emm ad cenam uocati) ad
cenam (theo cenare)
nolite recumbere (ff ' h discumbere) in locis eminentioribus
(g^ emm theo superioribus loc, m honorificis locis)

14 THE SYRO-LATIN TEXT OF THE GOSPELS.
ne forte clarior (m g" emm theo dignior, e honoratior) te
superueniat et accedens (g'' emm add. is) qui ad cenam uocauit te (fP
qui imtitauit te, m inuitator) dicat tibi : adhuc deorsum (g^
emm inferius, m infra) accede,
et confundaris (ff ' theo et erit tibi confusio).
Si autem in loco inferiori recubueris (ff ' h disc!),
et superuenerit (g^ emm aduenerit) humilior te,
(e add. tunc) dicet tibi qui te ad cenam uocauit (and
inuitauit) : accede adhuc (e om) superius (b ff ' h and sursum,
m in superiori loco),
et erit hoc tibi utilius (e et tunc erit tibi gloriam coram
discumbentibus)! It is sufficient to give two reasons for the belief that we
cannot seek the original form of the gloss in the Latin.
(i) Putting aside those variations which imply difference
of reading, the number of synonymous variants seems to
imply different attempts to render a common original. (2)
In the first line an imperative is required. ' Seek ye from a
lower position to rise to a higher.' This is demanded by the
illustration of the feast which follows. The imperative then —
' seek ye ' — must be the original form. The Greek ^rjTecTe is
ambiguous. The Latin authorities agree in having the
indicative. The quaeritis then of the Latins has every
appearance of being a mistaken rendering of the Greek
^TjTeiTe ',
' In Lc. xxii. 27 D reads efw fAp 6N /wecw Y^WN hAGon oyx coc o
ANAKeiMGNOC AAA WC O AlAKONCON KAIJYM6IC HYSHGHTe SN TH AlA-
konia moy coc o Aiakoncon. The points are: (i) The passage is assimilated
to the passage in Matt. xx. 28 ; for the Bezan ^\Bov comes from p^Bev (Matt.).
(2) With the o6x '^'s ivaKei/ievos dXX' is StaK. compare the Curetonian of Lc. xxii.
26 'And-he-that(-is)-chief is as the-server and-not as he-that-redineth' {oiyX 0
avaK. ; being read as if it had been ovx 6 avax., and transplanted into an earlier
clause). This incorporation in the gloss of a reading peculiar to Cur. suggests that
the gloss was originally Syriac. (3) This suggestion is confirmed by rji^erire,
which would naturally represent the Syriac word ' ye-became-great,' derived from
' Whosoever (is) great among-you ' [v. 26), ' who-is great} ' {v. 27) ; see p. 11 n.

. SELECT PASSAGES FROM ST MATTHEW. 15
Matt. xxi. 28 ff.
YHAfE CHMepON epfAZOY eic TO AMneAOJNA
0 Ae AnoKpeiGeic eineN oy GeAoi
YcrepoN Ae MeTAMeTAiweAHGeic
ahhAGcn eic TON AMneAcONA
npoceAGcoN Ae toj erepto eineN cocaytcoc
o Ae AnoKpeiGeic eineN epoj kc YiiArcio
kai OYK AHHAGeN" TIC eK TCON AY-CO
TO GeAHMA toy HATpoc cnoiHceN
AerOYCIN 0 AICXATOC.
The true text has viraye a. ipy. iv Tm dfiTreX&vi, ' 6 Be
airoKpiOeh eViTev 'Eyco, Kvpie' Kai ovk dTTrjXdev. TrpoaeXdav
Be Tft5 BevTepay eiTrev o)aavTa><;' 6 Be aTTOKp. elTTev Ov deXto'
vaTepov fieTUfieXTjOeh dTrrjXOev. rt? e'/c TOiv Bvo eTToirjaev Tb
deXTjfia TOV TraTp6<; ; Xeyovcriv 'O f/ffrepo?.
The Sinaitic Syriac has : 'He-said to-the-first. Go, my-son
\Cur. add. to-day] work the-work in-the-vineyard. He-said
to-him I-will not; and-in-the-sequel there-repented-him his-
soul, and-he-went to-the-vineyard. And-he-said to-the-other
(r^Vjr^) likewise ; and-he-answered and-said, Yea, my-
Lord ; and he- went not. Which of these {Ctir. add. two
seemeth to-you that-he-] did the-will of-his-father .' Saying
(were they) to-Him, That last [Cur. first],'
The Bezan, it will be seen, agrees with the Old Syriac
(Sin, Cur.) in (a) the transposition of the two sons : so also
the Peshitta and Old Latin MSS. ; (b) the interpolation
' into-the-vineyard ' in line 4; so many Latin MSS. Further,
the Sinaitic agrees with D 604 and Latin MSS. in transposing
the order of the sons and at the same time in reading 'the
last' in the answer of the crowd.
Matt. xxii. 34. cynhxGhcan en ayton.
The true text has eTrl to avTo. The Sinaitic and the
Curetonian have ' there- were-assembled unto-Him (or against-
Him, coAicA).' The Arabic Tatian (Hill, p. 180) has: 'The

l6 THE SYRO-LATIN TEXT OF THE GOSPELS.
Pharisees... assembled themselves together against Him, to
strive with Him! As this reading differs from that of the
Peshitta ' there-assembled together ' (= eTrt to avTo), it is prob
able that the Arabic here preserves the true Tatianic reading.
The substitution of 'unto-Him' for 'together' (e'tti to uvto)
is quite in harmony with the pronoun-loving Syriac.
This apparently Syriac reading is preserved in b c e ff^ h ;
aeth. ; Hil. Like the Arabic Tatian, f has the conflate
reading in unum ad eum.
Matt, xxiii. 9.
KAI HATepA MH KAAeCHTC Y^eiN 6111 THC fHC
eic TAp eCTIN 0 HATHp Y^OON 0 CN OYpANOIC.
The true text has /cat Trarepa firj KaXe<T7)Te vfiwv cTri t^9
795?, eh ydp i(TTiv vjmSjv 6 TraTrjp 6 ovpdviov.
The Syriac (Sin. Cur. Pesh.) has : ' And-father ye shall not
call for-you (»__a^) on-earth : for one is (ocd) your-father,
who-(is-)in-heaven! Compare Aphraat's paraphrase (p. con):
' Father shall not we call for-us (^A) on-earth.' This use of
the preposition A with the reflexive pronoun is very common
in Syriac (Noldeke Gram. | 224) ; see e.g. above, p. 12, 11. 2, 7.
This 'for you' is found in 26^", Old and Vulgate Latin MSS.,
the Egyptian Versions, and in Clem. Alex. (Strom, iii. 12,
p. 551 ed. Potter).
Matt, xxv. 41. eic TO TIYP TO AICONION
O HTOIMACeN 0 HATHp MOY
TCO AIABoAcO KAI TOIC AfreAOIC AYTOY'
For the second line the true text has to r)ToifiaaiJi,evov.
Neither the Sinaitic nor the Curetonian is extant at this
point. Aphraat (p. rdaa.) has 'to that fire which-(is-)pre-
pared (r<'.T»Av^) for-the-evil-one and-for-his-angels.' The
mention of the Father however in the Bezan text shews
that in that text the passage has been assimilated to Matt.

SELECT PASSAGES FROM ST MATTHEW. 17
XX. 23^ (oh -fjToifiaa-Tai vtto tov Trarpo? /iou). This latter
passage is literally translated in the Curetonian and in the
Peshitta, but in the Sinaitic it runs thus : ' for whom my-
Father prepareth (ar>r<' .^_acrA ¦ai^'w.t).' Further, the
Arabic Tatian (Hill, p. 165) has : 'for whom my Father hath
prepared it,' This resolution of a passive verb (with the agent)
into an active verb (with the subject) is specially characteristic
of the Old Syriac version, I have noticed the following
instances in the Sinaitic Syriac of St Matthew : iii. 6
(i^aTrTi^ovTo...vTT' avrov) 'he was baptising them'; ix. 17
(prjyvvvTai 01 dcTKoi) ' lest the wine split those skins ' ; ix. 32
(Baifiovi^ofievov) 'whom a devil rode'; x. 22 (eaea-de fiiaov-
IMivob vTrb TTdvTcov) ' men shall be hating you ' ; xiv. 1 1
(rjvi'xPr) rj K€(f)aXrj avTov) ' they brought the head of John ' ;
xviii. 30 (to o^eiXofievov) 'what he owed'; xix. 12 (evvovxoi
oiTive^ evvovxicrdrjcrav vtto t&v dvdpcoTrcov) ' eunuchs whom
men have made (o.Ta^)'; xxiii. 7 (KaXeladai vtto Tmv dv6pd>-
TTwv) 'that men should be calling them'; xxvii. 12 (e'l' tw
KaTrjyopeta-Oai, avTov vtto tcov dp'^^uepeatv) 'when the chief
priests and Pharisees accused Him'; xxvii. 64. (KeXeva-ov odv
da-^aXi<T6fjvai tov to^ov) ' command that they watch the
sepulchre.' Compare also ii. 16, iii. 13, v, 13, xviii. 25,
xxiv. 9; see Baethgen, Evangelienfragmente, p. 29, for
similar instances in the Curetonian.
Thus a Syriac reading in Matt. xx. 23 has been intro
duced into the Bezan text of Matt. xxv. 41 ; or — may we
say.' — into the Syriac text which underlies parts at least
of the Bezan text.
The Bezan reading in Matt, xxv, 41 has a special interest
through its wide attestation in quite early Patristic authorities^
viz. Justin Dial. 301 D ; Clem. Hom. xix. 2 ; Iren, ii, 6 § i,
^ It is worth noting that in the parallel passage, Mc. a. 40 (dXX' ok TiTol/iaffTat),
Sin. taking dXX' oh as flXXois has 'for-others however it-is-prepared.' The
converse confusion is found in the Curetonian of Jn. iv. 38 (dXXoi /ce/co7rid(ca(rii')
'but those who laboured.' In Mc. x. 4oDabff^ k aeth have the same mis
reading as Sin. Comp. Jn. vi. 23 (below, p. 20).
" For the Patristic authorities see Dr Hort's note (Introduction, Notes on Select
Headings) and Resch, Aussercanonische Paralleltexte, p. 313 ff.
C. 2

1 8 THE SYRO-LATIN TEXT OF THE GOSPELS.
iii. § 33, 2, iv. 55 § i, 65, 66; Clem., Coh. ad Gentes, ix.; Tert.
Hermog. xi. (the true reading being found in De came xiv.);
Cyprian Test. ii. 30, iii. i, De opere 23. It is found also in
I 22 a b c fP'= gl h r r^ R.
Matt. xxvi. 15. OIC Ae ecTHCAN aytco .A. ctathpac.
The true text has dpyvpia.
There is no special reason why such a reading should
arise in the Greek. For Greek has the convenient neuter
plural dpyvpia. The Sinaitic and the Peshitta have here
'thirty of-silver (r<lajaaA3 ^a1^).' It would be very easy
for a Syriac reading or gloss to arise, inserting the coin
after the numeral. The cursives 1-209^ have o-raTTjpa? dpyv-
piov, the latter word exactly answering to the Syriac ' of-
silver'; compare h stateres argenteos. It should be further
noticed that the Arabic Tatian (Hill, p. 218) inserts a
mention of the coin — 'thirty dirhems (i.e. drachmas) of
money".' The Bezan reading is found in a b q and in Euseb. Dem.
' Mr F. C. Burkitt in his notice of the Sinaitic Syriac MS. (Guardian, Oct. 31,
1894) writes thus: 'Two groups of cursives with mixed texts stand out as having a
special affinity with Syr-vt. These are i-(i i8-i3i-)209 and the " Ferrar group ".'
Though these two cursives do not here coincide with Sin., their relation to the Old
Syriac text makes it not improbable that they preserve here an Old Syriac reading,
a supposition confirmed by the genitive dpyvplov.
^ I take this opportunity to notice the Bezan reading in Mc. xii. 14 Aoynai
emKAl<t)AAAlON (true text K^vaov) KAlCApi (d, dare tributum Caesari). The word
KTJvffos occurs in the N. T. only in Matt. xvii. 25, xxii. 17, 19, Mc. xii. 14. In
Matt. Sin. Cur. Pesh., in Mc. (where Cur. is wanting) Sin. Pesh. translate Krjva^os
by the words ' money-of the-head ' (r<lX»T r<l&Qa&), ' money of-the-heads '
(r<'jL,'t,t, Sin. in Mc). The word (p6pos is so translated by Sin. Cur. Pesh. in
Lc. XX. 22, xxiii. 2, and by the Peshitta in Rom. xiii. 6 f. The last named version
renders the words ^c rais ii/UpaiS t^s droypatfi^s (Acts v. 37) by the paraphrase
' in-the-days (in) which- written were the-men in-the-money of-the-head.' Thus the
regular Syriac equivalent of the Greek words denoting ' tribute ' is a phrase mean
ing 'poll-tax.' It would seem then that a bilingual scribe, familiar with this Syriac
phrase, introduced into the Bezan text the Greek word for 'poll-tax' {iTn.Ke(pa\aiov,
[Arist.] Oecon. i., xv.). The Old Latin k (which has a text closely akin to that of
e, the constant ally of D ; see Dr Sanday in Old-Latin Biblical Texts, No. 11.
pp. Ixvii ff., xciv ff.) has the corresponding Latin term — capitularium.

SELECT PASSAGES FROM ST JOHN. 1 9
Evan. (Migne, P. G. xxii, 743), Origen (lat, interp.: Migne,
P. G. xiii. 1726),
John iv. 42. OYKen Aia thn chn maptypian nicreYOMeN,
The true text has XaXidv in place of jxaprvpiav. The
Sinaitic Syriac is wanting iv. 37 — v. 6. The Curetonian in
V. 42 has : ' And-saying were- they to that woman, Now it- is
not because-of thy-word (.t^hi^sn \\pi) believing (are) we
in-Him! In v. 39 (¦.¦ttoXXoI iTTiareva-av eh avTbv twv ^a/j,.
Bid TOV Xoyov Trjt; yvvaiKb<; /jLapTvpovcrr)^...) the same version
has: 'And-from that city many believed in-Him... because-of
her-witness (cn^o.icnflo .iyw) (even) oi-that woman who-
saying was All that which-I-have-done He-told me.' In the
latter verse the Curetonian gives the ideas connoted by the
Greek, but changes the form of the phrase : the notion of
witness comes early in the sentence, being expressed by the
substantive. Thus the phraseology of the Curetonian in
V, 39 is perfectly natural : no other authority has the reading.
Clearly v. 42 is closely parallel to v. 39. In the Bezan text
we have an instance of context-assimilation, v. 42 being
apparently assimilated to the Old Syriac text of v. 39. Is
it not probable that in v. 42 D reproduces an old Syriac
reading.' The Bezan reading is found in X* b 1.
John vi. 17. KATeAABeN Ae aytoyc h ckotia.
The true text has kolI a-KOTia ¦IjBrj iyeyovet.
The Curetonian and Peshitta have' : coA i\ocn ^\&jc.uO
(and-darkness was to-it (i.e. the boat)). There are thus two
points common to the Bezan and the Syriac texts, (i) the
omission of now ; (2) the insertion of a pronoun.
Further, if in the Syriac sentence the word AxAi.tK' —
the verb used as the equivalent of KaraXa^elv in reference to
darkness in Jn. i. 5, xii. 35 — were inserted before 4>oen, then
1 The only words legible in Sin. at this point are : ' To-Capemaum because...
to-it {or it).'
2 — 2

20 THE SYRO-LATIN TEXT OF THE GOSPELS.
the Syriac (the A now denoting the object) becomes 'And-
darkness had taken (or took) it ' — the equivalent of the Bezan
Greek. Thus, while to produce the Greek Bezan reading
the whole sentence is remoulded, the corresponding reading
in Syriac would be generated by the simple insertion of a
single word, suggested by two parallel passages.
The Bezan reading is found elsewhere only in X-
This theory as to the reading under discussion is, I think,
confirmed when we remark that we have evidence that in the
Diatessaron the same verb apparently was inserted in a
similar passage. In Matt, xxvii. 45 we read (XKOTot iyevero
im-l TToaav (Mc. xv. 33, Lc. xxiii. 44 icf! oXrjv) Tr)v yrjv — a
sentence literally translated in the Syriac versions. The
Arabic Tatian has: 'Tenebrae occupauerunt un'mersam terram'
(Ciasca, p. 92); 'darkness covered the whole land' (Hill,
p. 248). Further, the Gospel according to Peter (ed. Swete,
p. 7) has o-«oTo? KaTeaye TTaaav ttjv 'lovSaiav^.
John vi. 23. aAAcon nAoiApeicoN cAGontcon.
The true text has aWa rfxOev ttXom.
It is evident that dXXcov comes from the dXXd of the true
text misread as dXXa. But this misreading, taken in con
nexion with the recasting of the sentence, implies the medium
of a version. The Curetonian", making this mistake of
reading dXXd as dXXa, has : ' And-when there-came boats
other (r^h\Xi\i»r^ from Tiberias.' Of this natural Syriac
representation of the true text (misread) the Bezan Greek is
the natural retranslation.
X has iTTeXdovTwv ovv twv TrXoiav. Here we must take
account of (i) the construction, which, like that in D, recalls
the Syriac; (2) the omission of 'other' and 'but'; (3) the
compound verb. Was the K'AuiiMrc'C other') either changed
in some Syriac text, or read by some Greek scribe as though
1 For indications that this document is in large part based on the Syriac
Diatessaron see my Old Syriac Element, pp. 116 ff.
^ The following words alone appear to be legible in Sin. ; ' Boats came from
T.,..'

SELECT PASSAGES FROM ST JOHN, 21
it had been changed, into r^^iut^ (' postremae ') or ^iMK*
(' postea ') ? If so, we have an explanation of the compound
verb iTTeXOovTwv. The Old Latin b has a conflate reading :
' et cum j«/^menissent aliae naues'; e has: 'uenerunt aliae
nauiculae.' In V. 23 the words ev')(apia-Trja-avTO'i rov Kvpiov are omitted
in the Curetonian D 69* a e arm.
John vi. 56. cN eMoi MeNei KAfco eN aytco
kaGooc eN eMoi 0 hathp kapco 6n tco nATpi
AMHN AMHN AeroJ YMeiN
eAN MH AaBHTC TO CCOMA TOY Y'OY

TOY ANGpconoY coc ton AproN thc zcohc
OYK exere zcohn 6n aytco.
The first line of the gloss is modelled on v. 57, x. 14,
xiv. 10, XV. 9. How easily these words would arise is seen in
the following passage of Aphraat (p. A^) : ' When a man
gathers his soul in the name of Christ, Christ abides in him,
and God abides in Christ. So then that one man is of
three parts, himself, and Christ who abides in him, and God
(v. I. the Father) who is in Christ, as our Lord said : I in my
Father and my Father in me.'
The last four lines are founded on v. 53 d/Mriv dfirjv Xeya
ipXv, idv fjJi] (fydyrjTe ttjv crdpKa tov vlov tov avdpwTTOv Kai
irirjTe avTov to alfia, ovk exere ^eorjv iv eavToh. The differences
between this verse and the gloss (over and above the substi
tution of iv avra for iv eavToh and the omission of the
reference to the blood) are (i) the substitution of to o-ou/cta
for TTJV ardpKa, (2) of Xd^rjTe for ^dyrjTe, (3) the insertion of
the words <m? tov dpTov r^? ^wrj^.
To take first the substitution of rb awfia for t»)v adpKa :
in V. S3 the Syriac (Sin. Cur. Pesh.) has: 'Verily, verily,
saying-(am-)I to-you that except ye-eat His-body (eoi.^;^)
(even) of-the-Son of-Man and-drink His-blood there-is-not
to-you life in-you\' The Syriac versions (Sin., which how-
1 ..^ft^rJ (Sin. Cur.) „.ftA5aQ.1n-i (Pesh.)

22 THE SYRO-LATIN TEXT OF THE GOSPELS.
ever is wanting in v. 52, Cur. Pesh,) have 'body' throughout
John vi. to translate (ydp%. So also in John i. 13, 14 the
Curetonian (not Pesh., Sin. wanting) ; Aphraat does not quote
i. 13, but he twice (pp. v^a, vflon) quotes i. 14 in this
form. No other authority, so far as I know, reads 'body'
in any of these passages, with the one exception of the Old
Latin m in John vi. 51 (et hie panis quem ego dabo pro
huius mundi uita corpus meum est). The word ' body ' then
is a frequent and characteristic rendering of crdp^ in the
Syriac versions of St John. The use of the word o-cG/ia in the
Bezan interpolation seems to be a clear proof of its Syriac
origin. We pass on to consider Xd^r/Te in place of ^dyrjre.
Compare the reading of D in i'. 53 6an mh AaBhtg (true text
(j)dyr)Te) thn cApKA and in v. 57 o AamBancon (true text
Tpcoycov) Me. Clearly the substitution of ' take ' for ' eat ' fol
lows upon the Syriac substitution of ' body ' for ' flesh,' since
the word 'body' at once recalls the 'take' of the words of
Institution (Matt. xxvi. 26 Xa/Sere, (f>dyeTe, tovto ia-Tiv to
arufid fiov, Mc. xiv. 22 Xd^ere, tovto K.T.X.y. We have more
over some direct evidence for the substitution of ' take ' for
'eat' in an Old Syriac text of John vi. 53, since Ephrem
(Moesinger, p. 245) has the words ' Si quis carnem meam
non sumpserit, uitam non habet.'
The interpolation is found in a somewhat different form in
the Old Latin a ff" : 'si acceperit homo corpus filii hominis
quemadmodum panem uitae, habebit uitam in eo (ff^ illo).'
Here the word homo should perhaps be compared with the
Syriac (Sin. Cur. Pesh.) of ^;. 50 : ' This is the bread which
came down from heaven that a man (jLJ»<'=Tt?) should eat
thereof It should be added that a and Victorinus in v. 53 have
the interpolated words sicut panem uitae.
1 Aphraat (p. i*^^^) gives the words of Institution thus : ' This is my body ;
take, eat of it all of you. '

SELECT PASSAGES FROM ST JOHN. 23
John viii. 53. mh cy MeizcoN ei toy a aBpaam- on AneGANeN.
The true text has (a) oa-Tit, not oTt, (b) Trar/ao? ruiwv
before 'A^pad/j,.
In regard to the oti, while it is of course possible that it
is to be explained as an itacism for oo-tk;, it should be
noticed that the Syriac iuJM.i means because he died or
who died^. The Old Latin a seems to be the only companion
of D in this reading.
The omission of ' our father ' would be easier in Syriac
than in Greek or Latin. For in Syriac (i) it would be the
omission of a single word ; (2) the word .^^tor^ (our-father)
would easily fall out before >JcnT=3rc'(Abraham), the words
beginning with the same two letters. As a matter of fact
'our-father' is omitted in the Sinaitic Syriac, which has:
'Art-Thou greater than Abraham and-than the-prophets
who-died (or because-they-died).'" The same omission is
found in the Old Latin a b c e ff" 1.
Johnxi. gf. oyx^i AcoAbka copAc exei h HMepA...
eAN Ae TIC nepinATH eN th nykti npocKonrei
on TO C|)COC OYK eCTIN 6N AYTH.
The true text is ovxi, BcoBeKa wpai ela-iv ttJ'; rjfiepa'i ; . . .iv
avTw. There are two points here, (i) The Syriac (Sin. Pesh.)
of the first line is : ' Not twelve hours are-there (AupC) in-the-
day.'' The Syriac .sd Av.rt' (there-is in), like A Aur^(there-is
to), is a not infrequent equivalent of e^^et. Thus Baifioviov
e^et, (Lc. vii. 33) becomes in the Syriac ' A-devil there-is in-
him.' If therefore the Bezan scribe were following the Syriac
at this point, he would naturally retranslate the Syriac by
its most obvious Greek equivalent, i.e. by the words of the
Bezan text. D here seems to stand alone. Compare p. 41 f.
1 The Bezan Latin is quoniam. Had the word there been quia, it might have
been urged that q^uia arose from qui.

24 THE SYRO-LATIN TEXT OF THE GOSPELS,
(2) The Sinaitic Syriac of the last two lines is : ' Whosoever
in-the-night however walketh stumbleth, because the-light
(or light) is not in-him or in-it (wis).' The last word cos
can grammatically refer either to the man or ta the night.
It is in itself quite ambiguous. In Latin, it should be noticed,
there is no ambiguity (nocte... in eo)'. The reading therefore
is important in view of somewhat similar Bezan readings,
which might be regarded as due to the influence of an am
biguity either in the Syriac or in the Latin. Thus in Lc. ii.
22 D has Ai HMepAi toy kaGapicmoy aytoy- Here the Syriac
suffix (cnAii.^,iA», his- or her-purification) and the Latin eius
are alike indeterminate. Compare Old Syriac Element, pp.
81, 152,
John xi. 28. KAI TAYTA einoYCA AnnAGeN kai ecfjcoNHCCN
THN AAeAct)HN AYTHC MApiAM CICOnH.
The true text has in the last line M. t^v dS. avTij'i XdOpa.
The Sinaitic Syriac has : ' And-when she-had-said these-
things she-went silently^ (A\*r<LoAvx5a) she-called Mary and
(was) saying.' In connexion with the verb she went the
word silently is natural and forcible. It becomes paradoxical
in the place to which it is transplanted in D, most Old
Latin MSS. and Latin Vulgate (iiocauit Mariam. sororem
eius silentio).
Two other readings in this chapter may be noticed :
(i) xi. 14. AAZApOC 0 (|)lA0C HMCON AHCGANCN.
The only other authority, so far as I know, in which the
words 'our friend' are added here from v. 11, is Ephrem's
quotation from the Diatessaron, ' Lazarus our friend is dead '
(Hill, p. 367). Here then we have a Tatianic reading, due to
context-assimilation, preserved in D alone.
1 Apparently the only other authority in which the Bezan reading is found is
the Thebaic (ed. Woid).
^ In Matt. i. 19 the Old Syriac (Sin. Cur.) translates \a6pa. by ^(<Ll>caa
(quietly, silently).

SELECT PASSAGES FROM ST JOHN. 25
(ii) xi. 35. KAI eAAKpYccN 0 inc.
The added 'and' appears in Ephrem's quotation — 'And
our Lord wept^.' The addition is also found in N 61**69-
346 (the two last belonging to the Ferrar-group) 6'^", Old
Latin MSS., Lat.-vg. me. arm. aeth. There is no doubt that
the 'and' was added in an early Syriac text; but clearly
such an addition might arise independently in different
copies and versions.
Similarly in v. 48 D has kai can A<t)coMeN with the Sinaitic,
Ephrem, and the Peshitta — ' And if we suffer Him.' It is found
also in 235 me. aeth. In the same verse X*ff^ arm. join the
Sinaitic (which has the simple participle believing) and
Ephrem (' all men believe on Him ') in having the present
tense (true text TTiaTevaovaiv).
John xii. 32. era san YYtoGco
AnO THC THC cAKYCCO nANTA npOC CMAYTON.
The true text has iK in place of utto and reads TTdvTa<;
eXKva-co. The Peshitta'' has 'And-I, when that-I-have-been-
raised from the-earth, will-draw every-man (jtJ,.\A) to-me.'
Two points are to be noted : (i) The order in D agrees with
that in the Peshitta — ' I-will-draw every-man,' (2) Traj/ra?
(true text) = jcjA^ (cf. e.g. Matt. xix. 11, xxvi. 33, Lc. xxi. 17,
John i. 7, xiii. 35) = TTdvTa (D). Thus, when these two points
are considered together, it seems likely that TTavTa is the
masculine singular, a retranslation of the Syriac jcjI^. The
reading TrdvTa is found in X* 56. The Latins (Old Latin and
Vulgate) took ttuvtu as neuter plural: hence their omnia.
If this explanation is correct, this reading has a peculiar
value as being an instance of a Syrism in the Latin text,
which has clearly come through a Greek medium.
John xiii. 14. nocco maAAon kai ymgic 0(t)eiAeTe.
The true text has Kai v^ieh oifyeiXeTe.
^ Sin. and Pesh. have ' And-coming there-were His-tears (even) of- Jesus.'
2 Sin. Cur, wanting.

26 THE SYRO-LATIN TEXT OF THE GOSPELS.
The Sinaitic Syriac has : ' How-much-more (rdsa^) fit
for-you that-also ye &c.' With this reading that found in
Aphraat' (p. lA-i) and in the Arabic Tatian (' How much more
fit is it.'' Hill, p. 220) coincides. The Peshitta has: ' How-
much-more ye debtors (are) ye?' In the Bezan text then we
have here an Old Syriac and Tatianic reading.
This reading is found in aff^glm mm.
John xxi. 7. Aerei oyn 0 maGhthc CKeiNOc
ON HfAHA IHC TOJ nerpco 0 KC eCTIN HMCON.
The Syriac (Sin. Pesh.) for the last clause is : ' This our-
Lord is.' Here D reproduces the regular Syriac equivalent
of Kvpwi and 6 Kvpio<;, when applied to Christ ".
John xxi. 7, KAI hAato
eic thn GaAaccan.
The true text has 'i^aXev eavTov. The Bezan Latin has a
conflate reading : ' misit se et salibit' The reading of the
Sinaitic Syriac is : ' And-he-fell in-the-sea and-swimming
was-he and-he-came'.' The word 'fell' is not an unnatural
equivalent of the true Greek text; for in Matt. xxi. 21 the
Sinaitic has : ' If ye-shall-say to this mountain Be-taken-up
and-fall (= ^XijOtjti) in-the-sea*.' Is not the Bezan rjXaTo an
attempt to give a Greek rendering of the Syriac 'he-fell,'
without the extreme baldness of a literal translation ?
No other authority, so far as I know, shews any sign of
disturbance in the text at this point.
' Aph. inserts 'ye' before 'fit,' and omits 'also.'
^ The only other authorities which have this reading are the Aethiopic and
Persian versions. The latter is ' obviously made from the Peshitto Syriac '
(Scrivener, Introduction, vol. iii. p. 165).
^ The Pesh. adds here 'that-he-might-come to Jesus ' from Matt. xiv. 29, a good
instance of the assimilation which is so characteristic of the Syriac texts. See the
note on Jn. xxi. 13.
¦¦ So in Matt. viii. 32, Mc. v. 13 Sin. has 'fell into-(Mc. in-)the-midst-of the-
sea' ( = eii T171' BiXa,<saav).

SELECT PASSAGES FROM ST JOHN. 27
John xxi. 13. epxeTAi iHc
KAI AamBaNBI ton ApTON €YX*'PICTHCAC CACOKC
AYTOIC KAI TO O^AplON OMOICOC.
The true text has ...tov dpTov Kai BLBwatv avToh k.t.X.
The Sinaitic Syriac is as follows: ' And-He-took-up
(even) Jesus the-bread and-the-fish and-blessed upon-them and-
gave to-them.' The passage is evidently assimilated to the
accounts of the earlier miracles. Thus compare e.g. Mc. vi.
41 (Kai Xa^eov tow TTevTe dpTov<; Kai Toi)? Bvo Ix^va^ dva^Xe-
yjra<; eh tov ovpavov evXoyrja-ev Kai KaTeKXaaev tow dpTov;
Kai iBlSov Toh p,adr]Tah), which runs thus in the Sinaitic
Syriac : ' And He took-up these five loaves and-two fishes and-
looked to-heaven and-blessed and-brake the-bread and-gave
to-His-disciples '.' It will be noticed that, while the Sinaitic
has ' Yie-blessed', D has the synonymous expression evyapi-
aTrfda^. The Jerusalem Lectionary (p. 423) has ' Yie-gave-
thanks and-gave.' The Old Latin authorities have the word
used in the Sinaitic", the gloss in them taking two forms
(a) df et benedicens — a reading which looks like the translation
of a Greek aorist participle ; (b) g mm et bettedixit et.
Luke i. 79. anatoAh e2 yyoyc eni(})ANAi 0coc.
It appears that D is the only authority which adds ^w?.
The word would not unnaturally arise in a text assimilated
to a Syriac text. The Sinaitic has ' it-shall-make-light
(icQii^),' the Peshitta 'to-make-light (aicoi-SjA)'; the Syriac
versions, that is, here use the causative of the verb which is
of the same family as the Syriac word for light — p^ieocu.
' It is worth noting that whereas in the Greek two words are used to describe
the fish — ai IxBim and t6 b-^apiov — the Syriac has only one word.
2 The Greek and Latin of D, it will be seen, differ (euxap'f'riiffa!, benedicens).
Are they independent representatives here of the Syro-Latin text ? Or is the Latin
an inaccurate reading of the Greek? The participle (benedicens) of the Latin
favours the latter alternative.

28 THE SYRO-LATIN TEXT OF THE GOSPELS.
Luke ii. 5. ...BnGAeeM ^ AnorpA(t)ec6Ai
CYN MAplA TH 6MNHCTeYMeNH AYTCO
OYCH eNKYco Aia to ginai ayton ei oikoy
KAI nATpiAC AAYeiA.
D appears to be the only' authority which places the last
clause (Bid rb elvai avrov k.t.X.) after, instead of (as in the
true text) immediately before, the clause ' to enrol himself
with Mary &c.'
It will be convenient to give the whole passage in the
English : ' And Joseph also went up from Galilee, out. of the
city of Nazareth, into Judsea, to the city of David, which is
called Bethlehem, because he was of the house and family of
David ; to enrol himself with Mary, who was betrothed to
him, being great with child. And it came to pass, while they
were there, the days &c.'
I hope to make it probable that D here gives the Tatianic
order of the clauses, though it fails to reproduce what seems
to have been the Tatianic reading.
There is an important passage in Aphraat (p. .a-v.^, Bert,
p. 388), which is as follows : ' And Jesus was born from Mary,
the Virgin, from the seed of the house of David, from the
Spirit of holiness, as it is written that Joseph and Mary his
espoused (were^) both of them (»__oca*i4») from the house of
David! With this passage we must compare the following
from Ephrem's commentary on the Diatessaron (Moesinger,
p. 16) : ' Quodsi, quia Scriptura dixit : " Elisabeth soror tua,"
ideo hoc dictum esse putas, ut manifestaretur, Mariam esse
ex domo Levi, alio loco eadent Scriptura dixit, utrumque,
Josephum et Mariam, esse ex domo David!
These two passages' seem to make it clear that the text
of the Diatessaron (note ' as it is written ' (Aph.), ' eadem
' Except Sin. ; see the end of this note.
^ Cod. A inserts CVOCO.
•* They are brought together in Zahn, Forschungen zur Gesch. des Neiitesta-
mentlichen Kanons, i. Theil, p. 118.

SELECT PASSAGES FROM ST LUKE. 29
Scriptura ' (Eph.)) expressly asserted that Joseph and Mary
were both of Davidic descent. It would seem probable then
that in place of 'because he was (Pesh. .^cooAupc's A^-a
T^aeo) of the house and family of David,' Tatian by a very
simple alteration read ' because they were (^_oca»Av»t^.T Ayw
oacn) of the house &c.' But this emendation of the text of
the clause must have been accompanied by a change in its
position. As emended it could not stand before the words
' with Mary his espoused one,' for Mary had not been men
tioned in the previous context. It would naturally be placed
where it stands in the Bezan text, after the mention of
Mary. The evidence then of Aphraat, Ephrem, and D enables us
with great probability to restore a Tatianic reading. D, it
will be seen, witnesses indirectly to the reading, though it has
not preserved the actual reading itself The Bezan scribe
gives the Tatianic order of the clauses ; but he simply tran
scribes the true Greek text Bid to elvai avrov (not avrow).
I have left this note precisely as it was written some months
ago. The subsequent publication of the Sinaitic Syriac reveals
to us (i) the actual readings, (2) the order of the clauses in
an Old Syriac text. The passage is as follows : ' And-also
Joseph... [went] from Nazareth, a-city of-Galilee, to-Judaea to-
the-city of-David which(-is)-called Bethlehem, he and-Mary
his-wife while great-with-child, that-there they-might-be enrolled,
because that-both-of-them (»._octx»iA<.T Aj^'sa) from his-house
were (even) of-David! Thus the discovery of the Old Syriac
text entirely confirms the conclusion reached on critical
grounds as to the position of the clause Bid rb elvai k.t.X. in
the Old Syriac text of St Luke.
Luke ii. 48. iAoy 0 nATHp coy KAfto oAynoomgnoi
KAI AYnoYMeNOi ezHTOYMeN ce.
The Sinaitic Syriac has simply 'in-grief much seeking
were-we Thee.' The Curetonian however amplifies the phrase :

30 THE SYRO-LATIN TEXT OF THE GOSPELS.
' in-anxiety atid-in-grief^ much seeking were-we Thee,' Tatian,
as quoted by Ephrem (Hill, p. 337), has : ' Behold, I and Thy
father sorrowing (and) grieving were going about and seeking"
Thee.' In two Old Syriac texts then (the one using substan
tives, the other verbs) two words are employed as the equi
valent of a single Greek word. The Bezan reading coincides
with that of Tatian. This double rendering is found in many
Latin authorities — a e ff " 1 q r D'"^ G L gat.
It will be convenient to bring together the double render
ings found in the Bezan text of the Gospels, and to discuss
briefly this characteristic feature of the Syro-Latin text.
Matt. xix. 25. elenAHCcoNTO kai ect)0BHGHCAN ccfioApA.
The Sinaitic is fragmentary at this point. The Curetonian
has : ' Wondering were-they and-they-f eared much.' The
Arabic Tatian (Hill, p. 157) has in this place: 'And they
that heard were the more astonished, saying among them
selves, being now afraid. Who, think you, can be saved ? ' As
the Arabic Tatian here differs from the Peshitta, which has
not the interpolated words, we probably have here the
genuine Tatianic reading. The words et timebant are added
in a large number of Latin texts — a b c e ff " g" 3* L Q R.
Matt. xxv. I, eic ahanthcin toy nym(})Ioy kai thc nym(!)Hc.
This reading is rather of the character of a deliberate
interpolation than of a double rendering. It may however
for convenience sake be noticed here. The added words are
found in the Sinaitic Syriac (the Curetonian is not extant
here), the Peshitta^; also in X* i*-209 262*, the Latin MSS,
(lat-vt-vg), arm., Origen, Hilary : on the reading of the Ferrar-
group see Ferrar's note in loco.
^ Comp. the Arabic Tatian (Hill, p. 234): 'Their eyes were weighed down
for sorrow and anxiety' (Matt. xxvi. 43, Mc. xiv. 40).
^ The amplification ' going about and seeking ' should be noticed.
^ The added words are given in the Arabic Tatian (Hill, p. 214). But the
addition may be due simply to assimilation to Pesh.

DOUBLE RENDERINGS. 3 1
Luke viii. 8. eni thn thn thn apaGhn kai kaAhn.
Ephrem, as represented by the Armenian translator, gives
Tatian's reading as 'fat (and) good ground' (Hill, p. 350); in
the commentary (Moesinger, p. 125) he has the epithets in the
reverse order. The Curetonian has a second epithet derived
from the context. The words are : ' And-other fell on-ground
good and-giving fruit, and-sprang-up and-gave fruit a-hun-
dred-fold.' The Old Latin MSS. c e r have 'bonam et optimam,'
a has 'optimam et bonam! I believe that the Bezan Greek
and Latin are independent here, the latter having, it would
appear, a genuine Old Latin reading — ' bonam et uberam^ ' —
coinciding with, perhaps derived from, the reading preserved
in Ephrem's Tatian. The relation of D to this series of
readings it seems impossible to settle. It has perhaps em
ployed the epithet koXt^v, the epithet used in Matt. xiii. 8,
23, Mc. iv. 8, 20, to represent the interpolated epithet of some
version^ The evidence at our disposal at present does not
seem to take us further than this point.
Luke ix. 16. npocHY^ATO kai eYAornceN.
It seems that D alone has this double phrase. See
below, p. 36.
Luke xxiii. 28. mh KAAiere eMe mhAc neNGeire.
The true text has irr ip,e. In regard to this interpolation,
so far as I know, D stands alone.
Luke xxiii. 48, TYnroNrec ta cthGh kai ta MerconA.
No other authority, so far as I know, has this interpolation.
The Old Latin c however has frontes suas in place of pectora
sua. ^ Is it possible that in some Latin MS., high in the stream of descent, an
original opimam was emended into optimam ? In that case opimam and uberam
(d) might be divergent representations of the reading found in Tatian (Eph.).
2 Pesh. has: 'land good (T^ovaA,, the word used in Matt., Mc.) and-
beautiful ( r^OtT^^JCO )¦ ' This reading has the appearance of being a rendering
of the reading which we find in D.

32 THE SYRO-LATIN TEXT OF THE GOSPELS.
So far I have cited only such double renderings as are
found in Codex Bezae. I proceed to give those which are
found in the Old Syriac textual authorities, limiting myself to
the Gospel of St Matthew.
ii. 8. dnayyelXare fj,oi. ' Come shew-me ' (Sin. Cur, Pesh.).
iv. 5- TTapaXa/ji^dvei avrov 6 Bid^. eh rrjv dy. troXtv.
' The-devil led- Him and-made-Him-to-go to-the-city of-holi-
ness' (Sin. Cur.). Compare v. 8 TTapaXafi^dvei, avrov. ..eh opo^.
' Satan led-Him and-made(-Him)-go-up and-placed-Him upon
a-mountain' (Sin., not Cur.). Tatian (Eph., Hill, p. 339) has:
' He brought (Him and) took (Him and) set Him on a corner
of the temple... Brought Him (and) took (Him) into an ex
ceeding high mountain.'
V. 13. fimpavdrj. ' Be-insipid and-he-fooVish' (Cur.).
viii. 5. TrapaKoXSiv avrov. ' Seeking was-he from-Him
and-entreating Him ' (Cur.).
xii. 43. Biepxerai. 'Going (is it) wandering' (Sin. Cur.).
In Lc. xi. 24 (the parallel passage) the Curetonian (Sin. want
ing) uses the same paraphrase to render the same Greek verb.
xii. 44. et? rbv oIkov fxov iTTiarpe^^o). ' I-will-return I-will-
go to-my-house ' (Sin. Cur.).
xiv. 32. iKOTTaaev o dvep^o'?. ' The winds rested arid ceased '
(Tatian (Eph.), Hill, p. 352).
XV. 23. Kpd^ei OTTiaOev ^/j,S)v. ' She(-is)-crying and-coming
after-us' (Sin. Cur.). Tatian (Eph., Hill, p. 353) has: 'The
woman was crying out and following Him.' The Old Latin b
has : quia sequitur et clamat post nos.
xvi. 21. TToXXd TTadeiv. 'Endure much ^w^-suffer' (Cur. ;
Sin. wanting),
xvi. 21, aTTOKravdrjvai (so Lc. ix. 22). Tatian (Eph., Hill,
p, 357) has : 'The Son of Man must be crucified and die and
rise again.'
xxiv. 20 (so Mc. xiii. 18). rrpoaevx&aQe. 'V ray ye and ask'
(Tatian (Eph.), Hill, p. 370).
xxvii. 5, aTTrjy^aro. ' He hung and-was-strangled' (Sin.;
Cur. wanting). Tatian (Eph., Hill, p. 374) has : ' hanged
himself and died!

DOUBLE RENDERINGS. 33
xxvii. 41. ifiTral^ovTe<i. 'Mocking were-they at-Him and-
insulting were-they Him ' (Sin.). Tatian (Arabic, Hill, p. 247)
has : ' mocked Him and laughed to each other!
For other examples in the Curetonian see Baethgen, Evan
gelienfragmente, p. 1 5 f
An examination of these double renderings', which are
clearly characteristic of the Syriac texts of the New Testa
ment, shews that they are chiefly due to (1) the essentially
pleonastic character of Syriac ; (2) its inability to render
Greek compound words except by some kind of periphrasis ;
(3) a desire to bring out the full force of Greek prepositions ;
(4) the principle of assimilation — a potent factor in the Syriac
texts of the New Testament (see above, p. 31, on the Cure
tonian reading in Lc. viii. 8).
The evidence seems clearly to lead to the conclusion that,
speaking broadly, double renderings found in the Greek and
Latin authorities for the Syro-Latin text are derived from a
Syriac text.
It is instructive to compare Bp Lightfoot's statement as
to the characteristics of the Syriac version of Clement's
Epistle. Here at least Latin influence can hardly be a
factor". The Syriac version, he says (Clement, vol. i., p.
136 f), 'has a tendency to run into paraphrase in the trans
lation of individual words and expressions. This tendency
most commonly takes the form of double renderings for a
word, more especially in the case of compounds.' Bp Light-
foot proceeds to give a large selection of examples, e.g. § i
TTepiTTTwaevi lapsus et damna [impedimenta^ ; § 6 TraOova-ai
quum passi essent et sustinuissent [passi] ; § 1 5 M^^' vTTOKpiae(o<;
cum assumptione personarum et illusione [simulatores : sentence
recast] ; § 19 iTTavaBpdfiwfiev curramus denuo (et) revertamus
1 Compare Old Syriac Element, p. 78. I have there collected instances of
such double renderings in the Peshitta of the Acts — readings in which Pesh.
seems to stand alone.
^ I have in each passage appended to Bp Lightfoot's translation of the Syriac
version the renderings given in the newly discovered Latin version of Clement
(Anecdota Maredsolana, vol. ii.), enclosing them in square brackets.
C. 3

34 THE SYRO-LATIN TEXT OF THE GOSPELS.
[recurramtis'\ ; dreviamfiev videamus et cofitemplemur [intu-
eamur\ ' Sometimes however,' he continues, ' the love of
paraphrase transgresses these limits and runs into greater
excesses.' Among other illustrations he quotes § 2 1 /^^ Xitto-
raKrelv rj/xdi; aTrb rov 6eXi]fiaro<; avrov ne rebellantes et deserentes
ordinem faciamus aliquid extra voluntatem ejus [non desertores
nos esse a uoluntate illius\ ' The characteristic,' he proceeds,
' which has been noticed arose from the desire to do full
justice to the Greek. The peculiarity of which I have now
to speak is a concession to the demands of the Syriac. The
translation not unfrequently transposes the order of words
connected together: e.g. raTTeivocftpoa-vvT] Kai Trpavrr]<;  This
transposition is most commonly found when the first word is
incapable of a simple rendering in Syriac, so that several
words are required in the translation, and it is advisable
therefore to throw it to the end in order to avoid an am
biguous or confused syntax (the Syriac having no case
endings). Thus...Ta7reti'o</)pocrvi"?y is humilitas cogitationis!
Luke iii. lo, 12, 14. n noiHccoMeN ina ccoGcomcn.
In vv. 12, 14 the gloss is found only in D. In v. 10 however
bq gat G have the interpolation in the following form ut
tiiuamus. How are we to account for this double form of the
gloss ? The answer is clear when we turn to the Curetonian
(v. 10). We there read:
and-live or and-be-saved shall-we-do what
The Syriac verb to live is the constant equivalent in
the N.T. of the Greek aaydrjvat (compare above, p. 7, on Matt.
xvi. 16). Here then it is indisputable that the Latin autho
rities have incorporated in the text the translation of an
Old Syriac gloss. The Syriac word being capable of two
interpretations, we find one of these in D^, the other in the
Latin MSS. ; compare the note on Matt. xxvi. 60 (to e^fj<;),
p. 78 ff. The interpolation is doubtless due to assimilation to
' The analogy of e.g. Matt. xxvi. 60 is against, but does not exclude, the sup
position that the gloss first arose in Greek, and passed thence into the Syriac.

SELECT PASSAGES FROM ST LUKE. 35
Acts xvi, 30 (ri /le Bel TToieiv "va a-(o6(S ;), where the Peshitta
has: 'What is-it-necessary for-me to-do in-order that-I-may-
be-saved (<7r-live ; p^-mpS'.i vyr*")?'
Luke V. 7 f eAGoNTec OYN enAHCAN AMcfjorepA
ta nAoiA ojTe nApA ti ByGizccGau
^ 0 Ae ciMCON ^ npoceneceN aytoy toic nociN
AercoN nApAKAAco eleAGe An eMOY.
The true text is : /cat rfXOav, koL eTrXijaav dfi^orepa rd
rrXola ware ^vdl^eaOat avrd. IBwv Be Xip-asv Tlerpo<; TTpoae-
TTearev roh ybvacriv 'Iija-ov Xeywv e^eXde drr i/j,ov.
It will be convenient at once to give the words of the
Sinaitic Syriac and of the Peshitta (Cur. being wanting):
Sin. Pesh.
And-when they-came, And-when they-came,
They-got-aboard the-fish,
And-they-filled the-ships both-of- They-fiUed those ships both-of-
them, them,
And-near were-they from-their- So that-near were-they to-sink.
weight to-sink.
And-when there-saw(it) Simon, When there-saw(it) however Simon
Peter,
He-fell on his-face before the-feet He-fell before the-feet qf-Jesus,
of-fesus,
And-said to-Him, And-said to-Him,
My-Lord, Asking (am) I from- Thee my-Lord,
Depart for-Thee from-me. Depart for-Thee from-me.
The points in the Bezan text are three, (i) The TTapd
TL represents a phrase found (Cur. being wanting) in the two
Syriac versions. Compare the Arabic Tatian (Hill, p. 62)
' They filled both the boats, so that they were almost sunk.'
The naturalness of the Syriac phrase here used is clear when
we turn to two other passages. In Lc. viii. 23 (koI crvveTrXrj-
povvro Kai iKivBvvevov), where practically there is no variation
of reading in Greek or Latin authorities, we find the Syriac
texts having — ' And-there-was-filled their-ship and-near were-
3—2

36 THE SYRO-LATIN TEXT OF THE GOSPELS.
they to-sink ' (Sin.), ' And-there-was-filled their-ship from the-
waves, and-near was-it to-sink' (Cur.), 'And-near was the-
ship to-sink ' (Pesh.). Again in viii. 42 the Greek text is koi
avrrj aTTeOvrja-Kev, and, except that D reads aTrodvija-KOva-a,
there appears to be no variation of reading. The Syriac
texts however (Sin. Cur. Pesh.) have: ' And-near was-she Io
dic! The reading in Lc. v. 7, which a comparison of these
passages seems to stamp as indigenous in the Syriac, passed
over into the Old Latin c e g" r (ut pene. . .), arm.' (2) avrov
roh TToalv. In this form of expression D coincides with the
two Syriac texts. In the similar passage Mc. i. 40 (yowTrerSsv
avrov) the Sinaitic and Peshitta both have : ' He-fell at-(lit.
upon-)His-feet ' — a reading for which no other authority is
quoted. It would seem therefore that such a rendering of
'to fall at (on) the knees' was natural in Syriac. In Lc.
i-i 1 8-1 3 1-209 c me. join with the Syriac texts and D.
(3) irapaKaXSi. This addition, common to the Peshitta and
D, appears in Old Latin authorities in two forms — oro te ce,
rogo te / Compare Acts viii. 19. The eXOovrei, the first
word of the extract, points to retranslation.
Luke ix. 16. ANABAevfAC eic ton oyPanon
npocHYJATO kai CYAorHceN en aytoyc.
There are two points to be considered, (i) What of the
construction evXoyelv ivi riva " ? When we turn to the Cure
tonian (the Sinaitic is wanting here) we find a phrase of
which the Bezan Greek is a literal translation ,^_acaA^ *'\'*^
(He-blessed upon-them). We find the same Syriac con
struction in Matt. xxvi. 26, where the Sinaitic renders evXoy-
rjaa's 'eKXaaev by ' He-blessed upon-it (the bread) and-broke ' ^ ;
in Mc, viii. 7, where the Sinaitic translates evXoy^a-af avrd by
^ ' Ita ut inciperent mergi ' is the reading of the Memphitic.
2 The object after eiXoyeiv is expressed (a) in the N.T. by the accus. ;
(/3) in the LXX. by the accus. or more rarely the dative (e.g. Dan. v. 23, Ecclus.
1. 22). 3 Aphraat (p. !<lAi) has simply 'He-blessed and-gave.'

SELECT PASSAGES FROM ST LUKE, 37
the phrase 'and-also upon-them when He-had-blessed' — a
phrase retained in the Peshitta ('and-also upon-^em He-
blessed'). In Jn. xxi. 13 the Sinaitic has ' He-took-up (even)
Jesus the-bread and-the-fish and-blessed upon-therrv! We
compare also the Sinaitic in Lc. xxii. 19, 17 (evxap'O'T-no-af
eKXacrev, evxapicrTrja-a^ eStoKev) 'He-gave-thanks ttpon-it (jSOr^
^cnft \ 'h.) and-hrake.... He-gave-thanks upon-it and-said.'
There can, I think, be no doubt that here we have a Syriac
idiom reproduced in the Bezan text. (2) In view of the
frequency of double renderings in the Syriac New Testament
and of the fact that, where we have two Old Syriac texts of
a passage, we find such a rendering in one and not in the
other (see above, p. 32), we can hardly resist the conclusion
that, though the Curetonian has simply ' He-blessed upon-
them,' yet in the Old Syriac text, which lies behind the Bezan
text at this point, the phrase was ' He-prayed ««^-blessed
upon-them ^'
It should be added that the Syrism ' He blessed upon '
reappears in the Latin MSS. a b ff " 1 q r G (super illos; d super
eos). Epiphanius (p. 313, comp, p. 327) includes the reading
dva^Xe^Jra<; eh rbv ovpavov euXoyrjaev eTT avrovt in his list of
what he considers as Marcion's wilful corruptions of the text
of St Luke,
Luke X. 5.
eic HN AN Ae eiceAGHTC' npcoTON oikian
Aerere eipHNH tco oikoo toytco.
The true text has et? rjv B' av elaeXOr^re oiKiav Trp&rov
Xe76Te Eiip-ijvij ra o'lKfo rovrw.
The Old Syriac (Sin. Cur.) has: ' And-into-whatsoever
house the-first-one entering are-ye into-it, be saying Peace in-
the-house (Cur., this).' The TTpmrov of the true text could be
taken with the first part of the clause — 'into whatsoever
house ye enter first' The actual displacement of the word
1 For the construction ' to-pray upon ' (though the preposition here bears a
different meaning) see e.g. the Sinaitic of Matt. v. 44 (Trpo<reiix^(r0e virip k.t.X.).

38 THE SYRO-LATIN TEXT OF THE GOSPELS.
'first' to an earlier position in the sentence would be likely
to arise in a version. It did occur, as we see, in two early
Syriac texts. The reading, whether.it arose independently,
as would seem not improbable, or not, is found in more than
one form in Old Latin MSS. : a primum domum intraueritis,
b 1 q domum primum intraueritis, c primam domum, intraue
ritis primum.
The Old Syriac reading ' Peace in the house ' is found
elsewhere only, so far as I know, in the Ferrar-group — eipi'jvt}
iv T& o'lKW Tovrip.
Luke xi. 52 ff. oYAi fMem TOIC
NOMIKOIC OTI CKpYyATC THN KACIN
THC TNOOcecoc KAI AYTOI OYK ichAGatc
KAI TOYC eicnopeYOMCNOYC eKcoAYCAre
AepONTGc Ae tayta npoc aytoyc
eNOiniON HANTOC TOY AAOY HplANTO
01 (fjAplCAIOI KAI 01 NOMIKOI AelNCOC
exeiN KAI cynBaAAcin aytco nepi
nAelONCON zhtoyntcc A0OpMHN
TINA AaBbin aytoy INA eYpcociN
KATHfOpHCAl aytoy UOAACON Ae
oxAcoN CYNnepiexoNTcoN kykAco
cocTe aAAhAoyc cYNnNipeiN k.t. A.
The true text is as follows: oval  oVt -Sjpare rrjv KXelBa
TJ7? yvQxreox;' avrol ovk elarjXdare Kai Tov<i eiaepxo/Jievovi;
iKwX. KUKeWev i^eX06vTo<; avrov rjp^avro 01 ypafi/jLareii Kai
01 (f>. Beiv&v evexeiv Kai dTroarofiari^eiv avrov TTepl TrXeiovcov,
eveBpevovre^ avrbv Oripeva-ai rt e/e rov <Tr6fJM,ro<s avrov. iv oh
iTTiavvaxdeicTwv roov fivpidBcov rov oxXov, ware KaraTTareiv
dXXijXov; K.T.X.
It will be convenient at once to give the Old Syriac, The
Sinaitic and the Curetonian agree here, except that the
former omits the clause, which is printed below in italics.

SELECT PASSAGES FROM ST LUKE, 39
'Woe to-you Scribes because-ye-hid («^_aAuJtA^.i) the-keys
of-knowledge. Ye entered not, and-those who-entering (were)
ye-hindered, And-while saying was-He these-things against-
them in-the-presence-of all the-people. He-began abominated
was-He to-the- Scribes and-to(om. Cur.)-the-Pharisees, and-
disputing were-they with-Him about many-things, and-seeking
were-they to-take (,Ti*r^aaA) against-Him a-cause (»<'AA.^)
that-they-might-be-able (or find) they-should-accuse Him. And-
when there-assembled unto-Him a-multitude great, so-that
they-trod one on-one &c.'
The chief points in the passage are these: (i) iKpv-^are
in place of ijpare, the former being found in 157, in the Old
Latin MSS. abed (abscondistis) e q (absconditis), and in
the Armenian ^ The reading is also found in Ephrem's
Commentary on the Diatessaron : ' Woe unto you, lawyers,
for ye hide the key.' It is clear that the reading is not
due to the Armenian translator of Ephrem, but is really
Tatian's, because Ephrem comments on the word. So too
Ciasca (not Pesh.), 'ye have hidden the keys' (Hill, pp. 203,
369). The concurrence of the Sinaitic, the Curetonian, and
Tatian goes far to shew that this was the primitive Syriac
equivalent of ijpare. (2) The Bezan text exactly coin
cides with the Old Syriac in the words XeyovTO'i...rov Xaov.
In the words which follow (rip^avTo...'exeiv) the Bezan scribe
in the main gives the true text substituting (a) vo/niKoi for
ypafifiareh' and (b) e^eiv for ivexeiv. (3) The clauses in
the true text d7rocrTOfiari^eiv...iK rov arofi. avrov are less
simple than is commonly the case with the Gospel narrative :
hence in a version they were almost certain to be more or
less paraphrased. I believe that an examination of the Old
1 The verb is in the present tense in the Armenian version of Ephrem's
Commentary on Tatian as in the Armenian Vulgate (see Hill, p. 369), and as in
the two Old Latin MSS (e q). The Aethiopic has a conflate reading — 'ye took
away and hid, '
^ This is perhaps due to the last 'woe' (v. 52). It should however further
be noticed that Sin. Cur. Pesh, translate i>oiuk6s by the word used to translate
ypafifiaTeiis except in Matt. xxii. 35 (Sin. Pesh.), Lc. x. 25 (Sin. Cur.).

40 THE SYRO-LATIN TEXT OF THE GOSPELS.
Syriac and the Bezan texts shews that the former is a natural
paraphrase of the true text, and that the Bezan is a natural
representation of the Old Syriac text. We may take the
points in order : (a) dTToarofiari^eiv avrov. The verb is an
unusual and ambiguous one. It would be natural roughly
to represent it in Syriac by an expression used in a like con
nexion elsewhere. This the Old Syriac does by the words
oocD ^AX.i.-icv (and-disputing were-they), the word JL-i.i being
used in Mc. ix. i6 (a-vv^Tjrelre), xii. 28 (avrcjv crvv^rjrovvrav).
Acts vi. 9, ix. 29, xvii. 18, xviii. 28. Further, the word
(TVfji^dXXeiv is a natural representation of the Syriac word,
the latter in fact being used in the Peshitta to render
avfi^dXXeiv in Acts xvii. 18. (b) iveBpevovTe<; avrbv Orjpevcrai
Tl iK rov err. avrov. Such words were sure in a version
to sink into commonplace. The word ' seeking ' is made to
do duty in representing iveBp. avrov. Again, the words Grip.
Ti iK TOV err. avrov are toned down into ' to-take against-Him
a-cause that-they-might-be-able (find) they-should-accuse
Him.' It will be remarked how perfectly natural the Syriac
word r^AiL^ (a-cause) is in the forensic sense, being defined
by the subsequent clause ' that they might be able to accuse
Him.' On the other hand the Bezan d(j)op/j,^v is not suited
itself to the context, but would most naturally arise from
the Syriac word, this latter being its equivalent in the
Peshitta every time d^opfirj occurs. Further, the avrov of
this clause suggests retranslation by a bungling hand. The
defining clause ' that-they-might-find, &c.' comes from Lc. vi.
7, where Greek and Syriac are the same as here '. (4) The
last two lines in D differ from the Old Syriac in three
respects : (a) the latter has the singular ' There-assembled a
great multitude ' : but, as the noun and adjective can both be
vocalized as plural, the addition of an unpronounced o to
the verb, making it the 3rd person plural, brings the Old
1 There is however a difference of reading in regard to one word. Many MSS.
have Kwr-qyopiav. All other MSS. (except D, which has the aorist mfin.) read xar-q-
yopeiv. The same phrase occurs in Cur., alone of all authorities, in Matt. xii. 10
('iva Karriyoprqaaffiv avrbv). Sin. being here wanting.

SELECT PASSAGES FROM ST LUKE. 41
Syriac into harmony with the Bezan text^; (b) the Syriac
has nothing to answer to the kvkXw reinforced by the -rrepi-
of the compound verb ; (c) the Syriac has ' so-that they-trod
one on-another' ; D has mare avvTTvi^eiv. This reading, whether
it arose in a Syriac or in a Greek text, is due to assimilation
to Lc. viii. 42 (ot o;;^\ot avveTTViyov avrov).
It remains to add a few notes to shew how the Syriacised
text of the passage (v. 53 f ) spread.
(i) The Ferrar-group have coincidences with D. e^eti^ (for
evexeiv) is found in 1 24, avfi^dXXetv (for aTToarofi.) in 69.
(ii) I subjoin the text of the Old Latin Cod. Brixianus (f),
noticing the chief variations in other MSS. ' Cum haec ad illos
diceret coram omni populo (plebe, b i 1 q ; in conspectu totius
populi, ce) coeperunt pharisaei et legisperiti (legis doctores,
c e ; tam scribae quam et legis doctores, a) contristari (male
(-1- se, a) habere, a b q ; grauiter habere, c e i ; grauiter ferre, 1 ;
moleste ferre, r) et altercari cum illo (comminare illi, a ; com-
mittere cum illo, b i 1 q r ; conferre cum eo, c ; conferre illi, e)
de multis (de pluribus, ace) interrogantes eum quaerentes
(4- de multis, b) capere aliquid ex ore eius. ut occasionem
inuenirent accusare eum (occasionem aliquam inuenire ab illo
(de illo, b q ; in illo, i ; om. eel) a b c e i 1 q r). The variety of
rendering in the Latin texts seems to indicate that they are
different representations of a common original.
Luke xiii. 11. kai iAoy tynh cn acGcncia hn
Wnc en- iH.
The true text has Kai IBov yvvrj Trvevfia exovaa da0eveia<;
err) BeKa oktco. The character of the reading suggests retransla
tion from the Syriac. For Syriac has no word which exactly
represents ej^et. Hence any such phrase as Baip-oviov exei
has to be paraphrased in Syriac, e.g. ' a devil is to (in) him '
(Matt. xi. 18, Mc. iii. 11, 30, Lc. iv. 3$, vii. 33, viii. 27); and
' The Arabic Tatian (Hill, p. 206) has : ' Now when many multitudes were
gathered together,' As this differs from the Peshitta (' and-when there-were-
assembled (plur.) an-abundance of-crowds inahy ') it very probably represents the
Tatianic reading.

42 THE SYRO-LATIN TEXT OF THE GOSPELS.
in the present passage the Old Syriac (Sin. Cur.) has ' to whom
there was a spirit ' (Pesh. adds ' of infirmity '). Now in (a) the
statement as to the number of years, and (b) the use of the
word 'infirmity' there are points of affinity between the
present passage and Jn. v. 5 (rpidKovra oktw 'errj ex^v iv rfi
daOeveia avrov^). In Jn. v. 5 the Peshitta (Sin. wanting;
Cur,, 'who. ..was infirm') has: 'There was there a-man
a-certain-one who-thirty and-eight years was («cno4v*K'
i<'ocn) in-infirmity^' We know how potent a factor assimi
lation was in the Syriac texts. Hence in view of Jn. v. 5 a
Syriac reading might easily arise in Lc. xiii. 11 — 'who-in-
infirmity of-spirit was eighteen years.'
Three other passages may be noticed where this Syriac
mode of dealing with exei seems to have affected other texts :
(i) Jn. ii. 3 ohov ovk exovaiv. This necessarily becomes in the
Peshitta (Sin. Cur. wanting) 'Wine is-not to-them,' X (whose
' Western ' readings deserve careful attention) has oii/o? ovk
eariv. (2) Lc. xix. 34 o Kvpio<i avrov xpe^av exei. The
Sinaitic and Curetonian have : ' For-his-Lord (Pesh. for-our-
Lord) required (is he).' The Old Latin a f have : " domino
(-1- suo, a) necessarius est.' (3) Mc. viii. 17 en TTeTTtopm-
(levrfv exere rrjv KapBiav vfimv ; The Sinaitic is not extant
here. The Peshitta has : ' Still the-heart hard is-it to-you ? '
D 2*^ have nenoopcoMeNn' ecTiN h KApAiA ymcon ; Among the
Latin MSS. (i) fg"lvg. have: caecatum habetis cor uestrum?
(ii) a q obtusum est cor uestrum ? (iii) b c d ff"^ i obtusa sunt
corda uestra ?
Luke xiii. 17. KAI nAc 0 oxAoc
exAipeN eN hacin oic eGecopoYN
^ eNAo2oic ^ Yir aytoy reiNOMeNoic.
^ Similarly in L Matt. ix. 20 is assimilated to Jn. v. 5 ; for after the words
SiiSexa ir-q L adds cxoi;<7a tv rq a/rOevqa.
2 The similarity of Lc. to Jn. is more striking in the Syriac than in the Greek;
for the Syriac (Sin. Cur. Pesh.) has in Lc. -. ' There was there (om. Sin.) a-woman
a-certain-one (om. Pesh.).'
^ D has Treir"ptiiix€vq, the correction apparently being made by the original
scribe.

SELECT PASSAGES FROM ST LUKE. 43
The true text is Kai Trd<; 6 oj^Xos exaipev ivl rrdaiv roh
ivBo^oig roh yivofxevoi'i vv' avrov. What account can be given
of the interpolated words oh ided>povv ? When we remember
the love which the ' Syro-Latin ' text has for assimilation, we
can hardly doubt that they are derived from the very similar
passage in Lc. xix. 37, fjp^avro dwav rb ttXtj^o? twv fjLaOrjrwv
XO'i'Povreii alveiv rbv debv (poovy fieydXrj (D om. <J3. fi.) TTepl
TTaacov wv elBov Bwdfjieav (D nepi hantcon con eiAoN reiNOMCNCON).
But the Bezan form of the interpolation must have come
through the medium of a version. This version cannot be the
Bezan Latin ; for that slavishly follows the Greek idiom : ' in
omnibus quibus uidebant mirabilibus ab eo fieri! We accord
ingly turn to the Syriac. The Curetonian (the Sinaitic being
illegible) has in xiii. 17: 'And-all the-people rejoicing was
in-aW the-wonders which-being(done) were in-His-hand.' The
preposition ' in-a\\! answering to the Bezan iv rrdaiv, will
be noticed. In xix, 37 the Old Syriac (Sin. Cur.) has:
'There-began all the-crowd of-the-disciples (om. Cur.) rejoicing
[were-they, Cur.J and-praising God with a-great voice about
everything which- they -saw (ovui "pxsa A& A^.).' If in an
Old Syriac text the single word ovmI (which-they-saw) were
interpolated in xiii. 17 after the word 'wonders! the passage
would read thus : ' rejoicing in-all the-wonders which-they-saw
that-being(done) were-they in-His-hand.' Thus the insertion
of the word 'which-they-saw' is very easy, and it at once,
without any alteration of the surrounding words, takes a
natural place in the sentence. Of that Syriac sentence the
Bezan Greek is a natural rendering. The Bezan scribe would
be likely to translate the Syriac ot** by Oempovaiv, for this
Syriac verb is the constant equivalent of this Greek verb
(see e.g. Matt, xxvii. 55, xxviii. i, Mc. iii. 11, v. 15).
The theory that Lc. xiii. 17 was assimilated to Lc. xix. 37
in an Old Syriac text is confirmed by the fact that in the
Curetonian text of Matt. xxi. 9 (Sin. is wanting here) we
find an interpolation based largely on Lc. xix. 37. The verse
is as follows: '...Hosanna in-the-highest. And-there-went-

44 THE SYRO-LATIN TEXT OF THE GOSPELS.
out to-meet-Him many, and-rejoicing were-they and-praising
God about all that which-they-saw (atw.i rclso Aa Aj^),'
This interpolation seems to be found elsewhere only in ^
(Codex Purpureus) : waavvd iv roh ir^iaroi';' uTT'^vrcov Be avrw
TToWot ;jj;aipofTe? koI Bo^d^ovre<; rbv debv TTepl irdvrwv wv
eiBov (MS. iBov). Kai eiaeX66vTO<; k.t.X!
The interpolation in Lc. xiii. 17 has found its way into
Latin texts : in praeclaris quae uiderant fieri ab ipso, b c (eo)
ff '^ (uidebant) i 1 q r (— ab ipso fieri) ; in omnibus quae uidebant
praeclara fieri ab illo, e; in uniuersis praeclaris uirtutibus
quae uidebantur fieri ab eo, I. The diversity of phrase
seems to imply that the Latin texts present here various
attempts to render a common original.
Luke xiii. 24 f.
on noAAoi Aerco ymcin zhthcoycin
eiceAGeiN kai oyx eYpncoYciN : a(}) otoy
AN 0 oiKoAecnoTHC eiceAGH kai
AnOKAeiCH THN GYpAN.
The true text is : on ttoXXol, Xeyco v/xiv, ^rjT'qaovaiv
eiaeXOeiv Kai ovk laxvaovaiv, def)' ov dv iyepOfj 6 olKoBeaTTOTTj';
Kai uttokX. TTjV 6.
The points are: (i) The reading ovx evptjaovaiv is
peculiar to D. The Syriac rendering of the true text ovk
laxvaovaiv is »__ajjL^zJt r<lA. The Syriac verb means both
' to be able ' and ' to find.' Hence the Bezan ovx evpijaovaiv
is a natural retranslation of the Syriac. (2) The reading
eiaeXdrj is found in the Ferrar-group, in most Old Latin
MSS., and in the Latin Vulgate. It is doubtless due to
assimilation to the very parallel passage in Matt. xxv. 10,
ai eroifioi elarjXOov fier avrov eh tou? ydfiov^, Kai eKXeiaOtj
¦fj Ovpa. Just below, the doubled Kvpie (Kvpie, Kvpie dvoi^ov
' It will be observed that the interpolation in Cur. is the original of 0 and not
vice versa ; for the words ' there- went-out to-meet-Him' (Cur.) are precisely those
of the parallel passage (Jn. xii. 13) as given in Sin. Pesh., and their origin is thus
accounted for. The Greek (Jn. xii. 13) is i^rjKBov els iirivTTiinv airifi Kai iKpaiya-
^ov (ji(7avvd.

SELECT PASSAGES FROM ST LUKE. 45
ij/iti/), found in a very large number of MSS., is derived
from Matt. xxv. 11.
Luke xiv. 9. kai rore ecn MerA
AICXYNHC ^ ecXATON TOHON KATeXCIN.
The true text has dp^y instead of eay, and inserts rbv
before eaxarov.
The Bezan Latin preserves the true text— 'et tunc incipiens
(= incipies) cum confusione nouissimum locum tenere.'
The Old Syriac (Sin. Cur.) represents the ap^rj...Karexeiv
by the simple future — 'And-then while ashamed thou shalt-
recline in-the-place the-last-one.' Elsewhere however the
Syriac versions represent a similar Greek phrase by the verb
'to be' followed by the participle'. Thus in Lc. xiii. 25
(Kai dp^Tjade e^w kardvai Kai Kpoveiv rrjv Ovpav) the Old
Syriac (Sin. Cur.) and the Peshitta have: ' And-ye-shall-be
standing without and-knocking at-the-door.' Again, in xiv. 29
(Jva fjLy...dp^(ovrai avra ifiTTai^eiv) they (Sin. Cur. Pesh.) read :
' that-not... they-be mocking at-him.' In Lc. xiv. 9 therefore
the Bezan construction (earj...KaTexeiv), which could not
arise simply in Greek, seems to witness to a similar rendering
of ap^r)...Karexeiv in an Old Syriac text lying at this
point behind the text of D — ' and-then thou-shalt-be reclining
in-the-place the-last-one.' The Bezan scribe, instead of
writing Karex^v, has suffered the Karkxew of the true text
to remain. The Old Latin e, the constant companion of D,
has a reading which exactly answers to that of D — ' et tunc
eris... tenere! The chances are infinite against this reading
having arisen independently in two allied texts. We are forced
to the conclusion that e has here simply translated the Greek
phrase which we find in D, Thus we are led once more to
notice the remarkable kinship which subsists between D and e.
' Similarly the Old Syriac neglects ju^XXei. Thus in Lc. xix. 4 (on eKctvr)s
ij/ieWev SUpxeirBai) Cur. has ' because thus passing was Jesus.' In John vii. 35
(iroO oBtos /iiWei TropeieaBai ;) Sin. and Cur. have ' Whither then going(is) this-
man?' The word Set is dealt with in a similar way in Matt, xxiii. 23 (Sin.
Cur.), Lc. xi. 42 (Cur., Sin. has the proper equivalent of Set), xviii. i (Sin. Cur.).
See Baethgen, Evangelienfragmente, p. 14.

46 THE SYRO-LATIN TEXT OF THE GOSPELS.
Luke XV. 4. KAI AneAOcoN to ahoAcoAoc zhtci.
The true text has Kai TTopeverai e'Trt to aTroXwXo?.
The Old Syriac (Sin. Cur.) and the Peshitta have : ' He-
goeth (Atp^) seeketh that which-perished.' Here there are
two points: (i) The Bezan drTeXdrnv is clearly derived from
the TTopeverai of the true text ; but it arises through the
medium of a version. The Syriac Airs' is the natural
rendering of TTopeverai (see e.g. Matt. ii. 8, 9). But the
verb aTTeXdelv is an equally natural retranslation of this
Syriac verb (see e.g. Matt. viii. 18, 19, 21, 33). (2) The
words ' goeth seeketh ' is a characteristic Syriac periphrasis
to bring out the meaning of the preposition e'^rt. Note the
following renderings in the Old Syriac — Lc. viii. 33 (wpnyaev
...Kard rov Kprjfivov eh ttjv Xifivrjv) 'there-rushed all that
flock to-the-precipice and-they-f ell in-the-sea ' (Sin. Cur.) ;
xix. 29 (riyyiaev et? ^r)6(f)ayrj...'7Tpb<; to opo<;) ' He-came to-
Beth Phage... and-came to the-mount' (Sin. Cur.); xxiv. 5
(KXivovawv rd TTpoawrra eh rrjv yrjv) ' They-bowed their-heads
and-looking were on-the-earth ' (Sin. Cur.) ; Jn. i. 42 (ijyayev
avrbv TTpb<; rbv 'Irjaovv) ' He-led-him and-came to Jesus '
(Sin. Cur.); Jn. iv. 35 (XevKal elaiv tt/so? 6epiap.6v) 'they-are-
white and-they-have-come to-the-harvest ' (Sin. Cur.). See
above p. 32 f., and Baethgen, Evangelienfragmente, p. 17.
In the present passage the periphrasis, which we have
seen to be characteristically Syriac, has passed in different
forms into the Old Latin MSS. : — a e uadit ad illam quae
perit (e perierat) quaerens ; f uadit quaerere earn quae errauit.
The Bezan Latin is : uadit et quaerit quod perierat,
Luke XV. 29 f. KAI OYAenoTe
nApeBHN COY eNToAHN kai OYAenore
eACOKAC MOI epl(t)ON eZ AlptON ina MeTA TCON
(|)lACON MOY ApiCTHCCO TCO Ae YI<J^> COY
TO) KA(J)ArONTI HANTA MCTA TCON nOpNCON
KAI eAGoNTi eGYCAC ^ ton ceireYTON mocxon.

ASSIMILATION TO THE OLD TESTAMENT. 47
The true text is koX ovBerrore ivroXtjv aov TraprjXOov, Kai
efioi oiiBeiTore eBcoKa^ epi<j)ov iva...ev(f>pav6w. ore Be 6 vi6<; aov
ovro<; o Kara^ajwv aov rbv 0iov /jierd [rwv^ TTopvwv rjXdev,
eovaai avrm rbv airevrbv fibaxov.
Two points call for attention, (i) We have here a good
example of a passage rewritten — the ideas preserved, the
language altogether changed. The phenomena imply re
translation. The Sinaitic and the Peshitta have^ : ' And-not
(ever, Pesh.) did-I-transgress against (om. Pesh.) thy-com-
mands (thy-command, Pesh.), and-from ever one kid (a-kid,
Pesh.) not didst-thou-give to-me that-I-might-be-merry with
my-friends ; and-this thy-son (to-this-man however thy-son,
Pesh.) when he-had -devoured thy-property with-harlots (and-
came, Pesh.), thou-didst-kill for-him that calf (the-calf, Pesh.)
of-fatting,' It will be seen that the recasting of the clauses in
the Sinaitic and the Peshitta is quite natural in a Syriac
translation, and that this form of the clauses is pretty closely
followed in D. The Trdvra of the last line but one seems
due to context-assimilation to v. 31 (Trdvra rd ifid) ; this
suggestion is confirmed by the fuller form in which the
reading is preserved in that constant ally of D, the Old
Latin e : ' filio autem tuo qui comedit omnia tua cum forni-
cariis adueniente laniasti saginatum uitulum.' (2) The
reading epi(f>ov i^ alywv is important as being a clear instance
of the assimilation in the Syro-Latin text of a passage in
the New Testament to the language of the Old Testament
(see Gen. xxvii. g, xxxviii. 17, 20, Judg. vi. 19, xiii. 15). The
phrase itself is too simple to reveal in what language the
assimilation was first made.

I take this opportunity of bringing together some passages
from Syro-Latin texts of the Gospels in which we can clearly
trace assimilation to the Old Testament.

' Cur. is wanting here.

48 THE SYRO-LATIN TEXT OF THE GOSPELS.
(i) Luke iii. 22. KAI (J)O0NHN CK TOY OYPANOY
reNecGAi YIOC moy ei cy a epm CHMepoN
rercNNHKA ce.
The interpolation from Ps. ii., as is well known, is found
in connexion with the Baptism in many early authorities — the
Old Latin MSS. abcfi"M (Lc. iii. 22), Justin Dial 88, 103,
Clem. Alex. Paed. i. 6, the ' Ebionite ' Gospel quoted by
Epiphanius, p. 138V It will be noticed that the words 'Thou
art My Son' are common to Lc. and the Ps. I cannot doubt
that these words were a link between Lc. and the Ps., which
led to the insertion in Lc. of the clause from the Ps. On the
use of the Old Testament in the Early Church see below, p. 5 1 f.
(ii) Mark x. 1 1 f.
oc AN ahoAych thn tYnaika aytoy
KAI aAAhn tamhch" moixatai en aythn
KAI eAN rYNH eSeAGi- auo toy ANApOC
KAI aAAon tamhch" moixatai.
For the words underlined the true text has idv avrrj
aTToXvaaaa rbv dvBpa avTrjt; yafiijarj dXXov. The Bezan
i^eXdrj is found in the Ferrar-group 28 2"° 604, a discesserit,
b exiet, ff^ exeat, q exierit ; compare c reliquerit, k relinquit.
There can, I think, be little doubt that it is due to assimilation
to Jer, iii. i ' They say. If a man put away (n7tJ'*) his wife,
and she go from him (IDf^D nDTTTl), and become another
man's, shall he return unto her again ? '
In what language did the reading in question arise ? The
versions are as follows :
LXX. : idv i^aTToareiXr) dvrjp rrjv yvvaiKa avrov, Kai diriXOtj
aTT avrov Kai yevrjTai dvBpl erepw k.t.X.
Latin Vulgate (Cod. Amiatinus) : Si dimiserit uir uxorem
suam, et recedens ab eo duxerit uirum alteram.,.
' For later Patristic evidence see Resch, Agrapha, p. 346 ff.

ASSIMILATION TO THE OLD TESTAMENT. 49
Syriac :
(io A\t^i«o cD^&uiK' rc'va-.^ j30.a.xJ »__t^o
from and-she-go-away his-wife a-man there-put-away and-if
.r<liiAH<' t^ja-^A rS'oeoAio 03^a\
.another to-a-man and-she-be with-him
In regard to the Syriac it should be added that Aphraat
(p. ¦\y\n) quotes Jer. iii. i in the following form^:
coitoA ^ jao&^o r^hxhxir^ rt'isL^ .-\tn.i n^
with-him from and-she-go-out a-wife a-man there-taketh when
.rtfji*»r^ rS'i.n.^k r<'ocnA\0
.another to-a-man and-she-be
It will be noticed that the Syriac of Jer. has two points of
contact with the Syriac of Mc.''; (i) the word jar. (Sin.)
answers to dvoXvaai (Mc, comp. e.g. Matt. v. 32, xix. 9) and
is used in Jer. ; (2) the phrase ' to be to another man,' in the
sense of 'to marry another man,' is common to Mc. (Sin.
Pesh.) and Jer. The evidence does not perhaps warrant a
decided verdict, but it certainly points to the Bezan reading
having arisen in an Old Syriac text.
(iii) Luke xxi. 25 (Kai c'tti t^? 7^? avvoxv idvwv iv aTTopia
rixov<s 6aXdaar)<; Kai aaXov, aTTO'yfrvxbvrwv avdpcoTTWV drTO <f>6^ov
K.T.X.). The Sinaitic Syriac has : ' And-distress on-the-
earth and -feebleness -of hands (rdj.tiK' Ajcaio) of-the-
peoples &c.' It is clear that the Syriac translator had before
him, or translated as if he had before him, a Greek text
as follows : e'vrt t^? yr\'; avvoxi], idvwv d-TTopia, and that he
represented this last word by the paraphrastic expression
' feebleness of hands.' What is the source of this phrase ?
I cannot doubt that it is a reminiscence of descriptions in
the prophets of the coming of judgment — Ezek. vii. 15 ff,':
^ The words ' when a man taketh a wife' are due to assimilation to Deut. xxiv. 1.
2 Sin. has (Mc. x. 11) : 'Wbatsoever woman (is) putting-away (f<lrLaJt.)
her-husband and being to-another-man (is) committing adultery.'
^ The whole passage in Ezek. should be compared with the context in Lc.
C, 4

50 THE SYRO-LATIN TEXT OF THE GOSPELS.
'The sword is without and the pestilence and the famine
within... All hands shall be feeble (^Ix-iAo re'-..T»*<'), and all
knees shall be weak as water ' ; xxi. 7 ' And it shall be,
when they say unto thee, Wherefore sighest thou } that thou
shalt say. Because of the tidings, for it cometh : and every
heart shall melt, and all hands shall be feeble (^cnl^ ^ix.ii\sn
r^j.T»r<'), and every spirit shall faint.' Compare Jer. 1. 43 ' The
king of Babylon hath heard the fame of them, and his hands
wax feeble (Ai.iAxr*').' Thus in the Syriac version of the
Old Testament a certain phrase is used in the pictures of
men's fear of divine judgment ; in the Syriac Gospel in our
Lord's discourse on the coming woes, an adaptation of that
phrase is employed, the verb giving place to the correspond
ing substantive.
It would seem that the Curetonian reading' vyJLCL^
r^.'UrS' (wavering of hands : see Brockelmann, Lex. Syr.) was
derived from that of the Sinaitic. Syriac words from the
root \^x.& are the constant equivalents of Greek words
belonging to the same family as dtropia (see Payne Smith,
Thes. Syr.). The word v^a& (wavering) seems to have
been substituted for Ax.oi (feebleness), the two words having
the same general meaning, but the former being nearer to
the Greek aTropia.
(iv) Luke xxiii. 9 (at/ro? Be ovBev dTTeKpivaro avrw). The
Old Latin c adds quasi non audiens. The source of this
gloss is suggested to us by a passage of Cyril's Lectures (Cat.
xiii. xvi.), where he is speaking of our Lord's silence before
Pilate : Kai 6 'Itjo-oO? iaiwTTa. Xeyei b '\fraXp,(pB6<; Kat iyevo/jbrjv
wael dydpcoTTot ovk aKovwv Kai ovk e^wi' iv ra arofian avrov
iXeyfiov<i (Ps. xxxvii. 1$)'.
With these passages, where the language of the Gospels is
' So Pesh. The Arabic Tatian (Hill, p. 211) has 'wringing of hands.'
' It is impossible not to connect this gloss in c with the gloss in the Curetonian
Syriac : ' But Jesus returned him not any answer, as though he had not been there.'
But on the relation between the two glosses it is vain to speculate. On a some
what similar phrase in the Gospel of Peter see Old Syriac Element, p. 123.

ASSIMILATION TO THE OLD TESTAMENT. 5 1
assimilated to that of the Old Testament, I may be allowed
to refer to the Bezan reading in Acts xii. lo. I have pointed
out elsewhere (Old Syriac Element, p. 86) how naturally the
appearance of the angel and St Peter's guidance by the angel
through the precincts of the prison would recall Ezekiel's
vision (Ezek. xl.) of the supernatural being who guided him
through the precincts of the Temple (vv. 5 ff.). It must
suffice here to place the two passages — Ezek. xl. 6 (22) and
Acts xii. 10 as it appears in Codex Bezae — side by side.
Ezek. xl. 6. Acts xii. 10 (D).
Then came he unto the gate hAGon em thn hyAhn thn ci-
AnpAN
which looketh toward the east, thn <|j6poYCAN eic thn noAiN

and went up the steps thereof.
and they went up unto it by
seven steps ^ (v. 22).

HTIC AYTOMATH HNYfH AYTOIC
KAI eleAGoNTBC katbBhcan TOYC
•z- BaGmoyc.

For other probable or possible cases of assimilation to the
language of the Old Testament in the Bezan text of the Acts
see Old Syriac Element, pp. 32 (Acts iii. 3), 60 (v. 38), loi
(xix. 29).
The interweaving into the text of the New Testament of
phrases taken from the Old Testament is seen to be most
absolutely natural, when we realize the position which the
Old Testament occupied in the Christian Church in the
second century — the century when the ' Syro-Latin ' (or
' Western ') text of the New Testament was gradually taking
shape. 'The Old Testament was still the great storehouse
from which the Christian teacher derived the sources of
consolation and conviction''.' At least in the earlier part of the
^ The LXX. introduces the numeral (h iirrci, ivapa8/u>ts) in v. 6. So also the
Codex Syro-Hexaplaris Ambrosianus (ed. Ceriani), which has in ». 6 ' And-he-
entered that gate which-looketh towards the-east by-the-seven steps (r<ls».ax.3
ri::iO."t).' ^ Bp Westcott, Introduction to the Study of the Gospels, p. 169. The works of
Justin Martyr are the best commentary on this statement. Compare also e.g.
Ignatius, Magn. ix., Philad. v., viii., ix, Smyr. v. ; Hegesippus (Eus., H. E., iv. 22)
iv iKdarri iroX« oVtus ^X" "' ° vojios Kffpitiiyet Kai oi vpo^rai xal i xipios.

52 THE SYRO-LATIN TEXT OF THE GOSPELS.
century it was the Books of the Old, rather than those of the
New, Testament which were regarded as possessed of primary
authority. With these the apostolic writings were gradually
becoming coordinated. The harmonies even in small points,
which the devout thought of the early Christians discovered
between the Old and the New Scriptures, were thus invested
with a peculiar importance. Coincidences in language were,
as we know from early Christian literature, reverently and
eagerly noted.
Luke xvi, 31. OYAe AN TIC CK NCKpcON
ANACTH KAI AneAGH npOC AYTOYC
niCT6YC0YCiN.
The true text is : ovB' idv ti<; ix veK. dvaarfj TreiaOtjaovrai.
The interpolated words are clearly a context-supplement,
and come from v. 30 TTopevOrj Trp6<; avrov'?. But the variation
in the verb (aTreXdy, rropevdrj) implies the intervention of a
version. When we turn to the Sinaitic Syriac we read
(vv. 30, 31), 'If one from the-dead go (AtrSli) to-them,
repenting (are they). He-said to-him If Moses and-the-
prophets they-hear not, not-even if one frotn the-dead go
(AtndJ) (will they be) believing him.' Thus -Tropevdr} (true
text, V. 30)= A\r«:j = aTTeXOr) (D, V. ^\).
The reading appears in various authorities in different
forms : 225 245 TTopevOfj ; a ffM 1 ad illos ierit (-ittt i 1), b c q
ad illos abierit (-int b), e abierit a mortuis, d r surrexerit et ierit
ad eos; Iren. iv. ii. 3 (lat. int.) a mortuis resurgens ad illos
eat, credent ei ; Dial, contr. Marc' TTopev0y.
Luke xviii. 14. katcBh oytoc AeAiKAicoMeNOc ^
MAAAON nAp AlKeiNON TON (tlAplCAION.
The true text is: Kare^r) oi/ro<; BeB. eh rbv oIkov avrov
Trap iKeivov.
1 See Diet. Chr. Biog. (Adamantius), Prof. Robinson, Philocalia, p. xlviff.

SELECT PASSAGES FROM ST LUKE. 53
The Old Syriac (Sin. Cur.) and the Peshitta have : ' There-
went-down this-man to-his-house (Cur., to-his-house this-man;
Pesh., this-man justified to-his-house) justified rather than (lit.
from) that-man (Pesh., that Pharisee).'
There are three points : (i) The omission of ' to his house '
would be easy in Syriac, for it would be the omission of
a single word. The fact that the word has a different
position in the sentence in each of the three texts (Sin. Cur,
Pe.sh.) is an indication how easily it would fall out altogether.
As a matter of fact it has no place in Tatian as quoted
by Ephrem — 'This man went down justified more than
(he)' (Hill, p. 362). (2) Baethgen, Evangelienfrag., p. 18,
gives a long list of additions in the Old Syriac similar
to 'that Pharisee' here. (3) The Syriac rendering
of the idiomatic rrapd of comparison is the natural, indeed
the necessary, one. It is retranslated in the Bezan Greek
through the added fj,dXXov (= T»Av»). It is instructive to
notice that a literal rendering of the Syriac ^20 (from) has
passed into some Latin texts — magis ab illo, E ; magis ille
pharisaeus ab illo, gat; ab illo fariseo, T\ The sequel is
curious. This ab illo, a Syrism transplanted into t'he Latin,
was unintelligible. Hence, though it was retained, it was put
to a fresh use — ab illo magis quam ille fariseus, Q ; descendit
hie iustificatus in domum suam ab illo, vg. Thus Bede
ad loc, 'iniustus ad templum uenit, iustificatus a templo
rediit.' Or perhaps ab illo (taken with iustificatus) was re
ferred to God (v. 13).
Luke xix. 4. kai npoAABcoN
^ 6MnpOCG6N ANCBh em CYKOMCOpeAN
INA IAH AYTON on eKeiNH HMeAACN
AiepxecGAi KAI CYCNero cn tco
AiepxecGAi AYTON eiAeN kai ^ eincN aytco
ZAKXAie cneYCON kataBhGi.
The true text is : Kai TrpoBpa/jLonv eh to e/iTTpoaOev dve^rj
1 Similarly in Matt. xii. 6 the Bezan Latin (quia a templo maior est hie)
seems to preserve a Syriacised Old Latin reading.

54 THE SYRO-LATIN TEXT OF THE GOSPELS.
eTTt a. iva...iKeivr}';...Kal a)? ffxOev e'Trt toi' tottov, ai'a/SXei/ra?
[6] '\r)aov<;...e'iTTev tt/so? avrov ZaKXale, aTTevaa<; Kard^rjOi.
The Curetonian is: 'And -he -ran, anticipated - Him
(cQio.-ui), and-climbed-up into-a-fig-tree a-tasteless-one that-
he-might-see-Him, because-so passing was Jesus ; and-when
He-passed (even) Jesus, He-saw-him (.*oa.v»»). He-said,
Hasten, come-down, Zacai '.' The points in the passage
are these: (i) TrpoXa^wv. If the reading stood alone,
we should regard it as an itacism arising from TrpoBpa/jimv.
But it will be noticed that the Curetonian, as so often,
represents the tt^o- of the compound word TTpoBpaficov by the
addition of the verb 'he anticipated.' It would seem then
that the Bezan scribe, following the Syriac, retranslated this, the
most emphatic word in the sentenced (2) The Curetonian
and the Bezan texts agree in substituting for oj? ^X6ev e'Trt
toi' tottov the phrase (drawn from the context) 'And when
He passed,' the Bezan text being a little fuller. (3) elBev.
The (a) word and the (b) mood must alike be noticed.
(a) Syriac has no compound verbs. In representing the
compound verbs of the Greek it either has recourse to a
periphrasis or contents itself with an inadequate rendering
by a roughly equivalent (simple) verb. In the present case,
as elsewhere (see e.g. Matt. xi. 5, Mc. x. 51 f, Jn. ix. 18),
it used the common verb p^v*> (to-see) to represent dva/SXeyjrai.
Hence the Bezan translation elBev. (b) The Syriac regularly
resolves the Greek aorist participle into an indicative followed
by ' and ' (see below, p. 115): hence the Bezan elBev koI emev.
(4) o-Treuo-oi' Kard^r)6i. Here again the Syriac is unable
to represent exactly the participle aTrevaai (see below, p. 116).
It therefore, as so often, uses two imperatives asyndeta.
Hence the Bezan retranslation aTrevaov Kard^ydi.
1 The Sinaitic is only partially legible at this point. It has: 'And -he-ran
before-Him and-climbed-up into-a-fig-tree a-tasteless-one. ..because.... He was...
He-said to-him Hasten, come-down, Zacai.'
^ So e praecessit. Below (^7^ceTo...ouT6i') there agree with D the following:
157 a b c e ff^ i 1 q r s. In the following clause a large number of MSS. (Gr. Lat.)
have a conflate reading of some form, e.g. ijy elSev airdv dfajSX^^as Si k.t.X.

SELECT PASSAGES FROM ST LUKE. 55
The significance of these coincidences between the Bezan
and the Syriac texts lies in their combination,
Luke XX. 34.
01 Y'OI TOY AlCONOC TOYTOY feNNCONTAI
KAI reNNCOCIN TAMOYCIN KAI fAMOYNTAI.
The Old Syriac (Sin. Cur.) has : ' The-sons of this age
(are) bearing and-begetting (^.1X0,930 ^SJu), and(-are)-
taking wives and-becoming (^ooso) wives to-men.' The
word ^."U*, which I have ventured to translate (are) bearing,
seems to be commonly emended into ^tA>, and taken in the
sense of (are) begotten^. But the discovery of the Sinaitic
MS. furnishes what seems to be a conclusive reason against
this interpretation; for it is almost impossible to suppose that
the Sinaitic and the Curetonian should have independently
preserved the same itacism, (^."l" for ^."ul"). The interpre
tation which I propose introduces no new difficulty. For the
awkwardness of the phrase ' The sons of this age are bearing '
is parallel to the awkwardness of the phrase ' The sons of this
age... are becoming wives to men.' On the other hand the in
terpolated clause thus becomes strictly parallel to the clause
which follows it, both clauses speaking of the respective
parts which men and women play in this world. Such we
may, I think, say with certainty was the original meaning,
and such the original form, of the gloss. There are two
stages in its later history, (i) When it was transplanted
into a Greek text, where it was followed by a clause with an
active and a passive verb (ya/iovaiv Kai yap-iaKovrai), it was
natural to conform it to that clause and to render ^.iL> as
though it were ^.tA> (begotten) : hence the Bezan yevvwvrai
Kai yevvwaiv. The gloss is found in this form in some Latin
MSS. viz. ff" i q gat* E Q (generantur etgenerant% r (nascuntur
' So e.g. Cureton and Baethgen (p. 82) adopting the Bezan Greek (yevvuvrai
Kai yevvSunv). So too Mrs Lewis in her translation of the Sinaitic Syriac.
^ In E the interpolation stands after the clause : nubunt et traduntur ad
nuptias. Cyprian Aug. c e ff * i 1 q gat omit the clause they marry &c. altogether.
The Bezan Latin is : pariuntur et pariunt, nubunt et nubuntur.

56 THE SYRO-LATIN TEXT OF THE GOSPELS.
et generant). (2) The gloss was next further conformed
to the following clause, in which the active verb comes first.
In this form it is found in a eel Cypr. Test. iii. 32, de Hab.
Uirg., 22 (generant et generantur), in Clem., Strom., iii. 12,
Origen, Augustine.
Luke xxii. 12.
cKeiNoc YweiN Aeilei anapaion oikon
ecrpcoMeNON.
In place of oXkov the true text has fieya. The explanation
of this strange substitution is, I believe, simple if we look for
its origin in a Syriac text. The Syriac versions (Sin, Cur.
Pesh.) have .which(-is)-furnished large an (one) upper-room
In some Syriac text in place of «<'Av3i (large) the word
K'ixia.i (of-a-house) — 'an upper-room of-a-house which(-is)-
furnished' — was written, or was read by a copyist. The
confusion between .1 and i is too common to need illustration ;
in the present passage the substitution of .1 for \ would
be especially obvious, as the following word begins with .1.
The emendation, whether intentional or not, makes excellent
sense. The ' Bezan scribe ' however, when he reproduced this
Syriac reading, or perhaps his own misreading of the Syriac
word (large), in Greek, instead of oikoy wrote oikon, assimi
lating the termination to that of the previous word anafaion ^.
The Bezan Latin scribe took the word dvdyaiov as an
adjective meaning ' upper ' : hence his superiorem domum
(sup. locum, q).
In Mark xiv. 1 5 D has :
ANAfAION oikon eCTpO3M6N0N
MCfAN CTOIMON.
The true text is dvdyaiov fieya earpcofievov eroifiov. The
1 For such an assimilation in the Bezan text comp. e.g. Matt. v. 12 toyc
rrpoctJHTAC TOYC rrpo Y'V\wn YrrApxoNTCON.

SELECT PASSAGES FROM ST LUKE. 57
Bezan Latin is : stratum paratum grande. It would seem
that the Bezan (Greek) scribe repeated in Mc. the reading
which had been introduced into the parallel passage in Lc,
later in the sentence adding fieriav from the true text (iikya).
The word ' large ' has the same position in ff"- i q Or"'- as it
has in D. It is omitted in several cursives (see Tisch. in loc),
among these being 131 (see above, p, 18 n.) and 346 (see
above, p. 4 n.).
Luke xxiii. 36 f,
eNenezoN Ae aytco kai 01 CTpATiwTAi
TipocepxoMeNoi o2oc re npoce(t>epoN ^
^ AeroNTec" X'^ipg 0 BAC1A6YC TCON ioyAaioon
nepiTcGeNTec aytco kai akanGinon
crecfiANON.
The true text is iveTTai^av...6^o'; TTpoa^epovTe<; avrw Kai
Xeyovre'} Et ai) el 6 y8acrt\ei)? rwv 'lovBaiwv, awaov aeavrov.
The Old Syriac (Sin. Cur.) has : ' And-mocking were-they
at-Him also the-soldiers (Cur,, at-Him were they; and-also
the-soldiers) and-approaching (Cur., approaching) were-they
to-Him (Sin. cn^oA, Cur, crA) and-saying Peace to-Thee ; if
Thou art the-King of-the-Jews, save Thyself And-they-set
also (om. Cur.) on-His-head a-crown of-thorns.'
The Old Latin c has the same remarkable addition which
is found in D and the Old Syriac : ' aue rex iudaeorum,
libera te ; imposuerunt autem illi et spineam coronam,' In
a shorter form it stands at the beginning of the verse in a
Milan MS. (=M) (see Bp J. Wordsworth in loc): 'et in-
posuerunt in capud eius spineam coronam.'
There is nothing, so far as I know, in any of the different
forms of the gloss to indicate in what language it first
arose. But it is important to note that it is clearly due to
assimilation. Three points in Lc. xxiii. 36 f, viz. (i) the
mention of the soldiers ; (2) the mention of mockery ; (3) the
phrase 'king of the Jews,' link this passage with Matt, xxvii.
27 ff., John xix. I f , both which latter passages speak of the

58 THE SYRO-LATIN TEXT OF THE GOSPELS.
soldiers placing on the Lord's head the crown of thorns and
of their derisive salutation ' Hail king of the Jews.' The
addition at this point (Lc. xxiii. 36) then is natural.
We find a similar addition in one form of the Gesta Pilati
(Tischendorf, Evangelia Apoc, p. 231): ore dTTrfXQav irrl rbv
TOTTOV, i^eSvaav avrbv rd ifidria avrov Kai TTepie^waav avrov
Xevriov, Kai aricfiavov i^ dKavOwv rrepieO'qKav avrw TTepl rr)v
Ke<pdX'r]v. Kai iaravpmaav avrov. Here, it will be observed,
the addition is inserted at an earlier stage of the history.
If then its insertion where it occurs in the Old Syriac D c M
was natural, it seems likely that the interpolation was originally
made in the Syro-Latin text, and was thence taken by the
author of the Gesta Pilati and placed in a different setting '.
Luke xxiii. 40 ff. AnoKpiGeic Ae o erepoc
enereiMA aytco AercoN on oy <t)oBH cy
TON GN on CN TOO AYTCO KpiMATI CI
kai HMeic ecMeN kai hmbic mcn
AlKAICOC a5|A r^P WN enpAlAMCN
AnOAAMBANOMCN OYTOC Ae OYAeN
noNHpoN enpAlcN kai CTpAct)eic
npoc TON KN eineN aytco ^ mnhcGhti moy
eN th HMepA THC eAeYcecoc coy
AnoKpiGeic Ae 0 ihc eineN aytco too errAncoNn
GApcei ^ CHMepoN m6t eMOY £ch
eN TCO nApAAeicco.
The true text has : drroKpideh Be 6 erepo<; imnfiwv avrm
e(f)7] OvBe (j)o^fj av rbv Oeov, on iv rw avra Kpifxan el; koi
rjiMeh /ttev BiKaiwi, d^ta ydp wv iTrpd^afiev aTToXafji^dvofiev '
OVT09 Be ovBev droTTOv 'errpa^ev. Kai eXeyev 'I'qaov, /jLvijaOrjri
fiov orav eXOrj'i eh rrjv paaiXeiav (v. I. iv rfj /3.) crov. Kai emev
avrd) 'A/j.Tjv aoi Xe'^a), aijfiepov fier ifiov earj ev rai TTapaBeiaw.
The Old Syriac (Sin. Cur.) is as follows (v. 39 ff ) : ' One of
1 Yet the context in the Gesta Pilati has a point of contact with Matt, xxvii,
28 (Kai iKSijavres airiv xXajUi^Sa kokkIvtjv vepi.iBr)Kav airip).

SELECT PASSAGES FROM ST LUKE. 59
those however doers-of evil-things (or evil) who-crucified were
(with-Him, Cur.) blaspheming was at-Him, and-he-said to-Him
(om. Cur.) : Art not Thou the-Christ } save Thyself and-also
us (and-us also us, Cur.). And-there-rebuked him his-fellow
the-(that. Cur.) other, and-he-said to-him : Not-even of God
afraid-art-thou (afraid thou. Cur.), because-lo also we in-it we
in-the-judgment [i.e. we also are in the same judgment] 1 Aud
io we as deserving are-we (Sin. ^ocn. Cur. »^_oco ^ocn),
and-as we-did we-are-requited. But this-man not-even any
thing that-hateful (is) (is) done by-Him. And-he-said to-Jesus:
My-Lord remember-me when Thou-comest (coming (art)
Thou, Cur.) in-Thy-kingdom. There-said to-him Jesus: Verily
I-say to-thee that-to-day with-me thou-shalt-be in-Paradise (I
say to-thee to-day that-with-me thou-shalt-be in-the-garden-of
Eden, Cur.).'
The chief points in the passage are as follows: (i) koi
Tj/jueh iafiev^. When we turn to the Old Syriac we see that
the words ' because lo also we (are) in the same judgment
(re'ju.vs (iM COS ^1m Ai^* K'cn.i)' are due to assimilation
to the context as given in the Syriac, where in the previous
verse (awaov aeavrbv Kai fifids!) the Sinaitic has : ' Save Thy
self and also us (^A .ar**©),' the Curetonian the more em
phatic words ' Save Thyself and us also us (^A Ar^" ^r^o).'
Thus in the Old Syriac the ' and also us ' (Sin.), ' and us also
us ' (Cur.) of the one robber suggests the ' lo, also we ' of the
other. The Bezan scribe copied the et of the true text and
thus confused his assimilation to the Syriac. It should be
added that the simple iaixev in place of et is found in C* me
theb aeth, Gesta Pilati x. (Cod, A), Chrys. vii. 287 A, xi. 249 D,
760 C. (2) ovBkv TTovnjpbv eTTpa^ev. Chrysostom (ii. 480)
has ovBev Trovr/pbv iTToitjaev. One MS., viz. C (Tisch, p. Ixxi),
of the Gesta Pilati and Cyril, Cat., xiii. 3 have ovBev KaKov
eTToiTjaev. Another form of the Gesta (Tisch, p, 286) has
ovToi; Be TrdvTwi; ovBev KaKov eiTpa^e. Compare the Gospel
' If we considered this reading by itself, it might be plausibly suggested that it
arose from the following words Kai ii/ieU /iiv having been dittographed.

6o THE SYRO-LATIN TEXT OF THE GOSPELS.
of Peter iv. ij/tei? Bid rd KaKa a iTTOirjaafiev ovrw ireTTovdafiev .
It is easy to see how the word pc'^vx-lti (evil), which might
be translated by KaKov or vovrjpov, would naturally arise as
a reading or a gloss in Syriac. eirpa^ev is rendered by
ciA I--"- (done by-Him). This at once recalls the
phrase used to render KaKovpyoi (v. 39), viz. r^qir.i"^ •j.ias.
(doers-of evil-things, or evil). If the word ' evil ' were intro
duced into the speech of the penitent robber, this point would
be added to his words — ' We have been doers of evil things :
not any thing evil has been done by Him. He is not to be
ranked among us malefactors, us doers of evil things.' We
may compare the interpolation found in the Sinaitic Syriac at
Matt, xxvii. 16: 'A prisoner...whose name was Jesus Bar Abba.
He had been thrown into prison because of the evil things
which he had done (r^oeo .in'h .1 K'AuLn), and because he was
a murderer'; so in Lc. xxiii. 19 'because of evil things and
murder.' Thus in Syriac evil (Lc. xxiii. 41) would be due to
context-assimilation. (3) Koi arpa<f>eh tt/do? rbv Kvpiov eiirev.
This reading is, so far as I know, found only in two other
authorities. One form of the Gesta Pilati (Tisch. p. 286)
has Kai aTpa(j)eh tt/so? roi' 'Irjaovv Xe'^et avrw ILvpie, brav
^aaiXevay;, fir/ fiov iiTiXdOov. 0 Be eiirev avrS 'Er/fiepov Xe'^o)
aoi dXtjdetav iva ae e^^w et? rbv Trapdheiaov fier' ifiov. Again,
in the Armenian translation of the Acts of St Polyeuctes
(Conybeare, p. 138) we read as follows: 'Bethink thee of the
thief who was crucified on the right side ; what did he say to
the thief who was crucified on the left, and who reviled the
Lord .' " We suffer justly for what we have done, but our
Saviour'' was guiltless and sinless of the cross," and as he said
this he turned and said " Remember me, Lord, in Thy
kingdom'."... He said "This day art thou with me in Paradise."'
^ For this o8tu TreTrhvOaixev compare Gesta Pilati (Tisch. p. 286) rijieZi fifia
liv iirp6.iaiJ.ev i-rr6.Bop.ev, and the Old Latin b : 'et nos quidem iuste hsiec patimur' ;
see also the Armenian Acts of Polyeuctes quoted below.
" Comp. the Gospel of Peter oCtos Si a-wriip yevd/ievos t&v avBpibirav tI ^Sixijo-ec
iifias; ' This seems to have been Tatian's reading — ' Lord, remember me in Thy

SELECT PASSAGES FROM ST LUKE, 6 1
The addition seems due to the vivid fancy which would fain
picture all the details of a scene, which appealed profoundly
to Christian feeling. (4) fivqaOrirl fiov iv rfj '^fiepa rrj^
eXevaew<! aov. It should be noticed (see below, p. 94 f.)
that the Bezan scribe has the word eXevai<; in Lc. xxi. 7 in
a phrase which seems to come from Matt. xxiv. 3 through
the medium of the Syriac. The reading before us, whether
it arose in Greek or in Syriac, seems due to context-
assimilation. The robber is made to ask our Lord to
remember him 'in the day of His coming! The answer is
' To-day shalt thou be with me'.' (5) avrw rw iTTiTTXrjaaovri.
The addition seems to be a context-supplement and to be
derived from the eTTirifiwv of v. 40. But the variation of the
word, as indeed the phraseology (avrw rw iTTiTrX), implies
the medium of a version I I would suggest therefore that
the Bezan scribe is here following a Syriac reading or a
Syriac gloss: was who-rebuking to-him Jesus to-him there-said
kingdom' (Hill, p. 375). So Gesta Pilati (Tisch. p. 233), livrjcB-iyrl p.ov, KiJ/iie, iv
T^ ^affiXetg.' ffov.
' This is the connexion of to-day in Aphraat (p. Aflo^), as in the Sinaitic
Syriac, 'Verily I say unto-thee that-to-day with-me &c.' In another place (p.
\\ i<) Aph. omits to-day — ' And to one of them that were crucified with Him,
who believed in Him, He swore that " with-Me shalt-thou-be in the garden of
Eden." ' The omission is probably due to the fact that to-day is not required in
connexion with the purpose of the quotation. The Curetonian on the other hand
has a different connexion : ' Verily I say to-thee to-day that-with-me &c.' This
reading seems implied in Ephrem's Commentary on Tatian (Hill, p. 375).
Compare also Gesta Pilati (Tisch. p. 286), quoted above. This arrangement of the
words was perhaps due to an early misunderstanding of the word ' Paradise ' (or
' Garden of Eden '), as though it meant the final state of glory, whereas the Lord
' descended into Hades.' In later times we know that such a misunderstanding did
suggest this connexion of the words (see Archbp Trench, Studies in the Gospels,
p. 306 f.). Other difficulties were felt as to the words in early times; see the
passages from Origen and Chrysostom quoted by Tischendorf in loc. Marcion
according to Epiphanius omitted (apparently) the whole verse (see Dr Hort, Notes
on Select Readings, p. 68 f.).
2 The supposition that this version was Latin is excluded by the fact that the
Bezan Latin has alius increpabat eum (v. 40), qui obiurgabat eH (v. 43).

62 THE SYRO-LATIN TEXT OF THE GOSPELS.
The word r^r^lA is used in v. 40. The word eTriTTXrfaaeiv
is a word which the Bezan scribe uses in another passage,
where the phraseology makes it probable that he is re
translating. In Matt. xii. 16 (Kai iTrerifirjaev avroh) D reads
HANTAc Ae OYc eGepAncYceN enenAnSeN aytoic. The Curetonian
(Sin. wanting) and the Peshitta here render iTTeri/irjaev by
T^KlA. (6) Odpaei. The addition is obviously due to
assimilation to other words of absolution and promise ; see
Matt. ix. 2, 22'. No other authority for this reading is given
by Tischendorf It is found however in Cyril, Cat., xiii. 3 1 Sto
Kai BiKaiwi rjKovae %dpaei' ovx °''''' "^^ TTpdyfiard aov rov
dapaeiv d^ia, dXX' on ^aaiXev<; Trdpeari %a/3tfo^ej/o?.... ov
Xeyco aoi 'S.rniepov dTTepxi)' dXXd 'S,i]fiepov /ier' ifiov 'iarj. 6dp-
arjaov ov iK^7]0rjar}. Touttee in his note refers to the epistola
de uera circumcisione printed in the Appendix to Jerome's
works (v. p. 164): ' Forti animo esto : Amen dico tibi, hodie
mecum &c.' He adds ' Idem quoque uerbum ab aliis auc-
toribus citatum legi.' The only other reference however
which he gives is to Codex Bezae.
To sum up: the examination of the readings in this
passage seems to shew (i) that pious fancy was especially
active in regard to the story of the penitent robber, and was
not without effect on the Biblical text ; (ii) that the principle
of assimilation accounts for some of the Bezan readings ;
(iii) that there are signs that Syriac influence had at least
some share in the genesis of the Bezan text at this point.
Luke xxiii. 53. kai eGHKeN ayton cn mnhmciw
AeAATOMHMeNCO oy OYK HN OYHCO
OYAeic KeiMeNoc kai GeNTOC aytoy eneGHKe
TCO MNHMeiCO AeiGoN on Moric elKOCI

ckyAion.
' For a somewhat similar assimilation of words spoken by our Lord on the
cross to words spoken during His ministry, compare Tatian's version of our Lord's
commendation of the Virgin to St John (Eph. ; Hill, p. 375) : ' Thou young man,
behold, thy mother.' See Lc. vii. 14 f. ' And He said, Young man, I say unto
thee, Arise... And He gave him to his mother.'

SELECT PASSAGES FROM ST LUKE. 63
The true text is Kai edrjKev avrbv iv fiv^/iari Xa^evra ov
OVK Tjv ovBeh ovTTO) Keifievo^.
In regard to this remarkable passage there are two
questions for discussion. They are these: (i) To what
source or sources can we trace this strange gloss .' (2) Are
there in the passage and in the immediate context any signs
of retranslation } These two questions, it should be added,
are quite distinct.
(i) Whence did this perplexing gloss come? (a) First
of all it is plain that the account of St Luke is supplemented
by the addition of words from the parallel passage in
St Matthew (xxvii. 60) — Kai edrjKev avrb [rb o-oi/xa] iv rw
Kaivw avrov fivrj/jieia) b iXarofirjaev iv rfj Trerpa, Kai rrpoaKvXiaa's
Xidov fieyav rfj Ovpa rov fj,vrjp,eiov dTTrjXdev. Compare Mc.
XV. 46 Kai TTpoaeKvXiaev Xidov irrl rrjv dvpav to?) /ivrjfieiov.
Certain MSS., viz. U, the Ferrar-group and, according to
Tischendorf, fifteen others, together with the Memphitic
Version and two MSS. of the Aethiopic, insert words clearly
derived from Matt, (cf Mc.) — Kai TrpoaeKvXiaev Xidov fieyav
e'TTt rrfv dvpav rov fivrj/xeiov. If Ciasca's Arabic truly re
presents Tatian, his history of the burial was as follows (Hill,
p. 251 f) : Jn. xix, 38-42 (,,, There then, because the sabbath
had entered in, and because the tomb was nigh at hand, they
left Jesus) ; Matt, xxvii. 60 b (and they rolled a great stone
and thrust it to the door of the tomb, and departed) ;
Mc. XV. 47 a (And Mary Magdalene and Mary named after
Joses came after them unto the tomb) ; Matt, xxvii. 61 b
(and sat down over against the tomb); Lc. xxiii. 55b (and
saw how they brought in and placed the body there). It
appears then that in the Diatessaron, just before the mention
of the women, the phrase about the stone derived from Matt.
(cf. Mc.) had a place. Thus the interpolation is one of the
many instances in which, as it appears, a Tatianic reading
influenced the Bezan text. (b) But in the Bezan text the
jjueyav of Matt, gives place to the description — bv /i6yi<: e'Uoai
iKvXiov. The phrase is, I believe, derived from Joseph., de
Bello Jud., vi. 5. 3 (ed. Niese, vol. vi. p. 551), or, it may be, from

64 THE SYRO-LATIN TEXT OF THE GOSPELS.
a traditional account of what Josephus there records. The
passage is part of the description of the portents which, as
Josephus tells us, took place shortly before the destruction of
the Holy City. It runs as follows : ' During the same feast
[i.e. 'the feast of unleavened bread ']...the eastern gate of the
inner sanctuary (rov ivBorepw vaoii), which was of brass and
very solid (an^apwrdrrj), which in the evening was with
difficulty shut by twenty men (KXeiofiivrj Be irepl BeiXr/v fi6Xi<;
VTT dvdpwTTav e'lKoai), and which was supported by iron-bound
bars and had posts reaching far down, let into the floor of
solid stone, was seen about the sixth hour of the night to
have been opened of its own accord (avrofidrw; r]voiyfievrj).
The guards of the Temple ran and told the ofificer (rw
arparrjya) ; and he went up and was with difficulty able to
shut it (fi6Xi<; avrtjv 'iaxvae KXeiaai). This also seemed to
the ignorant a portent of most happy meaning ; for they
fancied that God had opened to them the door of His bles
sings. But the learned were of opinion that the security of the
sanctuary was of " its own accord " being broken up, and that
— a free gift to the foe — the gate was being opened, and among
themselves they explained the sign as indicative of desolation.'
That this story made a deep impression and was widely
known appears from the fact that it is referred to by the
Roman historian (Tac, Hist., v. 13): ' Euenerant prodigia....
Uisae per caelum concurrere acies, rutilantia arma, et subito
nubium igne collucere templum. Expassae repente delubri
fores, et audita maior humana uox, excedere deos.' Thus
there is nothing violent in the supposition that this story was
well known in the birthplace of the Bezan text, especially if, as
I believe, there are strong reasons for thinking that that birth
place was the Syrian Antioch. Moreover, the desolation
of the Holy City after the revolt in Hadrian's reign would
revive the memory of, and give special point to, the stories
current as to the siege of Jerusalem under Vespasian. It
would. appear that at this time, that is, in the second quarter
of the second century, the ' Syro-Latin ' text of the Gospels
was taking shape.

SELECT PASSAGES FROM ST LUKE. 65
But what are the links which connect the story of the
Temple gate in Josephus with the description of the tomb in
St Luke .'' What would carry the mind of a reader, or
transcriber, of St Luke's words, to the portent in the Temple
on the eve of the destruction of Jerusalem ? The points
of connexion are these: (i) In both cases the size and weight
of the barrier are emphasised (comp. Mc. xvi. 3 f ). In both
cases it is miraculously removed without human intervention'.
(2) The gate in Josephus closed the way into the sanctuary
(va6<;). The stone in the Gospel lay at the mouth of the
tomb where there lay 'the sanctuary of His body' (0 vab<;
TOV awfj,aTo<; avrov, Jn. ii. 2i). (3) The miraculous opening of
the Temple gate was an omen of the destruction of Jerusalem.
The murder of our Lord ensured the doom of Israel. Besides
these, other minor coincidences may be noted : (a) Both
events took place at the time of the Passover, (b) The gate
was shut at evening. The stone was placed at the door of
the sepulchre at evening, (c) The floor was of solid stone.
The tomb was hewn out of the rock, (d) When the gate
was miraculously opened, the temple guards (ot rov lepov
<pvXaKe<;) ran and told the officer (tw arparTjyw). When the
soldiers at the tomb saw the wonders of the Easter morning,
some of them went into the city and told the chief priests
what had happened.
The coincidences then are striking. Such coincidences
might well appeal to the imagination of second century
Christians, and their sense of them find expression in the
substitution of the description of the Temple gate for the
Evangelist's simple epithet fieyav.
The gloss is found in (a) the Old Latin c : ' et cum
positus esset in monumento, posuerunt lapidem quem uix
uiginti uoluebant.' The Bezan Latin is : ' et posito eo in-
posuit in monumento lapidem quem uix uiginti mouebant.'
(j8) the Thebaic Version : ' When he placed Him however, he
1 With the aiTO/idTois ¦qvoiy/iivri of Josephus compare the phrase in the
Gospel of Peter (ix.) : d0' ^auToD KvXiaBeU iirex'i'mae irapi, //.ipos.
c. 5

66 THE SYRO-LATIN TEXT OF THE GOSPELS.
placed a stone in the mouth of the tomb, which twenty men
would be able to roll.' It would seem that the two Latin
texts (c d) understood devro'^ as though it were redevro<; and,
though in different ways, gave a passive verb (cum positus
esset, posito eo). The difference of phraseology implies trans
lation from a common original. Thus all the phenomena
point to the Greek as prior to the Latin.
But if the Bezan Greek is prior to the Latin, is the Bezan
Greek itself the original form of the gloss .' There is of
course no a priori objection to this view. A careful exami
nation of the language of the most famous of the Bezan
glosses — the man working on the Sabbath (Lc. vi. 4 f ) — does
not reveal any indication of retranslation. The case might
be the same here. It is simply a matter of evidence. We
must examine (i) the immediate context, (ii) the gloss itself,
and see whether the language shews signs of retranslation.
(i) The context: (d) cn mnhmeico AeAATOMHMeNco (v. 53;
true text iv fivrjfian Xa^evrw). The Syriac (Cur. Pesh. ; Sin.
wanting) naturally renders Xa^evrw by the passive participle,
of which the Bezan XeXarofiTjijevm is an exact representation.
This coincidence however cannot be pressed into an argument,
since in Mc. xv. 46 we have iv fivqfian b ^v XeXarofiTj/jievov.
(b) In ^.55 we have the form apaAiAaiac ; see below, p. 102.
(c) kathkoAoyGhcan Ae Ayo' rYNAiKec... kai eGcACANTO (v. 55;
true text A;aTa/<;oXov^?^cra(7at...e^6acraz'To). The Bezan Greek
reproduces the Syriac rendering of the Greek participle:
' Those women who-went with-Him from Galilee wejit to the
sepulchre in-their-footsteps and-saw. . .' (Sin. Cur.); comp. p. 115,
(ii) The gloss itself: (a) The words Kai devro^ avrov
iTredr]Kev read to me like a somewhat halting piece of trans
lation. This however is a matter of impression, (b) In the
Greek Gospels the only word used to describe the placing the
stone at the sepulchre is TrpoaKvXiaai (Matt, xxvii. 60, Mc. xv.
46). The Old Syriac (Sin. ; Cur. wanting) however has in
' The Svo (also found in 29 a b e ff^ q r Q) is probably due to the mention of the
two Maries (see the Arabic Tatian quoted above, p. 63).

SELECT PASSAGES FROM ST LUKE. 6^
Matt, xxvii. 60 ' And-he-placed (lit. he-cast, jsoiri'o) a great
stone at (lit. upon, A:^) the-door of the sepulchre ' ; in
Mc. XV. 46 'He-rolled a-stone, he-placed (jSoirc') (it) at
(A:^) the-door of the sepulchre.' The Syriac words 'he-
placed... upon (at)' would precisely suggest the Bezan
iTTedr)Kev. (c) The word eKvXiov is to be noticed on two
grounds, (i) On the one hand we have here the simple verb.
In the Greek Gospels the compound forms of this verb
(rrpoaKvX., drroKvX., dvaKvX.) alone are used. The Syriac of
course has only an uncompounded verb (.\ \^s ). (ii) On the
other hand the imperfect tense is to be remarked. The
sense required is ' could roll ' rather than ' were rolling.'
Now this ' could roll ' would be concisely expressed by the
Syriac imperfect (see Noldeke, Syr. Gram., § 266). It seems
as if this idiomatic Syriac imperfect were literally translated
by the Greek imperfect.
Further, the story of the miraculously opened gate of the
temple, as told by Josephus, would to a reader of the Syriac
Gospels very naturally connect itself with the history of our
Lord's Passion. The phrase ro KaTarrkraajia rov vaov
iax''(^drj becomes in the Old Syriac ' there was rent the front
of the gate (r^l^iA\) of the sanctuary (Matt, Sin.), of the
temple (Mc, Sin. ; Lc, Sin. Cur.).' Josephus tells us that
the supernatural opening of the gate was regarded as pro
phetic of the destruction of the temple. Ephrem gives a
similar explanation of the rending 'of the veil' — 'in scisso
uelo imaginem templi diruendi proposuit, quia Spiritus eius
ex eo exierat' (Moesinger, p. 256).
When these indications of Syriac influence in the gloss
itself and in the context, in which it is embedded, are con
sidered together, there seems to be good, though not perhaps
conclusive, reason for thinking that it came into the Bezan
text from a Syriac source.
One question remains. Can the gloss, assuming that it
is derived from the story as to the Temple gate, have been
originally Syriac .' If the gloss is derived from oral tradition,
5—2

68 THE SYRO-LATIN TEXT OF THE GOSPELS.
there is no difficulty in giving an affirmative answer to this
question. But if the source of the gloss is literary, not
traditional, i.e. if it is derived from the narrative of Josephus,
is the supposition that it first found its way into a Syriac text
of St Luke excluded .' The story occurs in Josephus' treatise
The Jewish War. It appears from the Preface to that
treatise (comp. Contra Ap. i. 9) that Josephus first wrote
the history in his native Aramaic and circulated it in the East,
and then translated it into Greek for use in the Roman
Empire'. Thus there is nothing at all improbable in the
supposition that Syriac-speaking Christians knew the de Bello
Judaico of Josephus in its original Aramaic form.
Briefly to sum up : I have given reasons for thinking that
we have in this Bezan gloss, found also in one Latin MS., and
in one Egyptian version, a reference to a story connected
with the destruction of the Holy City. The Bezan gloss
itself has probably come from an Old Syriac text. If so,
it must remain uncertain whether the gloss was original in
this Syriac form, or whether it came into the Syriac from a
Greek text ; but there is nothing improbable in the former of
these two suppositions.
Luke xxiv. 32 f 01 Ae einoN npoc eAYTOYc
OYXI H KApAlA HN HMOJN KSKAAYMMCNH
coc eAAAei hmcin cn th oAco
(X)C HNYreN HMeiN tac rpACt)AC
KAI ANACTANTeC AyUOYMCNOI AYTH TH COpA
YnecTpeyAN.
The true text is /cat eirrav rrpb'i dXKriXov<i Oi5%t r) k. rjfiwv
Kaiofievr) rjv &>?... oSw, w? Birjvoiyev k.t.X.
The points to be considered are these : ( i ) KeKaXvfifievr].
The Syriac rendering of the true text (Kaiofievr)) is ^ja*,
which is the reading of the Peshitta. But this Syriac word
suggested either to the original Syriac translator or to an
^ 'EXXdSi yXiio'a-ri /lera^aXav a tois aviii /3ajO/3dpois rg Trarpiip a^vvrd^as a.viirep.yj/a
¦wporepov.

SELECT PASSAGES FROM ST LUKE. 69
early copyist an obvious and attractive emendation which
would assimilate this verse to v. 25 rtfLaA .^ixo* (=^paSeh
rfj KapB'ia). Hence both in the Sinaitic and in the Curetonian
we have 'Our-heart heavy (i*aj) was.' This obviously Syriac
reading appears in the Old Latin 1 (optusum^), the Thebaic
and the Armenian. This emendation in the Old Syriac text
seems to lie at the root of the other variations. It changed
the whole tone of the sentence : the predicate became one
expressive of disparagement. The remaining readings are
three in number, (a) The Old Latin e has exterminatum.
The word exterminare is very common in Old Latin BibUcal
texts and in early Latin Christian literature in the sense of to
destroy: see the instances quoted by Ronsch (ftala it. Vulgata,
p. 365 f., comp. pp. 56, 74_), and note especially the rendering
of Ps. xxii. 17, twice given by Tertullian (Adv. Jud. x. xiii.):
' Exterminauerunt (= wpv^av) manus meas et pedes.' I believe
that this exterminatum of e arose from a very obvious itacism
in some Syriac text which here lies behind e, or possibly from
a misreading of the Syriac word on the part of a bilingual
scribe. The common Old Syriac reading was, as we see in
the Sinaitic and the Curetonian, Tijv. (heavy). This word by
mistake was written or read as tloj (= hewn out, Matt, xxvii.
60, Mc. XV. 46, Lc. xxiii. 53), and this perplexing itacism is
somewhat skilfully represented by the exterminatum of e.
(b) The Old Latin c has excecatum. It would be very
natural, when once the adjective in the sentence under
discussion got a disparaging tone, to compare with this
passage, dealing with the Christian interpretation of the Old
Testament by Jews, the words of St Paul in which he treats
of precisely the same subject (2 Cor. iii. 13 ff) 'The children
of \sraA.. .their minds were hardened (iTTwpwdr] rd vot^fiara
avrwv) : for until this very day at the reading of the old
covenant the same veil remaineth unlifted.... Unto this day,
whensoever Moses is read, a veil lieth itpon their heart! Now
1 This optusuni may however come from 2 Cor. iii. 14 (optusi sunt sensus
eorum) : see below.

70 THE SYRO-LATIN TEXT OF THE GOSPELS,
in the Peshitta the phrase e-jrapwdr) rd v. avrwv is rendered
' But they-were-blinded (om^Axrc^ in-their-minds.' Hence
it appears likely that the Old Latin c excecatum is derived
from 2 Cor. iii. 14 through the medium of a Syriac text.
(c) Lastly there is the Bezan reading (KeKoXviMfievrj). It also
comes from 2 Cor. iii. 13 ff.', like the reading of c But in itself
1 'Probably from 2 Cor. iii. 14 f.' (Dr Hort, Notes on Select Readings,
p. 72). With these readings derived from 2 Cor. iii. compare the Bezan text of
Acts XV. 29 A^ CON AiATHpoYNTec eAYTOYC eY rrpA^ATe 4>epoMeNOi eN
TCP Apico riNI eppcocGe. Irenaeus (iii. 17, ed. Harvey) has: '=> quibus
custodientes uos ipsos, bene agetis, ambulantes in Spiritu Sancto.' Tert., de
Pudic. xii, gives the gloss in the form ' uectante uos Spiritu Sancto.' On this
interpolation I wrote (Old Syriac Element, p. 95) thus : ' I believe that the desire
to make the Apostolic decree more spiritual led to the introduction into the Old
Syriac text [which here lies behind the Bezan text] of a phrase from a Pauline
Epistle, which deals with the Judaistic controversy. See Gal. v. 18.' Mr Rendel
Harris (Four Lectures on the Western Text, p. 75 ff.) makes two criticisms on this
position. (i) He does not allow that in the Bezan text an attempt is made
to spiritualize the decree. In answer to this criticism I can only appeal to the
decree as a whole in the Bezan text, and to the amplification in that text of the
reference to the decree in xvi. 4 (true text irapedldofrav adroh (puXdaaeiv rd S6yp.aTa
rd KCKpip-iva iiirb tCiv diroaT. Kai irpea^.): BKHpYCCON KAI HApeAlAoCAN AYTOIC
MBTA TTACHC HAppHCIAC TON KN IHN XPN AMA TTApAAl AONTBC KAI TAC
bntoAac AnoCToAcoN KAI npecBYTepcON. The 'tendency' here cannot be
mistaken. On ras ivroXdt (which exactly represents the Syriac equivalent in
the N. T. of rd S6yp,aTa) see Old Syriac Element, p. 95 n. (ii) Mr Harris gives
his own theory thus (p. 77) : " The gloss does not belong where Mr Chase
imagines and where I first thought it to belong, but is a part of the following
sentence, describing the Apostolic Mission to Antioch. The current text of this
passage is ol /Mev odv diroXvdivTES KaTTJXdov els ' Avnoxei-av,
with which we must compare the parallel passage (xiii. 4),
oi [sic : lege avrol) /iiv ovv iKirep,(p6ivTes vvb tov dylov TrveipMTOs KaTTJXBov
els ^eXevKeiav.
Accordingly, the sentence in Acts xv. 30 should run, ' So they were led by the
Holy Spirit, and came down to Antioch.' " I note in passing that the real
difficulty of Mr Harris' theory lurks under the English phrase ' they were led.'
Later on (p. 79) Mr Harris notices that " the two passages are in harmony, as far
as the principal verb is concerned, in the Peshito." Thus, ' And-they vfhen they-
were-sent-forth (cL»jAA\X,r<') by (..'w) the-Spirit of-Holiness ' (xiii. 4); 'They
however who-were-sent-forth ' (xv. 30). " The same approximation of the account,"

SELECT PASSAGES FROM ST LUKE. 71
it offers no indication whether or no it is a reading originally
Greek. (2) XviTov/u-evoi (so c e theb.). This gloss is so
meaningless that it must have arisen from some transcriptional
accident. Can any clue to its genesis be found .' In this
verse the Kai dvdaravrev becomes in the Syriac Versions (Sin.
Cur. Pesh.) cutajao (and-they-rose-up). In v. 17 the true
text has Kai iarddrjaav aKvdpwwoi. The Syriac Versions
(Sin. Cur. Pesh.) have the common reading kuL iare aKvdpa-
TToi — 'while (Pesh. and-) sad (^iiSiA) (are)ye.' If however
we put the reading of the true Greek text into Syriac, it is
^TisoA ."»A otoao (and-they-stood-still while sad) ; for )tia
means ' to stand still ' as well as ' to rise up ' (see e.g. Lc. vii.
he continues, "appears in Cod. Bezae [i.e. the Bezan Latin] which reads in
xiii. 4 ipsi uero dismissi ab spo sancto, and in xv. 30 dlli quidem dismissi."
Whatever may be thought in general of Mr Harris' new theory of foundling
glosses, it is clear that in the present case (in whatever language the gloss may be
supposed to have arisen) all that he has shewn is that under the influence of xiii. 4
the gloss ' by the Holy Spirit ' might have been naturally added in xv. 30. His
theory accounts for the words ' by the Holy Spirit,' but not for the word (pep6p.evot,
ambulantes, since there is already in xv. 30 a participle (an indicative in Syriac).
It remains that I should very briefly explain my own position. The decree
(Acts XV. 23-29) deals with two chief points : (i) Were the Gentile converts
under the law? (ii) What were their duties? In regard to the latter point
it should be noticed that in the Bezan text, in which koX ttviktuv (v. 28, cf. v. 20)
is omitted, the words dirixeffSai elSaXoBiroiv Kai atfi,aTOS Kai iropvelas are most
naturally interpreted as enjoining abstinence from idolatry, murder, fornication^
three ' works of the flesh.' The whole passage — Gal. v. 13-25 — deals precisely
with these two subjects of the decree. I transcribe the chief phrases, italicising the
words which, as I believe, suggested the gloss in Acts xv. 29 in D and Irenaeus :
'(v. 13) For ye, brethren, were called for freedom ; only use not your freedom for
an occasion to the flesh.... (16) But I say, Walk by the Spirit (irvevp.aTL
irepnraTeiTe), and ye shall not fulfil the lust of the flesh.... (i8) But if ye are led
by the Spirit (Trvetj/ian dyeoBe), ye are not under the law. (19) Now the works
of the flesh are manifest, which are these, fornication. ..idolatry.. .enmities, strife,
etc  (22) But the finit of the Spirit is love.... (23) Against such there is no
law.' There are indications that the gloss comes through the medium of a
Syriac text, (i) The word (f)ep6p,evoi points to retranslation. The medium
cannot be the Bezan Latin ; for ferentes in d is obviously a meaningless translation
of <pep6iJ.evoi (taken as the middle voice). The Syriac |.*^s.10\93 ( = dyeaBe,
Gal. V. 18) would be very naturally translated by <j>ep6p,evoi.. (2) The preposition
(D ^i*, Iren. in) will be noticed. The Peshitta has ' ;'«-the-Spirit ' (r^lijois)
in Gal. v. 16, 18 (vveiiMri).

']2 THE SYRO-LATIN TEXT OF THE GOSPELS.
14). If we may assume, and it is no great assumption^ that
the true reading oi v. 17 appeared in some Syriac text, we
have a natural explanation of the Bezan reading in v. 33. In
this latter verse some Syriac scribe, with v. 17 in his mind,
after asa.no (and-they-rose-up) wrote the words ^T>.sa.& s^
(while sad) which followed the O-sajia (and-they-stood-still)
of V. 17: the oisxa<s, common to both verses, brought with it
into V. 11 the 'while sad' of v. if^. We may compare a
somewhat similar phenomenon in v. 13 (see below, p. 109).
Thus in a Syriac text, which here lies behind the Bezan text,
there arose in w. 33 the reading 'and-they-rose-up while sad
(^ioSaSk).' The Bezan translation of the last two words by
XvTTOvfievoi is very natural, Xvrreladai being translated by
the Ethpeel of i:2a& in Matt. xxvi. 37.
Luke xxiv. 37. aytoi Ae nroHGeNTec
KAI eMC|)0B0l reNOMeNOI eAOKOYN Ct)ANTACMA
GecopeiN.
(i) The Sinaitic and the Peshitta begin the sentence with
the word .___aicno (and-they). (2) What of (pdvraafia in
place of TTvevfia ? We compare at once the somewhat similar
passage Matt. xiv. 26 (comp. Mc. vi. 49) : ot Be fiadrjral IBovre'i
avrbv iirl ti}? daXdaar]<; TrepiTTarovvra irapdxdrjaav Xeyovre^
on ^dvraafid iariv, Kai dirb rov <f)6^ov eKpa^av. It is a
1 The Old Latin e (et steterunt tristes) alone among Latin MSS. has this
reading. If e had been lost, there would have been no Latin authority for this
reading. ^ In connexion with this verb I take the opportunity of noticing the Bezan
reading in Mc. vii. 9 ina thn nApAAociN ymcon cthchtai. The true text has
TTipri<rr)Te. The Bezan Latin reads : ut traditionem uestram tradatis. The Old
Syriac (Sin.) has : ' Forsaking (are) ye the-comraandments of God, that-ye-OToj/-
establish ( aS9.U3^,l) your-commandments ' (for the word commandments =
tradition cf. Matt. xv. 2ff.). Pesh. also has ' that-ye-may-establish.' It would
seem as if the translation arose through a. misreading of Ti\pi\<nirt. For such a
misreading in the Old Syriac (Sin.) compare ' that-they-might-hang Him ( = Kpe-
Udoai for Kp-qp-vlirai., Lc. iv. 29)'; see Baethgen, p. 8. The reading is also found
in 1-209 28 abcf ff^iqr (.f/atea/M) arm., Cyprian De Cath. Eccl. unitate 19, Epist.
xliii. 6.

SELECT PASSAGES FROM ST LUKE. 73
sufficient explanation of the Bezan reading to suppose that
the word (jjdvraafia is due to assimilation of this passage to
Matt. Mc. In connexion however with this reading it is im
possible not to take into consideration the remarkable passage
in Ignat. Smyr. 3 : e^»/ avroh Adhere, i{rr)Xa(l)ijaari fie, Kai
'iBere '6ri ovk el/il Bai/ibviov daco/iarov^. Jerome (de Vir.
III. 2) ascribes this saying to the Gospel according to the
Hebrews, the Aramaic original of which he himself translated
into Greek and Latin. Bp Lightfoot points out in regard to
the Ignatian passage that 'the reference is plainly to the
same incident which is related in Luke xxiv. 36 sq.; see esp.
vv. 38, 39.' What then of the strange word Baifioviov? It is
difficult to suppose that it would have been chosen for its
own sake as the word which our Lord used of Himself
When however we turn to the Sinaitic of Mc. vi. 49 (eBo^av
on (j)dvraafid iariv), we find the words ' they-thought that-«-
devil (r^:ir^j.^) (was)He'; and in Matt. xiv. 26 (where the
Sinaitic is illegible) the Curetonian has: 'And-saying were-
they that-a-devil (was)He.' The word p^.irslx., here used
to denote a spectral form, is a common equivalent of
Baifibviov (see Matt. vii. 22, xvii. 18; Mc. vi. 13, xvi. 9).
Hence we are led to ask whether the Baifioviov of Ignatius is
not the exact rendering of the Syriac t<'.ir<lx. ; whether in
fact this saying of our Lord's, preserved to us by Ignatius
of bilingual Antioch, is not derived from some Syriac account
of the Resurrection, identical with, or closely allied to, the
account of St Luke.
The evidence of Ignatius seems to shew that in some
Syriac gloss on, if not in some Syriac text of, Lc. xxiv. 37
the word devil (in the sense of apparition) was introduced from
Matt. xiv. 26, Mc. vi. 49. It is then at least possible that the
Bezan word (f)dvraafia may he a retranslation of a Syriac
^ Compare Ephrem's comment on Matt. xiv. 26 (Mc. vi. 49) : ' Cur ergo
mirati sunt? Si eum incorporeum nouerunt, imprudenter admirati sunt.... Si
uero corporalis erat, recte obstupuerunt.... Et quia ipse Dominus sciuit, quod
recte obstupuerunt, animum eorum confortauit dicens : Ego sum, nolite timere,
i.e. ego sum ille corporalis, quem uos cognoscitis' (Moesinger, p. 135).

74 THE SYRO-LATIN TEXT OF THE GOSPELS.
reading or gloss. There are indeed indications of Syriac
influence in the context, (i) D has octa oyk exei kai cApKAc
(?'. 39 ; true text, adpKa Kai barea ovk exei). The Syriac (Sin.
Cur. Pesh.) has the common order ('flesh and bones'). It is
however in Syriac, in whatever order the words stand, that
we see how easily the plural adpKa<; would arise. The Syriac
words (Sin. Cur. Pesh.) are these; K'jsai.^^ r^h ttt-t (flesh
and-bones). Even when the words stand in this order, we
see at once how readily the terminations would become
assimilated and r<'i w-i be pointed as a plural (compare The
Old Syriac Element (p. i8) on Acts ii. 17, eni uacac capkac D).
The reading is found in N* and appears to have made its
way into an Old Latin text : for in the Latin translation of
Irenaeus (v. 2) we read: '...Spiritus enim neque ossa neque
carnes (Gr. adpKa) habet; sed de ea dispositione, quae est
secundum uerum hominem quae ex carnibus (Gr. iKaapKo^;)...!
It occurs also in the Dialogue printed among Origen's works
(i., p. 857 (ed. Delarue); see above, p. 52), barea Kai adpKa<;
ovK exei. (2) Again, D has in v. 39 BAenere (true text dewpeire).
The Syriac (Sin. Cur. Pesh.) has the common word ^V4»
(seeing). The word dewpeiv is rendered by this verb in Syriac
e.g. in Matt, xxvii. 55, xxviii. i; Mc. iii. 11. Further, this
Syriac verb is the constant equivalent of /SXeTreti/ ; see e.g.
Matt. V. 28, vi. 4, vii. 3.
It is impossible to refrain from considering a remarkable
gloss in the immediate context (v. 43), though D does not
contain it. The true text is ot Be iTTeBwKav avrw ixdvo<;
otttov fiepo'i' Kai Xa^wv ivwmov avrwv e^ayev. The inter
polation in question, which is inserted after e<f>ayev, comes to
us in several forms: (i) The Ferrar-group, KII*, and
many cursives have kai ta eniAoinA eAcoKeN aytoic'. (2) 88 has
the same reading with ta nepicceYMATA in place of rd irriXonTa.
(3) 130 with TO enANAAei^eeN in the same position. (4) r has
' It will be noted that the interpolation is not found in Sin. Some MSS. of
the Memphitic have the following words : ' He ate and He took the remainder.
He gave to them.' Epiph. Haer. i. ii. xxx. ch. xix. has : Xo(3wc '4<l>aye /coi ISoiKev
Tols paBifTaXs.

SELECT PASSAGES FROM ST LUKE. 75
' reliqua accepit et dedit illis.' (5) c has a reading found also
in Aug. de Consejisu (iii. 74) and the Latin Vulgate : ' sumens
reliquias dedit eis.' (6) The Curetonian Syriac has :
K__ocnl j3cnL> i^.i yi'\Si Aaz-o
to-them He-gave which- (was)-Ieft that and- He-took
When we look at the variations in the form of the gloss
as found in Greek authorities (rd iTTiXoiTra, rd rrepiaaevfiara,
rb iTTavaXei(f)0ev), it is impossible not to feel that we are
dealing with retranslations of a gloss in some other language
than Greek, based on the expressions used in connexion with
the two miracles of feeding the Thousands. When, in order
to gain light on the question — -Through the medium of what
language did the gloss come .' — we compare the gloss with the
passages in the Gospels (Matt. xiv. 20, xv. ^? ; Mc. vi. 43,
viii. 8 ; Lc ix. 17), we are struck with the constancy in the
phraseology of the Syriac Versions: for (i) the word cvlax.
(= they-took-up), (2) some word from the root iiu, are always
used. Such constancy is not found in the Latin renderings
of the several passages — thus (a) sustulerunt, coUegerunt,
sublatum est, (b) reliquias, reliquum, quod superfuit, quod
superauit, quod abundauit — all these phrases are found in
their respective places. Thus, so far as the indications go, it
appears that the Syriac has a better title than the Latin to
be the source whence this gloss found its way into Greek and
other authorities.

HARMONISTIC INFLUENCE,
In this Chapter I shall call attention in the main to three
points in regard to the Bezan text of the Gospels. They are
these: (i) The Bezan text shews constant indications of
harmonistic influence. (2) In such harmonized passages
readings occur which we are justified by other evidence in
considering as Tatianic readings. (3) There are often
clear signs of the influence of Syriac phraseology in, or in
the neighbourhood of, readings due to harmonistic influence.
The discovery of the Sinaitic MS. of the Gospels reopens
the question of the relation of the Diatessaron to the different
forms of the Old Syriac text. The number and character of
harmonized readings in the Sinaitic text seem to point to
the priority of Tatian. If however the Sinaitic text is prior to
Tatian's work, it would seem that Tatian gave definite and
practical shape to tendencies already at work in moulding
Old Syriac texts^
^ Mr Burkitt in his valuable paper on the Sinai Palimpsest in the Guardian of
October 31, 1894, maintains the priority of the Old Syriac. On the other hand
Dr Zahn in the second of his articles in the Theol. Literaturblatt (Jan. 4, 11, 18,
1895) arrives at the verdict ' T[Tatian] ist und bleibt das iilteste nachweisbare
Evangelium der Syrischen Kirche. Man konnte hiernach berechtigt scheinen
die Genealogie : T-Ss[Sinaitic]-Sc[Curetonian]-P aufzustellen,'

HARMONISTIC INFLUENCE. "J"]
Matt, xxi. 1 8. npcoi Ae nApAfWN eic thn noAiN
For TTapdywv the true text has iiTavayaywv. The Sinaitic
Syriac is wanting here. The Curetonian has : ' In-the-morn-
ing however when passing was-He to-the-city.' The reading
seems due to assimilation to Mc. xi. 20 (/cat TrapaTTopevofievoi
TTpwl elBov). The difference of the words (Trapdywv, rrapa-
TTopevofievoi) points to the medium of a version. The word
used in the Curetonian is the same as is used in Mc. by the
Sinaitic ('and-when passing were-they in-the-morning') and
by the Peshitta (' and-in-the-morning when passing'). In
Matt, some Old Latin MSS. (e.g. e transiens ciuitatem) have
transiens. In Mc. however most Latin authorities have a
circumstantial clause, not a participle — et cum mane transirent
(q cum transiret, gat transfretasset ; k et praetereuntes illi qui
cum eo erant).
Matt. xxiv. 3 1 f.
AnO AKpCON OYpANCON
eCOC AKpCON AYTCON
APX0M6NCON Ae TOYTCON re'NECGAI
ANABAe^ATC KAI CUApATe
TAC Ke([)AAAC YM^'N" AioTi erpeizei
H AnoAYTpooceic ym^jon
AnO Ae THC CYKHC K.T.A.
With the single variation of dva^erifare for dvaKv^frare,
the interpolated words come from the parallel passage in
Lc xxi. 28. They stand in precisely the same position in
the Arabic Tatian (Hill, p, 211 f) as in D— ' ...from the end
of heaven even to the end thereof But when these things
begin to come to pass, be of good cheer, and lift up your heads ;
because your deliverance draweth nigh. From the fig-tree
learn the parable.'
The same interpolation is found in the Old Latin MSS,
b c hq.

78 THE .SYRO-LATIN TEXT OF THE GOSPELS.
Matt. xxvi. 59 ff. czhtoyn
VEYAOMApTYpeiAN ¦ KATA TOY IHY
oncoc AYTON Ganatcocoycin
KAI OYK eYpoN TO e5HC
KAI rroAAoi npocHAGoN " yeYAOMApTYpec
KAI OYK eYpoN TO elHC
YCTepoN Ae hAGon Ayo ycYAoMApTYpec
KAI einON' TOYTON HKOYCAMCN ACfONTA.
The important words from the true text of Matt, and
Mc. (xiv. 55 f) are as follows:
Matthew Mark
Kat ovx f^poi/ Koi ovx evpio-Kov '
TToWav TTpocreKdoPTcov yj/evBofiapTV- ttoXXoI yap e^evhop.apTvpovv Kar
pdiv. avTov,
Kai 'ifrai al pLapTvplai ovk rjaav.
varepov de npoaeXdovres bvo elnav kol rives avaa-rdvres e^j/evBofiaprvpovv
Kar avrov Xeyovres on
Ovros ecjjrj. 'Hfiels ^K0V<rafj,fv aijroC Xeyovros.
The point which is of special importance is the twice
repeated to e^rj<;. When we turn to the Old Latin MSS., we
find that (a) where the words to e^rj<; first occur, ff^ has exitum,
h exitum rei'^; (b) where the words recur, a has exitum rei,
ff^ has quicquam in eo, h in eo quicquam, f culpam, Q in eum
quincam. Thus we seem to have two glosses, corresponding
to the single repeated gloss of D (to e^?;?), the one gloss
meaning 'the issue,' the other 'a fault in Him.' The
phenomena of the passage however are all explained when
we remark that there is a Syriac word which could give rise
to both forms of the gloss. The word iivrj means both after
and against. Thus it occurs in the former sense in e.g.
Matt. iv. 19 'Come after-me (.*ii\=j)' ; in the latter sense in
the Peshitta of Acts xxv, 7 ' accusations many and-hard
' The special phrase was no doubt suggested by v. 58 sedebat cum ministris ut
uideretfinem (vg), where aff^hn qr have exitum rei. The Bezan Latin has : non
inuenertmt sequentia.. ,rei stquentia.

HARMONISTIC INFLUENCE. 79
bringing were-they against-him (coiAxa),' xxviii. 18 'be
cause that-not they-found against-me (.^Wrai) any accusation
which-worthy (was) for-death.' Further, this preposition or
a kindred phrase is used in the New Testament to render ef^?.
In Lc. vii. 11 (iv rw e^^?) the Sinaitic has ^i&ua ^JMO
(' and afterwards '), the Peshitta ' on-the-day which(-was)-
after-it (cniAva,!)'; in Lc. ix. 37 (rfi |^^9 vP'epci) the Peshitta
has the same phrase as in vii. 1 1. Thus the rb 6^179 of the
Bezan gloss is a quite natural rendering of the supposed
Syriac word.
The source of the gloss is doubtless to be found in
Jn. xix. 4, where the Peshitta (the Sinaitic and the Curetonian
being both wanting here) has : 'not finding (am) I against-
Him (eniAxa) not-even one fault (K'Au^.).' The form of the
gloss perhaps varied : oni Aval >s,"vs»3 (anything which(-was)-
against-Him) or K'Au^. cniAua (against-Him a-fault).
It is unfortunate that neither the Curetonian nor the quota
tions in Ephrem's Commentary or in Aphraat supply evidence
as to any Syriac reading in Matt. xxvi. 59, and that the Sinaitic
MS. is largely illegible here. But the few words which do
remain in the latter are important. They are these :
 K',1ori09  59
witness
cux^z.re' r^o  60
they-found and-not
..•.icouo T<r^j.^ao 
witnesses-of many
they-found and-not falseness
K'A*  X  KlsaA
^..•iAt oAtrC  ctLa
two there-came in-the-s[equel]

8o THE SYRO-LATIN TEXT OF THE GOSPELS.
K'iiiOX.  iwri'
falseness ot[her]
said this-man and-saying
Thus the Sinaitic agrees with D in inserting a second 'and
they found not! It has nothing to answer to the first rb e^^?;
but there is a line, more or less illegible, which evidently
answers in sense to the words ' anything against Him.' The
letters r^h\ at the end of the line have the appearance of being
a fragment of the word re'Avi^. (fault). If this be so, is it
possible that the letters earlier in the line have been wrongly
deciphered from the photograph, and that the whole line should
be restored (see Jn. xix. 4) thus 1
fault one not-even against-Him
We pass on to the remaining points, (i) In the fifth line
the Bezan diverges from the true text. The Syriac is obliged
to paraphrase the genitive absolute : hence the Peshitta
(the Sinaitic being illegible) has : ' And-there-came many
witnesses-of falsehood.' This verbal construction is followed
by D. (2) In line 7, where the Syriac (Sin. Pesh.)
naturally renders TrpoaeXdbvre'i. ..evnav by two verbs coupled
by and, the Bezan again follows the Syriac construction.
The simple verb rjXdov corresponds exactly with the ' there-
came (cth\T<)' of the Sinaitic. (3) In line 7 the 'two'
are defined both in the Sinaitic and in D as 'false-wit
nesses,' possibly through assimilation to Mc. (e-^evBofiaprv-
povv). (4) The last line (rovrov. . .^Kovaafiev) is derived
from Mc, the phrase from Mc. being used in the Arabic
Tatian, where the whole passage runs thus (Hill, p. 238) :
' And they took counsel against Jesus to put Him to death.
And they sought false witnesses, who should bear witness
against Jesu.s, that they might put Him to death ; and they
found them not, and many false witnesses came ; and their
witness was not in agreement. But at last came two false
witnesses, and said. We heard Him say.'

HARMONISTIC INFLUENCE. 8 1
Matt, xxvii. 28. kai cnAycantcc ayton- cimation nop(l)YpoYN
KAI x^amyAan kokkinhn nepieGHKAN aytco
The words Ifidriov TTop(f)vpovv come from Jn. xix. 2. This
conflate reading seems to have been invented or adopted by
Tatian, for in the Arabic Tatian (Hill, p. 243) we read : 'And
they stripped Him and clothed Him in a scarlet cloke, and
arrayed Him in a purple garment! Lower down (p. 245 f.)
we read : ' They took off from Him the purple and scarlet
garment with which He was clothed.' The Sinaitic (Matt.
xxvii. 28) has: 'And-they-clothed-Him in-garments of-scarlet'
and-of-purple (rSlaa.^rc'.TO K'Aiiicuit.'i K'Aijjj).' The
phrase used in the Peshitta of Jn. xix. 2 (Sin. and Cur. being
wanting) is rOo^jrc*.! i^AvmJ. In Lc. xxiii. 11 (iadrjra
XafiTTpdv) the Cutetonian (Sin. omitting vv. 10-12) has
' beautiful garments (K'iA^i. rC'Aviii),' the Peshitta ' garments
of scarlet (t<'Auio.M\i K'AvaJJ),' a phrase apparently taken
from the Old Syriac (Sin.) of Matt, xxvii. 28.
The conflation is found, the form slightly varying, in 157;
abcffPhqgatE a""^ Q Y ; and in Origen (lat.).
Luke iii. 23 — 38.
The Genealogy in D is a combination of that found in the
true text of Lc with that found in Matt.
From Abraham to Adam that of Lc. is followed, except
that in v. 36 between Sala and Arphaxad the name Cainan
is omitted, for which omission D appears to be the only
authority. From Joseph to Abraham the genealogy of
Matt, is followed, except that (i) between Jechoniah and
Josiah two names are inserted, viz. Joakim and Eliakim ;
(2) between Ozias and Joram three names are inserted, viz.
Amasiah, Joas, and Ochozias. In this latter portion the
notes of time and circumstance given in Matt. (vv. 2, 3, 5, 6,
II, 12) are omitted, and the enumeration is gonformed to
Luke's method.
No other text of the New Testament, so far as I know,
except D gives this combination of genealogies.
C. 6

82 THE SYRO-LATIN TEXT OF THE GOSPELS.
The genealogy however found in D is also found in the
Homily of Aphraat ' On the Cluster ' (Wright, p. -^^Af,
Bert, p. 392)'. The only point, of difference between D and
Aphraat is that the former inserts Eliakim between Josias
and Joakim (= Jechonias or Jehoiakim, Matt, i, 1 1 f ). Since
however Eliakim is another name for Jehoiakim (2 Kings
xxiii. 34), it was probably first added as an alternative gloss
and then crept into the genealogical series. It should further
be added that the name Cainan, which has no place in D, is
deliberately omitted by Aphraat, for he expressly notes
(p. V-^-Ax) that Arphaxad begat Sala.
Thus we have here a piece of harmonizing common to D
and Aphraat. It is well known that Tatian's !Diatessaron^\d
not_include_the_genealQgies._ We may then draw one or other
of two conclusions, either (i) that both D and Aphraat derived
the_genealogv from some very early work on the genealogies,
which harmonized Matt, and Lc. (comp. Bert, p. 391 n.), or
(2) that both used some recension^jf^^atianVwork^like the
anonymous Harmony which Victor of Capua believed to be
Tatian's, which contained a harmonized genealogy.
Luke iv. 31. kai kathAGcn
eic KACt)ApNAOYM nOAiN THC T^-^IAAIAC
THN uapaGaAaccion eN opioic
zaBoyAoon kai Necj)GAAeiM kai hn
AiAackcon aytoyc eN toic caBBatoic
The interpolated words come from Matt. iv. 13 and, it
appears, are added in Lc. by no authority except D.
The Arabic Tatian (Hill, p. 65) at this point is as follows:
' This is the second sign, that Jesus did, when He returned
out of Judaea into Galilee (Jn. iv. 54). And He was
preaching in the synagogues of Galilee (Lc. iv. 44) : and
' The date of tTiis Homily is A.D. 345 (Wright, Preface, p. 6). The genealogy
is quoted by Cureton, Gospels, p. vii. f., as if the Homily were the work of Jacob
the Persian Sage. It appears however that Jacob died in A. D. 338 and that the
Homily should be assigned to Aphraat (see Wright, ubi supra, and Syriac
Litei-ature, p. 31 f.).

HARMONISTIC INFLUENCE. 83
leaving Nazareth He came and dwelt in Capernaum, in the
seaside parts, in the borders of Zebulun and Naphtali : that it
might be fulfilled  to them did light spring up (Matt. iv.
13-16). And He was teaching them on the sabbaths (Lc.
iv. 31 b).' Thus the words from Matt. iv. 13 stand in the
same position (if the prophecy from Isaiah be omitted) in
Tatian and in D.
Luke v, 10 f. 0 Ae eineN
AYTOIC AeYTe KAI MH fciNecGe AAieic
iXGycon noiHcco pAp ymac AAieic ANGpconco
01 Ae AKOYCANTeC nANTA KATeAeiyAN
em THC THC kai hkoAoyGhcan aytco
The true text is /cat eirrev Trpb<; rbv ^ifiwva 'lrjaov<; Mr)
<f>o0ov' flTTo rov vvv dvdpwTTOVi 'ear; ^wypwv. koi Karayayovre'i
ra TTXola iiri rr)v yrjv d<j)evre<; Trdvra -^KoXovdrjaav avrw.
The parallel account in Matt. iv. 19 f. is : /cat Xe76t avroh
(Mc. i, 17 Kai eiTTev avroh b 'iTjaow) AeuTe oTTiaw fiov, Kai
TTOlrjaw vfia<; (Mc yeveadai) dXeeh dvdpwTTWv. oi Be evdewi
(Mc, Kai evdhs) d^evTe<i rd BUrva rJKoXovdrjaav avrw.
The only text which agrees with D here is its constant
companion, the Old Latin e, which has : ' qui ait ad simonem
ihs [d, ille autem dixit illis uenite et] nolite esse [d, fieri]
piscatores piscium faciam enim uos piscatores hominum.
Illi autem [d, ad illi] cum audissent [d, audientes] omnia
dimiserunt [d, dereliquerunt] super terram [d, super terra] et
secuti sunt eum.'
The chief points are as follows, (i) The passage is
obviously the_result of an attempt to weave into one the
Synoptic accounts of the__Apostles' call. It is clear from
Ephrem's fragments that Tatian used parts at any rate of the
Lucan account (Hill, p. 340). In Ciasca's Arabic Tatian
(Hill, p. 62) we have both accounts — Matt, iv, i?-22 followed
by Lc. V. i-i I. (2) We have an indication of retrans-
lation in TTdvra KareXeiAJrap. For (a) the word d^evre^,
common to the three Synoptists, is changed ; (b) the
6—2

84 THE SYRO-LATIN TEXT OF THE GOSPELS.
participle has become an indicative. The Sinaitic and the
Peshitta have in Lc. v. 1 1 : ' and-they-left (cinn.T.o) every
thing and-vfent after-Him.' It will be noted that (a) the
word Si~\ T, would naturally be retranslated by KaraXeiTreiv,
\of which it is the constant equivalent, see e.g. Matt. iv. 13,
pcvi. 4, xix. 5 ; (b) the Bezan construction (they left... and)
/tallies with the Syriac construction (p. 115). (3) The
evidence which we now possess in the Sinaitic Syriac makes
it clear, I think, that the gloss fiij yiveade dXieh ixdvwv
sprang up in, or in connexion with, a Syriac text. In
Matt. iv. 18 f. (rjaav ydp dXeeh ¦ ¦ .TToiijaw iifid<; dXeet? dvdpwTTcov)
the Sinaitic has : ' Because that-fishers (lit. hunters) were-they
offish (r^JiCO^ oocn k'.tj^s): He said to-them Come after-
me and-I-will-make-you fishers^ (lit. hunters) of-men (r^.T»^
KlxJr*' jAs.i).' The very natural translation of the phrase
' they were fishers ' by the words ' they were hunters of fish',
found in the Sinaitic alone, supplies the material for the Bezan
interpolation ' Be 7iot fishers of fish!
Luke V. 14 f. kaGcoc npoceTAleN mcoychc ina eic
MApTYplON HN YMEIN TOYTO O Ae
eleAGcoN nplATO KHpYcceiN kai
AIA({)HMeiZeiN TON AOfON COCTe MHKCTI
AynacGai AYTON (fiANepcoc 6IC noAiN
eiceAGeiN aAAa eJa hn cn epHMoic
TOnOIC KAI CYNHpXONTO TipOC AYTON
KAI HAGeN elC KA(|)ApNAOYM
AiHpxeTO Ae 0 Aofoc maAAon nepi aytoy
^ It is worth while to notice that the root S»» (hunt, capture) runs all through
this history as given in the Syriac Gospels. It is not only used to translate aXiefs ;
but (a) the phrase in Lc. v. 9 (iirl ti; dypa rCiv IxBvav wv avviXa^ov) becomes in
Syriac (Sin.) OnuK'.l (<llOJl OCn T*^H- "^ (at that capture of-fishes
which-they-look ; ¦wb.ic\i-ih£y-captured (Ci^ -«1 ) Pesh.) : (j3) the words avSpiirovs lorj
^uypuv (Lev. 10) are in Syriac (Sin.) »<L».ii,\ .IK'^ K'OOoAt r9fv\rtf-^\^
(men shalt-thou-be capturing for-life (or -salvation)).

HARMONISTIC INFLUENCE. 85
The interpolated words come from Mc. i. 45, where the
true text has d rrpoaer. yiwvafj'; eh fiapr. avroh. 0 Be i^eXdwv
rjp^. KTjp. TToXXa (om. D) Kai Biatf). rbv X., ware firfKen avrbv
(om. D) Bvvaadai (f}avepw<; eh tt. elaeXdelv (D (j). elaeX. eh rr.),
aXXa e^co c'tt (D ev) ip. r. rjv. Kai vpx- tt/so? avrbv rrdvrodev.
Kai elaeXdwv rrdXiv eh K. Bi' rjfiepwv (D /cat) rjKovadrf on
ev o'lKw iariv.
In the Arabic Tatian (Hill, p. 129 f.) the account in Mc.
is followed. It would appear therefore that the interpolation
from Mc. in D is to be traced to Tatian. For the quotation
in Ephrem see below.
The phrase iva eh fiaprvpiov fj vfilv rovro is of special
interest and importance. (i) The phrase et? fiaprvpiov
avroh occurs in the New Testament in the following places :
Matt. viii. 4, x. 18; Mc. i. 44, vi, ii, xiii. 9; Lc. v. 14,
compare Lc. ix. 5 (et? fi. irr avrovf). Of these passages the
Sinaitic is wanting in Mc. i. 44 ; in the rest it has varying
translations, viz. for-a-testimony to-them (Mc. xiii. g),for-their-
testijnony (Matt. x. 18, Mc. vi. 11), that-it-may-be to-them (to-
you, Lc.) a-testimony (Matt. viii. 4, Lc. ix. 5), that-it-may-be to-
them for-a-testimony (Lc. V. 14). Of these passages only two,
viz. Matt. viii. 4, Lc. ix. 5, have a place among the Curetonian
fragments. In both these passages the Curetonian has ' that-
it-may-be to-tlum for-a-testimony! Neither in Matt. viii. 4
nor in Lc. ix. 5 has any other authority this form of
the phrase, so that it appears to be an indigenous Syriac
growth. For the toOto of D compare the Curetonian
rendering of Lc. xxi, 13 (dTTo^rjaerai vfiiv eh fiaprvpiov) —
rc'.icn r<'A»o.icaoo ^.1 »_oaA r^ocnA\
this-thing a-testimony however to-you it-shall-be
(2) The vfiiv of D (instead of avroh) points back to a
confusion in Syriac between »_aofA (to-them) and ...ft^
(to-you), the confusion being facilitated by a reminiscence of
Lc. xxi. 13. We have the same phenomenon in the Sinaitic
of Lc. ix. 5, where »__eLaA (to-you) answers to the true text
ctt' avTov<; (N* and several cursives avroh), no other authority.

86 THE SYRO-LATIN TEXT OF THE GOSPELS.
so far as I know, reading to you^. The fact that the corruption
points to a Syriac source is obviously a strong confirmation
^ The confusion between the suffixes ^_Cl^ and ^__O0D may be
compared with the confusion in Greek MSS. between ymcon and hmcon. Note
the following passages. In Matt. v. 12 some MSS. (see Tisch.) add ol waripes
avTuv from Lc. vi. 26 ; Cur. (see Baethgen, Evangelienfrag., p. 8) has 'your-
fathers (^^cy^jCna (<")•' In Lc. xi. 44 Sin. has 'men walk over them
(^  nm\ '-ty AiA),' Cur. has 'over joa (__f\'>.< -giV' In Acts xiv. 17
(vixw...Tds KapSlas vp,av) the Peshitta (with arab. theb. Ath.) has ' to-them... their-
hearls (^OoA, ,.^OeoA\aaA)-' In i Cor. vii.'i^. (iirel dpa rd riKva vp.uv
aKdBaprd ioriv) the Peshitta has ' their-sons (^^OCTlxlra).' In discussing
Acts ii. 17 I urged (Old Syriac Element, p. 18) that the Bezan reading 01 yiOl
AYTCON KAI GyrATCpec AYTCON points to an underlying Syriac text.
Mr Rendel Harris (Four Lectures on the Western Text, p. 90 n.) criticises my
position thus : 'In the foregoing remarks I have avoided the discussion of certain
test passages which Mr Chase considers decisive, because they are not, at all
events as presented by him, of the nature of proof. It is not fair, for example,
to quote the reading " their sons and their daughters " in Acts ii. 1 7, in proof of a
Syriac origin of the Bezan text of the Acts, and to support the statement by
reference to Tertullian (Adv. Marc, v. 8), without at the same time informing the
reader that Tertullian is expressly, and from the necessities of the case, quoting
Joel against Marcion, and that the Bezan text shews signs of having been corrected
to the text of Joel ! The argument needs re-statement, to say the least.'
The gravity of the accusation made in this paragraph is plain. It will be a
sufficient apology for some fulness of treatment. I pass at once to the two counts
of Mr Harris' indictment.
(i) It is perfectly true' that Tertullian 'from the necessities of the case'
appeals to Joel. But the only point which has any bearing upon the question
at issue is What text of Joel's prophecy does he quote ? Now if an English
writer were to quote Joel's prophecy of Pentecost without turning to the passage
in the Old Testament, the probability is great that he would give the familiar
words of the New Testament quotation (Acts ii. 17) : 'And it shall come to pass
in the last days ' and not the original text ' And it shall come to pass afterward.'
That Tertullian as a matter of fact does this, i.e. quotes the version of the
prophecy given in Acts ii., is placed beyond possibility of doubt when we turn to
the passage (Adv. Marc, v. 8). It runs thus : ' lam nunc et ilia promissio
spiritus absolute facta per Joelem : In noiiissiinis diebus effundam de meo spiritu
in omnem carnem, et prophetabunt filii filiaeque eorum, et super seruos et ancillas
meas de meo spiritu effundam. Et utique si in nouissimos dies gratiam sj^iritus
creator repromisit, Christus autem spiritalium dispensator in nouissimis diebus
apparuit, dicente apostolo. At ubi tempics expletum est, misit deus filium suum, et
rursus, Quia tetnpus iam in collecto est, apparet et de tcmporum ultimoi-um

HARMONISTIC INFLUENCE, 87
of the position taken up above that the reading itself is a
Syriac reading, (3) In Latin authorities we find alike
the reading and the corruption of the reading : (i) g^-^ e have
' ut in testimonium sit illis ' ; (ii) c has ' ut sit in test, nobis
istud' \ abff^qr Ambr, (in Luc) 'ut sit in test, hoc nobis!
1 ' ut sit in test, nobis! (4) But there is a further point.
The concurrence of Tert. Adv. Marc. iv. 9 (ut sit nobis in
testimonium) and Epiph. p. 322 f. (ti^a ^ fiaprvpiov rovro vfilv)
praedicatione hanc gratiam spiritus ad Christum praedicatoris pertinere.' Thus
Tertullian bases an argument on the words in nouissimis diebus, words which do
not occur in the Hebrew or in the LXX. text of Joel, but which are found in the
version of Joel's prophecy given in Acts ii. If any further argument were needed
to prove that Tertullian is quoting the text of Acts ii., it is to be found in
the coincidence of his quotation with that in the Passion of St Perpetua (ed.
Robinson, Texts and Studies, vol. i., No. x, p. 60 ff.). The whole passage is as
follows : ' Sed uiderint qui unam uirtutem Spiritus unius Sancti pro aetatibus
iudicent temporum : cum maiora reputanda sunt nouitiora quaeque ut nouissimiora,
secundum exuperationem gratiae in ultima saeculi spatia decretam. In nouissimis
enim diebus, dicit Dominus, effundam de Spiritu meo super omnem carnem, et
prophetabunt filii filiaeque eorum : et super seruos et ancillas meas de meo Spiritu
effundam: et iuuenes uisiones uidebunt, et senes somnia somniabunt.'
(2) I pass on to examine Mr Harris' second proposition, viz. ' that the Bezan
text shews signs of having been corrected to the text of Joel.' The only
satisfactory way of dealing with this question is to write out in four parallel
columns (i) the Hebrew text of Joel ii. 28 ff., (2) the LXX., (3) the ' true text' of
Acts ii. I7ff., (4) the Bezan text. Any one who will take the trouble thus to
compare these four texts will see that the position that the Bezan text has been
' corrected to the text of Joel ' is supported by one, and only one, piece of
evidence, viz. the omission in the Bezan text of the words Kai npo^Tp-eioovai
(v. 18), words not found in the Hebrew or the LXX., but inserted in the 'true
text' of the Acts. But further examination of the Bezan text shews that this
argument is worthless. This omission in the Bezan text of the quotation from
Joel is only one of many omissions. Abbreviation is the characteristic feature of
the Bezan text at this point. The following words are omitted in D — v. 17,
Koi (before iaraC), al (before Bvyaripes), i/iSv (after oi veavla-Koi and after oi Trpea--
pirepoi), ivuTvlois; v. 18, iv tois 7ip.ipai.s iKeivais, Kai vpo<l>y)Tevaov(rLV ; v. 19, aXfia
Kai irSp Kai drpiSa Kairvov ; v. 20, Kai iTrupavfi. In view of all these omissions,
it is impossible from the omission of the words Kai ¦irpo<prirev<Tova-iv to deduce
the conclusion that the Bezan text has ' been corrected to the text of Joel.'
The case indeed may be stated thus. Between the text of Joel (Hebr. LXX.) and
the true text of Acts there are some seven points of difference. Between the
former and the Bezan text of Acts there are, I believe, fifteen.
Mr Harris' case then for a Hebrew origin of the Bezan avTQv...airuv (in
place of the true text ip,uv...vp.Qv) breaks down on examination.

88 THE SYRO-LATIN TEXT OF THE GOSPELS,
in ascribing this Syriacised reading to Marcion shews that it
was really his ; and thus this Syriacised reading must
have been current as early as the middle of the second
century'. A few clauses of the passage are found in the fragments
of Ephrem (Hill, p. 355): 'Lord, if Thou wilt. Thou canst
heal me... and He stretched forth (His) hand (and) touched
him..,. [Tell] no man, (but) go, shew thyself to the priests, and
offer a gift, as Moses commanded, for a testimony unto them.'
The context in Ephrem shews that he had before him in Tatian
the reading ' He was angry and stretched forth His hand,'
whereas the true text (Mc. i. 41) has aTrXayxvitrdeh eKreiva<;
K.T.X. Thus (see Moesinger, p. 143 ff ) we have in Ephrem
the following phrases (for the literal translation of which from
the Armenian I have to thank Prof J. A. Robinson): "On
account of two things our Lord was angry.. ..¥or ?liis that he
said 'If thou wilt,' He was angry; and for that 'Thou
canst,' He healed him..., Wherefore the Lord by wrath [a
different Armenian word] shewed that He healed not with
respect of persons.,.. Again it is said that not with him, but
with the leprosy, Christ was wroth!' There can, I think, be
no doubt that Tatian had the reading ' He was angry.' Can
we explain it by a reference to the Syriac .'' In the account
of the miracle in Mc. i. 40 ff., the account which we have
above seen reason for thinking that Tatian mainly followed,
we have the words /cat arrXayxyiadeh. For this the Sinaitic,
the Peshitta, the Harklean, and the Jerusalem versions have:
(Jer. ..iOJl:^) ..aOfsolv. ^Mi^r^ (lit. He-pitied upon-him).
Now if a Syriac scribe in writing the word )a*jiA^r^ omitted
by mistake the i before m, the word would become )a*jA\p«'.
An obvious way to convert this vox nihili into a real word
would be to add h\ and make it into Av5aMA\t<' (He-was-
angry). In some such way it would be very easy for ^tOM^i^'
(He-was-angry) to arise from ^njjiA^rC (He-pitied). It should
1 The evidence of Epiph. is express : Siia-rpef as Si rd pryrbv, u MapKlav, dvri
Si rov elirecv E^s p.apr. airots, Mapr. Xiywv vfiiv.

HARMONISTIC INFLUENCE. 89
further be added, if it is assumed that .^caai^ (upon-him)
was part of the original Syriac reading, that the Syriac verb
' to be angry ^ like the verb ' to pity' is constructed with this
preposition ; see Acts iv. 2.
This Syriac corruption is found in the text of Mc. in D
(kai opricGeic cKTeiNAc), a ff^ r* (iratus) ; while b g\ having no
participle at this point, seem to afford evidence of the
presence in their respective archetypes of a strange and
difficult reading, which the copyists desired to avoid.
Whether this reading was due to Tatian, or whether it
already existed in the Syriac text of Mc. which he in
corporated in the Diatessaron, is a matter which cannot be
certainly decided. The evidence is quite consistent with the
latter alternative.

Luke vi. 42.

H ncoc Aynacai AereiN tco AAeA4)a coy
^ A(})ec eKBAAo to KAp(t)OC eK toy
0(t)GAAMOY COY KAI lAOY H AOKOC eN too cco
ocpGaAmco YTTOKeiTAi YUOKpeiTA eKBAAe
npCOTON THN AOKON CK TOY 0(t)GAAMOY COY
KAI TOTe AiABAeyeic eKBAAeiN
TO KAp(})OC eK TOY o(()GaAmoy toy AAeA(})OY coY-

The true text of Matt, and Lc. is as follows :
Matt. vii. 4 Luke vi. 42
I? TTws epetff TG> ah. aov ttoJs hvv. X. rta ab. trov
"A0ES CK^. TO K. eK TOV 6(f)d. o'ov, 'A6eX</)e', acjies eK^, to it. to ev ra
0(^6. crov,
Kat toov rf 0. ev tS o(j)$. (rov; avTos rrjv ev ra 6(^6. (rov boKov ov
^Xeiroiv;
vTTOKpiTa, CK^, Trp, eK TOV o(j)6. (rov VTTOKpiTo., eK/3. irp. Trjv 8. eK tou
Tiji/ 8., 6(j)d. trov,
Kai Tore 8ia^\. c'k(3. to k. ck tov koI totc 8(aj3\. to k. to ev t<3 ocfid.
6(pB. TOV aS. (rov. tov aS. (rov exfi.

90 THE SYRO-LATIN TEXT OF THE GOSPELS.
It is clear that D here takes certain phrases from Matt.
But obviously the problem of the passage lies in the word
vTTOKeirai. The points to be considered are the following.
(i) The Bezan Latin is: trabis in tuo oculo est; comp. e in
oculo tuo trabis est. The reading therefore is not taken over
into the Greek from the accompanying Latin, (2) A
possible explanation is that YuoKpiTA was dittographed, and
that the word at its first occurrence was read by a scribe
as YTTOKeiTAi. To some this suggestion may commend itself
(3) When we turn to the Curetonian of Matt. vii. 4 (where the
Sinaitic is wanting) and to the Sinaitic of Lc. vi. 42, in both
places we read : 'because- (Sin., and-) lo in-thine-eye which-is-
thine a-beam (is) set (r^'saxJto)-' There is, it appears, no
other authority for this reading in Matt. It will be specially
observed that the reading of the Sinaitic in Lc. is assimilated
to Matt, for the sentence is wholly different in the true
text of Lc. (aLrT09 r-qv iv ra> bcf)d. aov BoKbv ov ^XeTTWv).
Here then in the Sinaitic a harmonizing influence has
operated. It must be noticed that (a) r^isixs^ is quite a
simple and obvious word in this connexion, being used in
reference to building (see e.g. Matt. vii. 25, Lc. vi. 48, i Cor,
iii. 10, I Tim. vi. 19); (b) that vTTOKenai would be a not
unnatural rendering of this Syriac word, for the latter, besides
being the constant equivalent of the simple verb Keladai, also
represents drTOK. (Lc. xix. 20, Heb. ix. 27), irriK. (i Cor. ix. 16),
TTpoK. (Heb. xii. i). These two points just noticed, together
with the fact that the Bezan Latin did not suggest the
reading, and the further fact that this Old Syriac reading
occurs in Matt. vii. 4, with which Lc. vi. 42 is harmonized,
make it most probable that the Bezan vTTOKeirai represents
an Old Syriac, possibly a Tatianic, reading.
A large group of Old Latin authorities in Lc. (a b c ff^* 1 q
aur.) has in oculo tuo trabes subiacet.
Luke viii. 35. nApAreNOMeNooN Ae eK thc noAecjoc
KAI GecOpHCANTCON KAGHMCNON

HARMONISTIC INFLUENCE. 9 1
TON AAIMONIZOMCNON CCO(t)pONOYNTA
KAI IMATICMeNON KAGHMGNON

nApA TOYC noAAC toy ihy /v e{t)oBHGHCAN.
The Bezan text is the result of harmonizing with Mc. v. 1 5.
The true text of Lc. and Mc. is as follows :
Luke viii. 35 Mark v. 15
e^rjXdov 8e i8eii' to yeyovbs Kol ^dav Ka\ epxovrai npos rov 'Irjo-ovv,
irpos TOV lr](rovv,
Kai evpav KaStjpievov rov avBpamov K(u 6e(opov(riv rov Saiiiovi^o/ievov
a(j) ov ra Saifuivta e^rjXBev KaOrjuevov
ipLaTKrjjievov Kai (r(0<j)povovvTa IfiaTurpevuv Koi cr(i>(^povovvTa,
Trapa Toiis iroSas tov 'li;croC, Kai e(f)0- rov ea-x^Kora tov Xeyiava, Kai e(j)o-
^ridrj(rav. ^ij6ri(Tav.
It seems clear that the Greek of Mc. has affected the
Bezan Greek of Lc. indirectly, and that the Bezan Greek is
the result of rough retranslation from a version. In the first
line the words e/c Ti79 TroXetos are due to context-assimilation
(aTT^yyeiXav eh ttjv ttoXiv V. 34), but throw no light on
the problem of the passage. We turn to the Kadrjfievov
repeated before the words iTapd roii^ Tro'Sa9. The Sinaitic
and the Curetonian have : ' And-they-went-out (even) the-
men and-saw (Cur., that-they-might-see) that which-was, and-
they-came to Jesus, and-found that man from whom there-
went-out those devils, while clothed and-sober and-sitting at-
the-feet of-fesus! The Peshitta, differing in some small points
from the Old Syriac text, has the words 'and-sitting' before
the words 'at-the-feet of-Jesus^' The Peshitta in Mc. has:
' And-they-came-out to-see that which-was, and-they-came to^
Jesus, and-they-saw that-man whose(-were)-the-devils while
clothed and-sober and-sitting^, him in whom was the-legion
and-they-feared.' Thus the position of the words 'and-
sitting' late in the sentence seems to be characteristic of the
Syriac Versions, and for this position a reason can be given.
' The Armenian and Aethiopic versions seem to be the only other authorities
for this position of' and sitting ' in Lc.
^ Sin. agrees with Pesh, as far as the M'ord ' to.' It then becomes illegible.
5 There appears to be no other authority for this position of the word in Mc.

92 THE SYRO-LATIN TEXT OF THE GOSPELS.
The Syriac versions constantly give a paraphrastic translation
of such a preposition as rrapd in rrapd rov^ TrdSa?, inserting a
verb, participle or the like. Thus in Matt. iv. 21 (elBev dXXow
Bvo dBeX(pov<;...iv rw ttXoiw), the Sinaitic and the Curetonian
alike insert sitting to give the force of iv — ' He-saw two other
hret!nren... sitting in-the-ship'; so also in Matt, xv, 39 ive^T)
eh TO ttXoiov becomes in the Sinaitic and Curetonian ' He-
went-up. He-sat Him (ea\ csAv») in-the-ship ' (see above,
p. 32 E, and Baethgen, Evangelienfragmente, p. 16 f).
It should be noticed that earlier in the passage (Lc, viii. 27)
the Curetonian introduces an interpolation (absent from the
Sinaitic) from the parallel passage in Mc. — 'and-every cry
crying was-he and-smiting was-he himself with-stones ' (from
Mc. V. 5). So also X 64 /cat ev roh opeaiv rjv Kpd^wv Kai Kara-
KOTTTWV eavrbv Xidois:^.
Luke xi. 2. o Ae eineN otan
npoccYXHCGe mh BATToAoreire coc 01 Aoinoi
AoKOYCiN fAp TiNec OTI eN TH noAYAofeiA
AYTCON eiCAKOYCGHCONTAI aAAa npoceYXOMCNOi
Aerere nATep hmcon 0 cn toic oypanoic.
The interpolation comes from Matt, vi, 7 ff., where the true
text is Trpoaevxbfievoi Be firj /SarraXoyijarjTe wairep ol edviKol,
BoKovaiv ydp on iv rfj TToXvXoyla avrwv eiaaKovadrfaovrai...
0VTW<; ovv TTpoaevxeade vfieh TIdrep rjfiwv b iv roh ovpavoh-
The interpolation seems to be peculiar to D, The
Arabic Tatian at this point (Hill, p. 78) has the directions
about prayer (Matt. vi. 5 — 8), then the request of one of the
disciples (Lc. xi. i b, 2), then the Lord's Prayer. Thus the
Bezan reading here coincides substantially with the arrange
ment of the discourse in the Arabic Tatian.
1 In the Arabic Tatian (Hill, p. 87) this clause has a. place, but in the rest of
this narrative Lc. is in the main followed. The fragments -of Ephrem (Hill,
p. 344) are too scanty to afford evidence as to the composition of the Diatessaron
at this point.

HARMONISTIC INFLUENCE. 93
The remarkable points in the Bezan interpolation are
(i) the substitution of &)9 ol Xoittoi for warrep ol idviKoi,
(2) the insertion of Tti'e9 after BoKovaiv ydp.
In Matt. vi. 7 the Curetonian, assimilating to t/. 5, has 'as
the accepters of persons' (i.e. the hypocrites). It is remark
able that Cod. B is the only other authority, it appears, for
this reading. The Sinaitic and Peshitta have ' like the-profane
(rc*«Mjj)'; the Harklean 'like the-Gentiles (rsL».sa,Jsas».).' I
venture to suggest that behind the Bezan reading there lies
an Old Syriac reading or gloss^ — the result of assimilation —
' as the-rest of-men.' Compare Lc. xviii. 1 1 (God, I thank
Thee that I am not as the rest of men (r^lx.irt'.i rci^ix.)),
I Thess. iv. 13, v. 6, Eph. ii. 3, iv. 17 — in all which places the
phrases ' the rest,' ' the rest of men ' etc., are almost equivalent
to ' the Gentiles.' If this was an Old Syriac reading or
gloss, we have an explanation of the Bezan rive<;; for the
word p^JEJpfi'.i could easily be taken with the following clause
with the meaning ' because-some! It is obvious that this
suggestion has a higher probability, inasmuch as it offers an
explanation of both the characteristic phenomena of the
Bezan text at this point.
Luke XX. 20. KAI Ano)(6opHCANTec AnecreiAAN
CNKAGeroYC.
The true text has Kai TrapaTr)prfaavTe<; k.t.X. The Bezan
a7ro%ft)/3?7a-az/T69 (d, recedentes) implies harmonizing and re
translation. It seems to be derived from the Tropevdevre'i of
Matt. xxii. 15. The Arabic Tatian (Hill, f 178) has: 'Then
the Pharisees went away, and took counsel how they might
catch Him in His talk (Matt. xxii. 15), and deliver Him up to
the authority of the court and to the authority of the governor
(Lc. XX. 20). And they sent to Him their disciples' (Matt. xxii.
16). The Syriac versions (Sin. Cur. Pesh.) represent rropev-
devre<i (Matt. xxii. 15) by the word ciiiK*. The Greek aTro-
Xwpeiv would be a very natural translation of A\r^, the latter

94 THE SYRO-LATIN TEXT OF THE GOSPELS.
in fact being the representative of the former, e.g. in the
Curetonian of Matt. vii. 23.
It is remarkable that in Lc. xx. 20 neither the Peshitta
nor the Curetonian nor the Sinaitic has a verb to correspond
to d'TToxwpijaavre'; (TTaparrfprjaavTe<;) ; the two latter insert
the word 'afterwards.'
The Latin authorities are worth noting. In Matt, the
readings of the Old Latin MSS. are Tunc abeuntes (g' ; so vg),
tunc abieritnt et (f, q) ; in Lc. f i 1 q cum recessissent^ , a cum dis-
cessissent, c cum redissent, e et secesserunt et. It seems clear
therefore (i) that the Bezan drToxoip. cannot come from Matt.
through the medium of the Old Latin ; (2) that the Old Latin
renderings are different translations of the Greek aTToxcopTJ-
aavTe<;. An important reading at the end of the verse claims
attention. Instead of the true text (ware TTapaBovvai avrbv ry
^PXV '^"'' "^fi i^ovaia rov rjyefLovof;) Cod. D has oocTe uapaAoynai
AYTON TCO HreMONi. The only authorities, it appears, which
share this reading with D are its constant companions e et
traderent ilium legato and the Curetonian ' and-deliver-Him
to-the-governor.' The whole clause is omitted in the Old
Latin i.
Luke XXI. 7.
ennpcoTHCAN Ae ayton 01 maGhtai
AeroNTCc AiAACKAAe nore ^ tayta gctai
KAI Tl TO CHMeiON THC CHC cAeYcecoc.
The true text is...TroTe ovv'^... ; Instead of rrfi arj<; iXeii-
aewi it has brav fieXXrj ravra yiveadai.
The passage is assimilated to Matt. xxiv. 3 : TTpoarjXdov
avrw ol fiadrjral Kar IBiav Xeyovreif EtTTOi' rjfilv TTore ravra
earai, Kai n ro arffieiov T7J<; afj<; Trapovaia<; Kai avvreXeia<; rov
alwvot;. ^ Compare ff^ cmn recessit sesum.
2 The oi5;» is omitted in the Old Syriac (Sin. Cur.), me., Lat-vt-vg., arm.,
aeth., in 13 (of the Ferrar-group), 1-131-209 and a few other MSS. It has no
place in any text of Matt.

HARMONISTIC INFLUENCE. 95
The Arabic Tatian (Hill, p. 208) is at this point as follows :
' Simon Cephas and James and John and Andrew came unto
Him, and said unto Him between themselves and Him
(Mc, xiii, 3), Teacher (Lc. xxi. 7'), tell us, when shall these
things be ? and what shall be the sign of Thy coming and of
the end of the world?' (Matt. xxiv. 3). Further, the use of
the word eXevai<; — which occurs again (see p. 61) in a charac
teristic Bezan reading (Lc. xxiii. 42) — appears to be a sign
of retranslation — t^? 0-^9 TTapovaia<; (Matt. ; D t^9 tt. aov) =
v\oua\_sa.l (Sin.) = T'^9 0-179 iXevaea<; (D Lc). The Bezan
eXevai({ is an exact equivalent of the Syriac word (cf Acts
vii. 52). Thus the evidence points to the Syriac Tatian as the
source of this Bezan reading.
The only other authority for this reading, so far as I
know, is the Old Latin 1 (aduentus tui).
Luke xxiii. 45 ff. ecKOTicGH Ae 0 hAioc ^
KAI (])CONHCAC 0 IHC MefAAh (()CONH
eincN HATep eic ^eK^t^z coy hapatiGhmi
TO nNCYMA MOY TOYTO Ac eincoN eleuNCYce
KAI TO KATAneTACMA TOY N'>'OY
eC)(ICGH ^ : KAI ^ 0 CKATONTApXOC ((jCONHCAC
eAo2AzeN ton Gn.
The true text has the words e'o-%. Se rb KararTer. rov v.
fieaov between the mention of the darkness and the Lord's
commendation of Himself to the Father. Also, it has the
words in the second line in this order — /cat ^covrjaa'i (fiwvfj fi.
6 'Irjaovv. In the last line it has IBwv Be 6 eKarovrdpxv^-
The Syriac Versions (Sin. Cur. Pesh.) have : ' And-the-sun
was-darkened, and-there-was-rent the-front-of the-door of-the-
temple from its-midst, and-there-cried Jesus with-a-voice
great and-said etc'
The points are these: (i) In Syriac it is natural that
'Jesus' should stand immediately after 'and-there-cried.'
' No authority seems to have the word ' Teacher ' in Matt. Mc.

96 THE SYRO-LATIN TEXT OF THE GOSPELS.
This order is reproduced in CD. (2) Tatian, adopting
the order of events in Matt. Mc, places the rending of the veil
after our Lord's death (Hill, p. 249), compare Ephrem, ed.
Moesinger, p. 256. The Tatianic order is reproduced in Lc.
by D alone. (3) It should perhaps be noticed that
Ephrem (Moesinger, p. 256) simply has uelum scissum est.
This may be an abbreviation due to the expositor. It may
however reproduce a Tatianic reading, followed by D.
Luke xxiv. i.
HpxoNTO enei to mnhma (})AipoYCAi
A HTOIMACAN a KAI TINCC CYN AYTAIC
eAofizoNTO Ae 6n baytaic
TIC ApA AnoKYAicei TON AlGON
eAGOYCAi Ae eYpON k.t.A.
The true text has e'Trt rb fivrjfia -qXdav ^. a rjr. dpwfiara.
evpov Be K.T.X.
The Sinaitic and Curetonian are as follows : ' They-came
to-the-house-of sepulture, and-they-brought that which-they-
prepared and-there-came (Cur., -were) with-them other women.
And-they-found etc'
The main points are these : (i) D agrees with the Sinaitic
and the Curetonian in the omission of ' the spices ' : so a b c
eff^lr, theb. (2) The passage in the Arabic Tatian
(Hill, p. 252) runs thus: 'Came Mary Magdalene and the
other Mary and the other women to see the sepulchre (Matt,
xxviii. I b), carrying with them the spices which they had pre
pared (Lc. xxiv. I c). And they said among themselves. Who
shall remove for us the stone from the door of the tomb f for it
was exceeding great (Mc. xvi. 3). And when they said so, a
great earthquake took place, and an angel descended from
heaven, and came and rolled away the stone from the door
(Mc. xvi. 4 b, Matt, xxviii. 2 a). And they came and found
etc. (Lc. xxiv. 2).' In detail it should be noticed (a) that
Tatian, as Sin. and Cur., mentions ' other women ' ; D ' certain
(women) ' ; f has aliae, q r (taking Tti/69 as masculine) quidam :

HARMONISTIC INFLUENCE. 97
many authorities however have this addition ; (b) that the
addition from Mc. xvi. 3 has exactly the same position in D
as in Tatian ; it appears also in the Thebaic and in c cogita-
bant autem inter (intra d) se quisnam. esset qui (quis utique d)
teuolueret lapidem ; tt/oo? eavrd<; (Mc. xvi. 3, true text) =
^cn T^i-> (in-their-soul) = e'j/ lauTat? (D, Lc); (c) that the
Bezan eXdovaai (so Thebaic and c et cum uenisseni) reproduces
the Tatianic they came; for, as the Peshitta has not this
addition, it is probable that the Arabic here preserves the
actual reading of Tatian,
Mark viii. 10, eic ta opiA MeAepAAA.
The true text is et9 rd fiiprj AaXfiavovdd. The Bezan
Latin is : in partes Magidan. The parallel in Matt, (xv, 39) is
ei9 ra bpia MayaBdv. If Ciasca's Arabic accurately represents
Tatian at this point, he used Matt, xv, 39 (not Mc. viii. 10).
That this particular bit of harmonizing spread widely in the
Syro-Latin and related texts is clear from the evidence.
Note (i) that the name in an uncorrupted form is found in the
Ferrar-group, 1-209 (fiayBaXd), 28 2^" (fiayeBd), a ff^ (magedan),
b i (magedam), c (mageda) ; compare Aug. de Consensu ii. 3 ;
(2) that the reading ei9 rd opia is found in cf (in finibus),
arm., and is postulated by the '6p7) of Cod, N, We must
now add the evidence of the Sinaitic Syriac, which has ' to-the-
mountain(s i") of-Magedan (...jl-S^a.T p^'io,^^).' Whatever
else this remarkable reading in the Sinaitic may shew'*, it

' The word in the printed text has not the sey&mt of the plural. We cannot
however lay much stress on their omission.
^ This reading of the Sinaitic raises two questions. (a) Was there an early
Greek harmony of the Gospels ? Otherwise the reading ' mountain(s'i) of Magedan '
must almost certainly be due to a Syriac translator having before him a harmonized
Greek text which either had (cf. Cod. N), or was read by him as having,
op-q for Spia. I say almost certainly, for it is conceivable that the blunder
(' mountains ' for ' boundaries ') arose in a Syriac translation of Matt, and was thence
transplanted into the text of Mc This supposition is however improbable,
though not impossible, in face of the fact that Sin. Cur. Pesh. have in Matt, the
C. 7

98 THE SYRO-LATIN TEXT OF THE GOSPELS.
puts it beyond dispute that this harmonized reading in Mc.
was current in an Old Syriac text. The Bezan form fieXeyaBd,
which is obviously a corruption of fiayeBd, is easily explained
on the hypothesis that the Bezan scribe is here assimilating
his Greek to a Syriac text, and that this Syriac text either
actually had, or was read by the Bezan scribe as having,
the reading »_jt^^.='3, the A being generated by the right
hand stroke of the -^ being carried up somewhat high.
Mark xiii. 2. OY mh Act)eGH coAe AiGoc
eni AiGco oc oy mh kataAyGh
KAI Aia rpicoN HMepcoN
aAAoc anacthcctai ancy X€ip<J^N.
Just before the account of the poor widow, with which the
previous chapter closes (xii. 41 — 44), Tatian (Hill, p. 169 f)
inserted ' the cleansing of the Temple,' and (as he identified
the cleansing at the close of the ministry with the cleansing
recorded in John ii.) he incorporated in this history our
Lord's conversation with the Jews (Jn. ii. 19 — 21) about the
destruction and the resurrection of the Temple of His body —
' Destroy this temple, and in three days I will raise it up.'
Thus to those familiar with the Diatessaron the context in
St Mark, immediately preceding the interpolation, was closely
associated with the passage in Jn. ii. on which the interpola
tion is based. Among such readers the interpolation would
naturally take its rise \

Syriac word which answers to 8pm. (b) What is the relation of Sin. to
Tatian? Did Tatian inherit or coincide with or initiate the piece of harmonizing
found here in Sin. ? The question is a large one. Its treatment requires an
elaborate examination of the Sinaitic text.
1 A subordinate point is this. The word for 'temple' in Mc. xiii. i (twice
repeated in D) is lepbv: that in Jn. ii. 19 ff., Mc. xiv. 58 is valis. Both these
words are represented by a single Syriac word. Hence to a Syriac reader the
three passages would be more closely connected than to a Greek. The same key
word is common to them all.

HARMONISTIC INFLUENCE, 99
Mark xv. 25 ff. HN Ae COpA • r
KAI 6Ct)YAACC0N AYTON
HN Ae ^ enirpAct)H thc aitiac aytoy
enirerpAMMCNH oytoc cctin
0 BACiAeoYC TCON ioyAaicon
KAI CYN AYTCO CTAYpOYNTAI • B • AHCTAI.
The true text has ijv Be wpa rpirrj Kai earavpwaav avrov.
Kai rjv 7] iTTiypa<f)'^ Trj<; alria's avrov imr/eyp. 6 ^aa. Ttof
'lovBaiwv. Kai aiiv avrw aravpovaiv Bvo Xyard^.
The points are as follows : (i) The Arabic Tatian (Hill,
p. 247) between the division of the garments (= Mc. xv. 24)
and the account of the superscription has the words ' This
the soldiers did ; and they sat, and kept guard over Him
there' (Matt, xxvii. 36). The Greek of the words italicised
is irrjpovv avrbv, the Latin MSS. having seruabant, obserua-
bant (see Bp J. Wordsworth in loco). It appears clear then
that the Bezan iipvXaaaov avrov is the Tatianic reading.
It comes from Matt, but through the medium of retranslation.
Now in Matt, xxvii. 36 the Sinaitic and the Peshitta have
^iA^ (watching), the Peshitta adding cnl (Him). But
tA^ is the common equivalent of <j)vXdaaeiv; see e.g. Matt.
xix. 20, Mc. X. 20, Lc. ii. 8, xi. 21, 28, xviii. 21, Thus
crA ('iAg) would be naturally represented by i^vXaaaov
avrbv. This Tatianic reading appears in the custodiebant
of certain Latin MSS. viz. ff'' k n r\ (2) The addition
of 0VT09 'eanv likewise points to a Syriac medium. The
Bezan version of the superscription is that of the Sinaitic and
the Peshitta : 'This-is (rdJeo ; Pesh. euco) their-king (Pesh.
the-king) of-the-Jews,' It is probably based on the Tatianic
reading; for Tatian, as quoted by Ephrem (Hill, p, 375), read
'This is the Christ, the king of the Jews.' Compare Matt.
xxvii. 37 o5to9 eanv 'Irjaov'; 6 ^. rSv 'lovBaiwv. The Gospel
' Q has a conflate reading ; it adds the words ei custodiebant eum.
7—2

100 THE SYRO-LATIN TEXT OF THE GOSPELS,
of Peter (ed, Swete, p. 6) has ovto9 eanv b /3. rov 'lapaijX.
(3) The last line exactly corresponds with the phrase of the
Sinaitic (Mc. xv. 27): ' And-crucified^ there-were with-Him
two robbers.' There is, so far as I know, no other authority
for the passive in Mark^ (comp. Matt, xxvii. 38 (Gk.), Mc. xv.
32 (Sin. Pesh.)).
' There is a slip here in Mrs Lewis' translation ' And -with Him thev crucify
two thieves.'
^ The passive apparently is not the Tatianic reading : Ephrem has ' they cruci
fied -viiih Him two others' (Hill, p. 375 ; so the Arabic, p. 246).

3-
PROPER NAMES AND FORMS OF WORDS,
If the text of Codex Bezae or, to speak generally, the
Syro-Latin text took its rise in a bilingual (i.e. Graeco-
Syrian) Church, we should expect to find that scribes
sometimes Syriacised Greek words and proper names, and
that, probably through the medium of Greek MSS., this
Syriac influence made itself felt in Latin texts.
Such an influence would be due to two causes : (i) the
transcription of Graeco-Syriac bilingual MSS.; (ii) the
famiharity with Syriac forms on the part of bilingual scribes,
and the tendency of such scribes to shape Greek words in a
Syriac mould.
That such an interplay of influences operated in the case
of bilingual MSS. is clear from a study of Codex Bezae itself
We find a tendency to introduce hybrid words on both sides.
Thus in the Latin text appear such words as promeletantes
(npoMeAercoNTec, Lc. xxi. 14), dum aporiarentur (cn too AnopeicGAi
AYTAC, Lc, xxiv. 4); see Scrivener, p. xxxii: in the Greek
such forms as camapitancon (Samaritanorum, Matt, x, 5),
exerec (habetis, Mc. vi. 38); see Scrivener, p, xxx.
What is the evidence as to the influence of Syriac forms .?
Certain Bezan words seem to reproduce the Syriac pre
fixed K*, which 'in vetustioribus codicibus passim vocibus
praeponitur, ubi omittunt recentiores' (Payne Smith, Thes. Syr.,

I02 THE SYRO-LATIN TEXT OF THE GOSPELS,
p. 3). In Matt, xxvii, 55, Lc, xxiii. 55 D has awo thc
AfAAeiAAiAc (afaAiAaiac : d, de galilaea (Matt.), a galilaea
(Lc)). With this form compare the Jerusalem Syriac
Lectionary (p. 329) ^«^'raftiA\r<' = Tedarffiavei (Matt, xxvi
36); also .\ci\r^ = Twy (see Payne Smith, Thes. Syr),
r<'.Ti\\r<' = hS'.tA^ (Noldeke, Syr. Gram., § 51). In
Mc. xi. 8 we read in D aAAoi Ae bctiBaAac (d, frondes). Com
pare Noldeke, Syr. Gram., § 5 1 (comp. § 25) : ' Einem anlauten-
den Consonanten ohne vollen Vocal wird zuweilen ein | mit
Vocal vorgeschlagen. So 1 in 1A.»1 'sechs'... Haufig so bei
griechischen Wortern mit ctt, o-tt wie
(j-^^.^] oder ]j_4r4^ arpareia,
'jj_,_2l£0( und 1; I '^m aTTeipa u. s. w.'
Of these two forms the latter (cctiBaAac) seems due to a
bilingual scribe Syriacizing a Greek word ; the former
(apaAiAaia), since it is found in parallel descriptions of the
' faithful women ' (Matt., Lc), probably reproduces a Tatianic
form. It is natural to consider in this connexion the forms
which the name 'laKapiwrrji; takes in the Syro-Latin texts.
The forms found in D are :
(i) icKApicoA, Lc. xxii. 3.
(2) cKApiwTHc, Matt. X. 4, xxvi. 14, Mc. xiv, 10, 43.
(3) cKApicoG, Mc, iii, 19, Lc. vi. 16, Jn, vi. 71,
(4) Ano KApYOOTOY, Jn. xii. 4, xiii. 2, 26, xiv. 22.
Among other Greek MSS,, G has aKapiwrrfv (Lc. xxii. 3).
The interpretative reading is found (Jn. vi. 71) in t<* (aTro
KapvwTov), in the Ferrar-group (13 aTro aKapvwrov, 69
aTTOKapiwrov, 124 aTro Kapvwrov).
About the forms in the Latin texts two points call for
notice: (i) The first syllable: sometimes the name has the
initial syllable is; thus isscariotha (D, Matt, xxvi. 14),
iscarioth (a, Lc. xxii. 3). More often this syllable is is

PROPER NAMES AND FORMS OF WORDS. r03
wanting; thus scariothes, scariotes (most vg.-MSS, in Matt.
x, 4), cariotha (e, Mc iii. 19), shariotes (k, Mc. xiv, 10),
(2) The interpretative reading: in e (Jn. xiii. 2) we find
the reading: iudae simon a cariotha.
From the Greek and Latin forms we turn to the Syriac,
the same points still claiming our attention: (i) The first
syllable: (i) The Sinaitic has r^A^a»Taj» in every passage
where the name occurs in the Greek text except Matt. xxvi.
14, where there is a lacuna, and Jn. xiv. 22, where the clause
' not Iscariot ' is rendered needless by the introduction of a
distinctive name ' Thomas said to Him.' (ii) The Cure
tonian has r^a^h^SOT^ (Jn. vi. 71), rtfl^^Ojiaoor^ (Lc.
xxii. 3) : in Jn. xiv. 22, like the Sinaitic, it adds the distinctive
name 'Juda Thomas said to Him': in no other place where
the name occurs in the Greek text is this Syriac text
extant. (iii) The Peshitta always has the same form
(i^jjOrfO^as) as the Sinaitic. Thus, when the name is
fully written, the first syllable is r^. This K* is retained
in the Curetonian : it is omitted in the Sinaitic and the
Peshitta. Now in Syriac the evanescence of the first syllable
of the name finds a natural explanation ; for the first syllable
is an K", which would be regarded as the f^ frequently pre
fixed to Syriacised Greek words, as KIAqajjo rC (= axoXij),
r^m\n*r^ (= axvfia), see Noldeke, Gram., §§ 25, 51.
(2) The interpretative reading: such a paraphrastic represen
tation of the name would be likely to arise in Syriac. Thus
raXtXato9 (Lc. xxiii. 6) becomes in the Curetonian (Sin.
wanting) 'from Galilee (is He)'; TajOo-ev9 (Acts ix. 11, cf xxi.
39) becomes in the Peshitta 'who-is from Tarsus the-city';
K/3^Te9 (Acts ii. 11) 'those-from Crete'; raro9 Ae/)/8ato9
(Acts XX. 4) 'Gains who-(was-)from Derbe the-city.' Such
indeed is the interpretation of the name Iscariot preserved by
the Syriac lexicographers : ' ex urbe Scariot nomen ducunt
lexx., sc v^h\^\n J^a.*i.&ju ^ .en r<'-\^CU'i^a>'
(Payne Smith, Thes. Syr., p. 2637), It is found also in the

104 THE SYRO-LATIN TEXT OF THE GOSPELS.
margin of the Harklean' Version (Jn. vi. 71) fSn^ am
<xA^x>o^ir^.o 'he who-(was-)from Carioitu.' As however
'the margin contains various readings taken from Greek
MSS.' (Dr Hort, Introduction, p. 85), this is probably not a
genuine Syriac reading.
We turn to certain termination-forms in the Bezan text.
In Mc. iii. 21 we have a Greek noun with the Syriac plural
termination — rpAMMAreiN. It seems clear that this form be
trays the hand of a scribe familiar with Syriac. The
genealogy in Matt. i. is wanting in the Bezan Greek text.
A large part however of Matthew's genealogy is interpolated
in the Bezan text of Lc. iii. In Lc. iii. we have the form
lAxeiN (true text in Matt. i. 14 ' Kxe'ifi), with which we compare
the form in A (Matt. i. 14) ' k^xew (see Tisch. in loco). In
these two forms (^Yaxelv, 'A-xeiv) we have, as far as the
termination is concerned, a reproduction of the Syriac form
^.lArc' (Sin., Curet., Pesh., Aphraat p. .VwAx) : the Harklean
in Matt. i. 14 has this form once, and the form ^Xi^K'once in
the text and once in the margin, the latter form also appear
ing in the Jerusalem Lectionary (p. 483). The Bezan Latin
has achim in Matt., iachin in Lc. With this we may
perhaps compare Nect)GAAeiN (true text ve<j)daXeifi) in D (Matt.
iv. 15, d neptalim), a form which, so far as I know, elsewhere
is found only in the Lat. vg. MS., Cod. Harleianus, neptha-
lin (Matt. iv. 13). It would seem that here the Bezan scribe
instinctively made the alteration of a single letter that he
might assimilate the termination of the Greek name to the
familiar Syriac plural form^ Compare cyx^n for Sv%e/i
(Acts vii. 16, d sychem) and cAMct)OYpeiN (p. 108).
I pass on to the consideration of certain passages in
Codex Bezae,
^ The Harklean has varying forms of the name, some with the prefixed
syllable (<5a*t<', 00r<'), some without it. The form in the Jerusalem
Lectionary (pp. 313, 323, 342, 555) is r^i^CV»iA\,
^ It is of course a spurious form, the Syriac always being »iiov^.

PROPER NAMES AND FORMS OF WORDS. 105
Matt. xiv. 34 eic reNNHCAp.
The name occurs (besides this passage) in Mc. vi. 53,
Lc V. I. In Mc, D has the same form as in Matt, the
Latin being gennasar (Matt.), gennesar (Mc) ; in Lc we find
reNNHCApeA (d gennesared).
The longer form is that found in all Greek MSS. with, so
far as I know, one single exception, viz, the remarkable Cod,
604 (=700 Gregory), which in Matt, has yevvrjadp. The
Latin authorities vary, (i) In Matt. Old Latin and vg.-MSS,
(except R, genessareth) have gennesar (genesar). (ii) In Mc,
most vg.-MSS. have gennesareth; but bcff^ have the shorter
form, (iii) In Lc. most vg.-MSS. have the longer form.
Of the Old Latins however f, and of the vg.-MSS. gat Z
have the shorter form ; D (' Book of Armagh ') has genitzar.
The form of the name in the Old Syriac (Sin. Cur.') and
in the Peshitta is 'Wm_^ (genesar). The shorter form there
fore is distinctly the Syriac form, and the evidence seems
clearly to point to the conclusion that from the Syriac this
form passed into other authorities, Greek and Latin.
This conclusion receives confirmation from the fact that
the two other Greek books in which alone (so far as I know)
the shorter form Gennesar is found are both reproductions of
Aramaic originals, (a) In i Mace xi. 6j we have the words
TO vBwp rov Tevvffadp (X* Tevvqaat). But this book 'was
written . originally in Hebrew (Aramaic), as may be con
fidently inferred from its grammatical peculiarities, and as is
further confirmed by the testimony of Origen and Jerome'
(Schiirer, Hist, of the fewish People, Div, ii, vol. iii. p. 8,
Eng. Trans.), (b) Again, in Josephus, Bellum fud. (ed. Niese),
iii. 10. 7, 8, we several times find the form Tevvrjadp. Jose
phus in the Preface to this treatise (comp. Contra Ap. i. 9)
tells us that he first wrote and circulated the work in Aramaic.
With the Syriacised name Tevvrjadp we must compare
AAZAp found in H* (Jn. xi. 43), and the Old Latin Lazar found
' Cur. is extant only in Matt. xiv. 34.

I06 THE SYRO-LATIN TEXT OF THE GOSPELS.
in a and in the Bezan Latin (Jn. xi. 14)'. The regular
Syriac form in the Sinaitic and the Peshitta (the Curetonian
being wanting in Lc. xvi., Jn. xi.) is "iv^^A, which is simply
transliterated in Xa^dp, Lazar.
Matt. xxvi. 3 TOY AeroMCNOY kaV<})a.
This is the regular form of the name in D ; in Jn. xi. 49
we find Ki(t)Ac. In Matt. xxvi. 57 we find the form npoc kackJjan.
The Bezan Latin has caiphas (caifas, Jn. xi. 49, Acts iv. 6).
This form has the appearance of being an attempt to
transliterate the regular Syriac form rd^^a.
The shorter form is found in a few Greek MSS., e.g. C
(Lc. iii. 2), 64 (Matt. xxvi. 3). The longer form occurs in
most of the Latin Vulgate MSS. ; the shorter form how
ever is found in some Vulgate and most Old Latin MSS.
(caiphas, caphas, caifas, cayfas, chayphas, chaiphas).
Matt, xxvii. 46,
HAei HAei AaMA ZACf)GAN6l . TOYT eCTIN
Ge MOY Ge moy • 'nati Me eNKATcAinec,
We at once compare the parallel passage in Mc. xv. 34 :
HAei HAei Aama ZA(})GANei
0 ecTiN MeGepMHNeYOMeNON
0 Gc MOY 0 Gc MOY eic ti coniAicac Me.
Here there are presented two points of great interest :
(i) the form ^a^davei both in Matt, and Mc. in place of the
true text aafBaxdavei; (2) the interpretation wveiBiad<; fie
(Mc.) in place of iyKareXnre<; fie.
It is obvious that an explanation of the form ^a^davei
1 Mi- Rendel Harris, A Study of Cod. Bezae, p. 183, draws attention to the
reading of ad. This Lazar in d is a genuine Old Latin reading, like many Old
Latin readings, of Syriac origin, not smoothed away by assimilation to the Greek.
Compare Dr Hort, Introduction, p. 82: 'Here and there the assimilation has
accidentally been incomplete, and the scattered discrepant readings thus left are the
only direct Old Latin evidence for the Greek text of the New Testament which
the bilingual MSS. supply.'

PROPER NAMES AND FORMS OF WORDS. 107
which solves the problem involved in the interpretation wvei-
Biad<; fie will thereby be strongly confirmed. The points are :
(i) The Bezan ^a^davei transliterated into Syriac would
be .xii\&^\. I suggest then provisionally that this was
a Syriac reading or a Syriac gloss in Matt, and Mc.
(2) How did this Syriac reading arise .'' The original
Hebrew word in Ps. xxii. 2 (of the Aramaic equivalent of which
aa^axdavei is the transliteration) is "'iDlTy. There is no Syriac
root corresponding to the Hebrew root ^T^. Hence if it were
wished to represent the original Hebrew word, it would be
natural to use some actual Syriac word, roughly correspond
ing to it both in form and in meaning: hence .xs&v^^t.
(3) The Peal of the Syriac verb .^^\ is used in the Jeru
salem Version as an equivalent of ifi^pifidadai (Matt. ix. 30,
Mc. i. 43), of iTTirifidv (Matt. xvii. 18, xx. 31, Mc. viii. 30,
Lc. iv. 35, ix. 42, xvii. 3, xxiii. 40), the Ethpeel of dyavaKreiv
(Mc. X. 41). In the Peshitta the verb is only found once and
then in the Ethpeel to translate ifi^pifidaOai (Mc. xiv. 5),
Thus the Bezan wveiBiad<; fie would be a very natural equiva
lent of jjiva^ti.
(4) This Syriasm has spread widely in the Latin texts :
(i) As to the transliteration, we find (a) in Matt, h
zapthani, ff^ sapthani, J" zeptani, bMTKV zaptani, a zahthani,
T zabthani ; (b) in Mc. ff^ sapthani, KV zaptani, T sabtani,
k zaphani, i* izpthani.
(ii) As to the interpretation (Mc), c has exprobrasti me,
i me in opprobrium dedisti.
John i. 6. HN ONOMA AYTCO ¦ ICOANNHN.
'ImdvvTfv reproduces the Syriac ^XmCU. In Acts iii. 4,
Codd. DE have cyn (tco, E) icoanhn.
1 The difficulty of this explanation lies in the fact that the verb -^*^ \ 's
followed by the preposition .s ; compare t\'!11 followed by ?J? (Prov. xix. 3), by
DV (2 Chron. xxvi. 19). The abnormal construction with the suffix may however
be justified on the ground that the word was meant to be a rough transliteration of
¦"jnaTV-

I08 THE SYRO-LATIN TEXT OF THE GOSPELS.
John xi. 54.
aAAa AHHAGeN eic thn X'^P'''
CAM(l)OYpeiN errYC thc epHMOY eic e4)pAiM AeroMeNHN
noAiN.
The Bezan Latin has in regionem sapfurim. Perhaps the
only certain thing which can be said as to this puzzling
interpolation is that the termination -eiv is a clear sign of
Syriac influence (comp. p. 104).
The problem is a tempting one and invites suggestions
for its solution. I venture to hazard the following, (i) As
regards the first part of the interpolated word (aafi-), I
adopt Mr Rendel Harris' suggestion that it is a corruption of
cfa:az..i (whose-name). Mr Harris (A Study of Cod. Bezae,
p. 184) adduces an argument in support of this theory in the
word Samgriazim in Ephrem's Commentary on Tatian (ed.
Moesinger, p. 142) — ' Patres nostri in hoc monte adoraverunt.
Haec de Jacob et filiis ejus dixit, quia in Monte Sichem aut
in Bethel aut in Monte Samgriazim adorarunt' The name
here seems to be generated by a misunderstanding of the
Syriac words ' whose name is Gerizim.' (2) Is the latter
part of the interpolated word (-(f>ovpeiv) a corruption of ^ios ?
For the confusion between A and •= see e.g, the Syriac
version of Clem, Rom. xvi, where rr^wiftr, (glory) is written
instead of r^i^ai. (= KdXXo<;). The word t^ias or iaa
has the meaning of barren, waste : it is the equivalent of
Xepao^ several times in the Syriac Hexapla (for the words
of this root see Payne Smith, Thes. Syr., and compare
Buxtorf, Lex. Chald., p. 275 f, Levy, Chald. Worterbuch,
p. 86 f ). I suggest then that it is possible that we have in
the word aafi<f)ovpeiv a relic of a Syriac gloss — ' whose name
was Burin (desert places)' — a gloss on the words 'into the
country near to the wilderness,' partially corresponding in
form to the clause which follows, ' into a town which was
called Ephraim.'

PROPER NAMES AND FORMS OF WORDS, IO9
Luke xxiv. 13. onomati oyAammaoyc.
The Old Latin MSS. eff'' read here ammaus et cleopas
(ff^ -phas), b has cleofas et ammaus. These two names appear
in Ambr., Expos, in Luc, vii. 132, 173 : ' Hunc ignem in se
etiam Amaon et Cleophas a Domino missum esse testantur...
se Ammaoni et Cleophae seorsum iam uespere demonstra-
uerat' (see Tisch. on v. 18). The combination of names pro
bably arose in v. 18.
If we turn the reading of b into Syriac we have the words
oooK^so^o r^^cxiis. A slight blunder in writing the
second of these two names (and-Emmaus) would generate
the word ooor^ialo (oulemmaus). As to this corruption of
.a*, into A we may note that (i) it is very natural in itself;
so in an Ignatian letter (Hero 6, see Bp Lightfoot, Ignatius, i.,
p. 87) we have in the Syriac version t^."u.2a,AA» (disciple)
written instead of r<'ii*7iN,A< (= o'lKJirrfpiov): (ii) the corrup
tion would be suggested to a scribe by the initial letters of
thc preceding word (-^o, -Vo).
When in the above clause the obviously intrusive rcL^cxxia
(Cleopas) is ej'ected, there then remains the word, the cor
ruption in which has j'ust been explained, ooordsolo. This
transliterated into Greek gives the Bezan ovXafifiaovs-
It must be noticed that we have the same name in the
LXX, of Gen, xxviii. 19 /cat OvXafifiain (OvXafifiaoix; DE*)
•^v ovofia r^ TToXei rb TTpbrepov. It seems to me most probable
that the remembrance of the LXX. name facilitated, if it did
not suggest, the corruption of Ouemmaus into Oulemmaus.
It is however possible that the Bezan reading is due simply
to a reminiscence of the LXX. reading in Gen. I.e.
Mark v, 41, Aerei ayth pABBi • GaBita koymi.
The true text is Xeyei avrrj TaXeidd Kovfi. The Sinaitic, as
well as the Curetonian, is unfortunately wanting here. What
of the name in the Bezan text (dajSird) and in several
Old Latin MSS.— c thabita, aff^g'l tabitha, bi thabitha^t
1 This reading (tabitha, thabitha, tabita) is found in many vg.-MSS. (see Bp
J, Wordsworth's note in loco).

no THE SYRO-LATIN TEXT OF THE GOSPELS.
It appears that the Aramaic word in Mc. — TaXetdd (=Tb
Kopdaiov) — is changed into the name found in Acts ix. 40
(Ta^eidd, dvdarrfdi). How easy the substitution would be in
a Syriac text is clear when we place the passages side by
side: Mc. V. 41 .i^Ckfl r^^A^}^ (TaXet^a Kovfi).
Acts ix. 40 .0930,13 K'^Aa!^^ (Ta^eidd, dvdarrjdi).
The Old Latin MS. e has in this place the remarkable
reading : ' et dixit ei tabea acultha quod est interpretatum
puella puella tibi dico exsurge.' There are here two points
to be considered, (i) There is the reiterated puella puella.
The only other authority for the repetition of the word is
Aphraat (p. cafloo) : ' And our Lord too at His first coming
revived three dead persons... And by two words He raised
each one of them. For when He revived the widow's son.
He called him twice, since He said to him Young man, young
man, arise^... And again the daughter of the ruler of the
Synagogue He called twice, since He said to her Maiden,
maiden, arise! Further, the strange Bezan reading pa^^l
da^ird finds its explanation in the reduplication (' maiden,
maiden ') coupled with the corruption of Ta/itha into Tai^itha.
But the reading pa^0i seems to bear in itself evidence of
having arisen in, or in connexion with, a Syriac text. For,
while it is hard to see how the 00 of pa00i could arise from
the Greek da0idd (ra0idd), the Syriac equivalent of pa00'i is
jai, and .x=3i would be generated as a corruption of the
first two syllables of the word rCAxii^ (Tabitha),
(2) What of the words Tabea acultha f The word acultha
bears upon its face signs of a Syriac origin. It can hardly
be doubted that it is a relic of the word K'AucvAr^^a
(macultha = food). There is evidence that the -word food had
1 In Lc. vii. 14 D has nganickb NBANlCKe; the Old Latin aff^ have
adolescens adolescens. The repetition in Mc. v. 41, Lc. vii. 14 is doubtless due to
assimilation to such passages as Lc. a. 41 l^dpBa JidpBa), xxii. 31 (Si/jwi/ Si/xuj/).
A similar repetition is found in Jn. xi. 43 Adt^ape Ad^pe (C^ aeth'''"" ; see Tisch.) —
a reading implied by Aphraat's words, though (in the context) he has 'Lazarus,
come forth.'

PROPER NAMES AND FORMS OF WORDS. Ill
a place in an Old Syriac version of the Lord's command to
the parents, for Ephrem (Hill, p. 344) gives that command
in this form : 'And He commanded to give her food to eatV
The reading of e ('tabea acultha quod est interpretatum
puella puella tibi dico exsurge ') is important on two grounds :
(i) we have here in an Old Latin MS. what is clearly a Syriac
word ; (2) the text of e must be due to a scribe to whom the
meaning of the Syriac word macultha was not known; it
would seem therefore that this Old Latin text did not spring
up on Syrian soil,
¦• I am conscious that I am on less secure ground when I give expression to my
suspicion that a further corruption lurks in the reading of e. I suspect that (i) in
some Old Syriac text our Lord's words to the parents were assimilated to His
command to the disciples (Matt. xiv. i6) ^__OOVJr^ .  OCoA 0.303
¦\«\^»^«^\ (give to-them yourselves to-eat), and that thus there arose the
reading in Mc. v. 43 'Give to-her food to-eat'; for this introduction of the
oratio recta comp. Lc. viii. 29 (Trapr^yyeXXev ydp r<} Tvei/ian r^ dKaBdfrrip i^eXBeiv),
where D has eAereN pAp-.-eleAGe (so e); Jn. v. 15 (in 'Ir/aovs iariv b ¦jvoi.iittas
airhv iyif/), where many Syro-Latin authorities (D 1-118, Cur., aeff^ Iq, me. arm.)
read fii ; (ii) when this direct command was first compared with, then substituted
for, the direct command 'Maiden, arise,' the word m\-^m (give her) — for
the imperative of this verb ' passim cum praep. J conjunctim scriptum est '
(Payne Smith, Thes. Syr., p. 1565) — was conflated with Tabitha, and hence the
corrupt tabea of e.

4-
GRAMMATICAL POINTS.
In this last chapter I propose to consider some gram
matical peculiarities of the text of Codex Bezae. Many of
these have been already incidentally noticed. It will be well
however to gain some connected view of them. The points
to be discussed fall under three heads: (i) the definite
article ; (2) prepositions ; (3) verbal constructions.
(i) The definite article.
The Syriac texts of the New Testament not unfrequently
use the pronouns ' this,' ' that ' to represent the Greek definite
article, for which indeed Syriac supplied no other distinctive
equivalent. Thus in the Sinaitic Syriac of St Matthew we
find, e.g. ii. 7 'those Magi' (tou? fidyovi); viii. 31 'those
demons ' (ot Be Baifiove<;); ix. 22 ' that woman ' (rj yvvrj) ; xiv. 19
' these five loaves ' (rov'i TTevre dprovi) ; xv. 32 ' this multitude '
(rbv oxXov) ; xv. 36 ' these seven loaves ' (toi)9 eTTTa dprovi) ;
xvi. 10 ' these seven loaves ' (tov<s erTrd dprovs) ; xx. 24 ' those
two brethren ' (twv Bvo dBeX^wv) ; xxi. 20 ' that figtree ' (ij
avKTJ) ; xxv. 9 ' these wise ones ' (at (ppbvifioi) ; xxvi. 72 ' this
man ' (rbv dvdpwTTov). ' Sehr zahlreich,' writes Baethgen
(p. 20), 'sind die Beispiele fiir den Gebrauch des Demon-
strativpronomens, fiir welches kein griechisches oiiT09 u. dgl.
vorausgesetzt werden darf '
We have seen abundant reason for the belief that as
similation to a Syriac text is an influence constantly at work
in the Bezan text. We are not therefore surprised to find an

GRAMMATICAL POINTS. II 3
insertion of outo9, iKeivo<; in the Bezan text similar to the
insertion of ' this,' ' that ' in the Syriac texts. Thus Matt.
xiii. 27 TOY oiKoAecnoTOY ckbinoy; xv. 24 ta npoBATA tayta (Cur.
' those Hocks'); xv. 32 ton oxAon toyton (Sin. Cur. Pesh. lat-vt.) ;
Lc. xvii. 17 oytoi acka (Sin. Cur. wanting^; lat-vt); xvii. 22
T05N HMepCON TOYTCON ; Mc V. 36 TOYTON TON AofON (Sin. Cur.
wanting; fff^iq); viii. 2 toy oxAoy toytoy (Sin. Pesh.; af
(huic), q (hanc), b c g^ ff^ i (istam)) ; x. 22 toytoo too Aopco
(Sin. Pesh.; Ferrar-group, 2^ abcfff"kq)l
Taken by themselves these Bezan readings do not decide
between the Syriac and the Latin texts as rival claimants for
the honour of having influenced the Bezan text. For the
necessity of inserting 'this,' 'that,' when it was desired to
represent the Greek definite article, is common to the Syriac
and Latin versions. The evidence of these readings however
may be fairly claimed as corroborating the theory of Syriac
influence, if that theory is established on other grounds.
(2) Prepositions.
In Syriac a preposition is commonly repeated before a
second noun in sentences where in Greek a single occurrence
of the preposition would have sufficed. Thus in Lc. x. 1 3 (iv
Tvpw Kai XiSwvi) Sin. has ' in-Tyre and-?w-Sidon ' ; in Matt.
iv. 25 (aTro T7J9 raXt\at'a9 /cat AeKaTroXew<; Kai lepoaoXvfiwv
' Pesh. 'Were not ten these who- were-cleansed ? '
^ In Jn. xvii. 3 D (Greek and Latin) has IC toyton ton kocmon, in hunc
mundum. A similar insertion of ovros, hie, is found in w. 11, 13, 15, 16, 18,
25 in D (Gr. Lat.) and in many Latin texts. Mr Rendel Harris (A Study of
Codex Bezae, p. 66) notices that the insertion of 'this' before 'world' is found in
the Bezan Latin (not Greek) of Jn. viii. 26, xiv. 22, 30, xvi. 21. The 'this' is
not found in Sin. (which is extant in the whole series of passages except xvi. 21,
xvii. 13, 16, 18). On the other hand two points should be remembered : (i) the
phrase 'this world' is so frequent in St John (viii. 23, ix. 39, xi. 9, xii. 25, 31,
xiii. 1 , xvi. 1 1 , xviii. 36) that the prefixing of ' this ' to ' world ' in other passages is a
matter of obvious assimilation rather than of idiom ; (2) of the passages in which
D adds 'this' before 'world' Cur. is extant only in xiv. 22. Further knowledge
of early Syriac texts might well alter the balance of evidence.
C. 8

114 THE SYRO-LATIN TEXT OF THE GOSPELS.
/cat 'lovBaia'i) it has ' from Galilee and-from the-ten cities
and-from Jerusalem and-from Judaea.'
We turn to Codex Bezae. In Matt. xiv. 9 D has Aia toyc
opKOYc kai Aia toyc cYNANAKeiMeNOYc. The preposition (AA^)
is repeated here in the Old Syriac (Sin. Cur.), also in many
Latin MSS.; so in the parallel, Mc. vi. 26. Lc. ii. 34 eic
nrcociN KAI eic anactacin. The Syriac (Sin. Pesh. ; Cur.
wanting) has ' for-the-fall and-_/br-the-rising': so also cff^gM.
Lc. ii, 52 nApA Geo kai nApA ANGpconoic. The Old Syriac (Sin.
Cur.) has ' with (A\oA) God and-with men.' No other
authority is quoted for the repetition of the preposition.
Mc. vi. 36 eic toyc erricTA ArpoYC kai eic tac kcomac (true text
et9 rov'i kvkXw dypov<; Kai Koifia^). The Peshitta has ' To-
the-fields that-are-near and-^o-the-villages.' No other au
thority is quoted for the repeated preposition. The Sinaitic
has a shortened text here : ' To-the-villages these that-are-
near.' Mc. viii. 31 Yno tcon npecBYrepcoN kai aho tcon
ApxiepecoN. The Syriac (Sin. Pesh.) repeats ^ (from) — 'from
the-elders and-from the-chief-of the-priests and-from the-
scribes.' Similarly many Latin texts repeat the preposi
tion a. Mc. xiv. 43 nApA tcon ApxiepecoN kai auo tcon rpAMMAreco.
D has here the alliance of several Latin texts (a... a). The
A>o\ ^sa (from with) of the Syriac (Sin. Pesh.) is not
repeated. Thus in five out of these six cases of the repeated
preposition D has the company of the Syriac; in two of
them D and the Syriac appear to stand alone.
One or two passages of the Bezan text where the
preposition e'Trt is used are worthy of notice. One of the
most striking of these — Lc. ix, 16 — has been already dis
cussed (p. 36 f ). Matt. xiii. 14 kai roTe nAnpcoGnceTAi [true
text dvaTTXrfpovrai] en aytoic. The Old Syriac (Sin. ; om. Cur.)
has ..^ocoAs^ (upon-them). The same reading appears in
M* and (alone apparently among Latin texts) k (super eos).
Lc. i. 21 eGAYMAZON eni (true text iv) tco xpoNizeiN ayton. The
Peshitta (Sin. Cur. wanting) has 'Wondering were-they at

GRAMMATICAL POINTS. II 5
(lit. Upon, Aj^) his-delay.' No other authority is quoted for
this reading. Jn. xi. 6 eweiNeN o ihc eni tco ronco (true text
iv w ^v TOTTw). The Old Syriac (Sin. ; Cur, wanting) has : ' He-
abode Upon (A^) His-place.' No other authority is quoted
for this reading: the Bezan Latin has in loco. Matt. xiv. 14
ecnAArXNicGH nepi aytcon (true text err avroh). The Syriac
(Cur. Pesh. ; Sin. wanting) has the obvious rendering ' He-
pitied upon-them (^^^oeaJL^).' No other authority is quoted
for the strange TrepL of D. Is not the explanation that the
Bezan scribe had the Syriac Aai. in his mind, and retranslated
it by TTepi, of which it is the regular equivalent .''
Two remarkable readings in Codex Bezae may be
mentioned here. (i) Jn. vii. 39 oyuco pAp hn to una apion
en AYTOIC. Compare Lc. ii. 25 /cat TTvevfia ¦^v dyiov iir avrbv.
The Old Latin f (in eis) and the Gothic are the only other
authorities given for this insertion. (ii) Lc. xi. 2 afiacGhtco
(true text to) onoma coy e(|) hmac. No other authority is
quoted for this addition. When the fondness of the Syriac
for adding to a verb a preposition with a pronominal suffix
(see Baethgen, Evangelienfrag, p. 21) and for the use of the
preposition A^ (upon) in various connexions is remembered',
the suspicion that these readings arose in a Syriac text does
not seem unreasonable.
(3) Constructions of the verb.
The participle (most commonly the aorist participle) is in
many passages of the Bezan text resolved into the indicative
or the imperative, as the case may be, followed by /cat.
Thus Matt, iv, 3 (D, d) :
kai npocHAGeN aytco o nipAzcoN Et accessit ad eum qui temptabat
KAI eirreN aytco. et dixit ei.
^ See above, pp. 36, 88. Thus e.g. in Matt. v. 7 i\ei]B-<]<jovrax = ' upon-them
shall-there-be mercies ' (Sin. Cur. Pesh.) ; in Matt. xii. 22 Sai.juivi.ibp.evov = ' a man
upon-whom (rT"^'*- s) there was a-devil' (Cur.), so 'ixtnv Saip.iivia
(Lc. viii. 27) in Sin. Cur,
8—2

Il6 THE SYRO-LATIN TEXT OF THE GOSPELS.
The true text is /cat TrpoaeXdwv b TTeipd^wv elrTev avrw. The
Old Syriac (Sin. Cur.) is : ' And there-drew-near to-Him the-
tempter (Cur, he that-tempteth) and-said to-Him.' Similar
passages in the Bezan text are Matt. v. 13, ix. 28, xiii. i, 4,
48, xvii. 7, XX. 6, 30, xxi. 6, xxv, 25, xxvi. 51, xxvii. 58;
John vi. II, ix. 35, xi. 17, xii. 36; Lc. viii. 27, xix. 35;
Mc. ii. 16, iv. 36, viii. 10, x. 22, xii. 20, xiv. 22.
As an instance of the corresponding resolution of the
participle in an imperative sentence Lc. xxii. 32 may be
cited (D, d) :
CY Ae emcTpeYON kai tu autem conuertere et
CTHpiloN TOYC AAeA(()0YC COY- confirma fratres tuos.
The Old Syriac (Sin. Cur.) is : ' And-also thou in-time be-
converted^ and- strengthen thy-brethren.' Similar instances
will be found in Matt, xxviii. 19, Lc. v. 14, 24, xv. 23
(cohortative), xix. 5, Mc. v, 23.
How are we to explain this phenomenon thus frequently
recurring in the Bezan text .'' It is not likely to have arisen
in the Greek, unaffected by any foreign influence.
Two arguments which make for assimilation in this
matter to a Syriac text, or at least a Syriac idiom, and
against assimilation to the Latin, must be considered.
(i) It will appear that, while such a resolution of the
aorist participle as we find in the Bezan text is not essentially
characteristic of the Latin, it is characteristic of the Syriac
translations of the New Testament, so that a scribe ac
customed to Syriac idiom would be likely, apart from definite
reference to any Syriac text, to introduce such a type of
phrase into the Greek text. It will be best to give a
concrete example of the mode of treatment of the aorist
participle in the Syriac and in the Latin texts. Any
historical passage will serve the purpose. The following
passage (Matt. ii. 7-12) is taken at random.
The Old Syriac (Sin. Cur.) runs thus: 'Then Herod
' Sin. and Cur. use different verbs.

GRAMMATICAL POINTS. II7
privily called those Magi and-asking was them (KaXeaa<s...
rjKpi^waev) that- he-might-know at-what time there-appeared
to-them the-star (roO (paivofievov darepo<s), and-he-sent them
to-Bethlehem, and-he-said to-them (/cat TTefi-^a<;...elTTev), Go
enquire (TTopevdevre<: e^erdaare) about-Him about the-child
(diligently. Cur.), and-when ye-have-found-Him (irTav Be
evprjTe) come shew-me (diTayyeiXare) that-I-may-go, also I,
worship Him (bTTW^...eXdwv TTpoaKvvijaw avrw). They how
ever when they-had-received the-command of-the-king (from
the-king. Cur.) went (ol Be dKovaavre^ rov ^aaiXew; irTopev-
drjaav), and-there-appeared to-them the-star that-one which-
they-had-seen in-the-east ; going was-it before-them until it-
came stood (and-stood, Cur.; iXdwv iarddrj) (at) the-place
(over, Cur.) where there-was the-child. They however when
they-saw-it (even) the-star, with-j"oy great rejoiced (iBbvre<;...
exdprjaav) ; and-they-entered the-house ««if-they-saw-Him
(eXd6vre<; eh...eiBov) (even) the child with Mary His-mother,
and-they-fell-down and-worshipped (Cur,, worshipped) Him
(rTeaovre^i TrpoaeKvvrjaav avrw). And-they-opened their-treasures
and-they-offered (dvoi^avre<; . . .rrpoa-qveyKav) to-Him an-offering,
gold and-myrrh and-frankincense. And-it-appeared to-them in-
a-vision that they should not return to-him Herod (to Herod,
Cur.) and-they by-a-way another went (xprffianadevreii...
dvexdipTiaav) to-their-place.'
From the Old Syriac we turn to the Latin version of this
passage. I transcribe the text of the Old Latin Codex
Brixianus (f), adding the variants, as far as the rendering of
the Greek participle is concerned, in d g k q and the vulgate.
The passage is as follows : ' Turn herodes occulte uocans
magos diligenter exquisiuit ab eis (g [i.e. gj k q, uocauit... et ex-
quisiuit; vg. uocatis magis) tempus quando apparuit eis stella
(d, apparentis stellae), et misit eos in bethleem dicens (k, etcum
mitteret... dixit; dvg., et mittens .. .dixit) : euntes requirite (d,
euntes interrogate; gq, ite...(et) interrogate; k, ite et quaerite;
vg., ite et interrogate) diligenter de puero, et cum inueneritis
renuntiate mihi, ut et ego ueniens adorem eum (g q, ut (et) ego
ueniam et adorem ; k, ueniens adirem) : qui cum audissent

Il8 THE SYRO-LATIN TEXT OF THE GOSPELS.
regem abierunt (d, audientes... abierunt; k, ubi audierunt
...abierunt), et ecce stella quam uiderant in orientem ante-
cedebat eos usque dum uenit et stetit (vg. ueniens staret) supra
ubi erat puer. Videntes autem stellam gauisi sunt (k, cum
uidissent...gauisi sunt) gaudio magno ualde: et intrantes in
domum uiderunt (d, uenientes...uiderunt; gvg., intrantes...
inuenerunt; k, cum introissent...uiderent) puerum cum maria
matre eius ; et procidentes adorauerunt (d, cadentes adorauerunt;
k, prostrati adorauerunt) eum, et apertis thensauris suis
optulerunt (d, aperientes .. .obtulerunt ; k, aperunt...et optulerunt)
ei munera, aurum thus et murram. Et admoniti per somnium
ne redirent ad herodem per aliam uiam reuersi sunt (d k,
moniti...reuersi sunt; vg., responso accepto... reuersi sunt) in
regionejn suam.'
A review of the Syriac version shews that in the large
majority of cases it resolves the Greek participle into a past
tense with or without a7zd appended. Sometimes, though
comparatively seldom, it employs a circumstantial clause.
On the other hand the Latin has several expedients ; it uses
in such cases a circumstantial clause, an ablative absolute,
a participle of a passive or of a deponent verb ; even when the
aorist participle describes what is clearly past, it employs
as its equivalent, and that very frequently, the present parti
ciple of the active verb.
Thus the resolution of the participle, which so often meets
us in the Bezan text, is essentially characteristic of the
Syriac versions.
(ii) The second point which claims notice is connected
with a series of passages in the Bezan Greek text where,
though the resolution of the participle has not taken place, a
/cat is prefixed to the following verb. Thus Mc. ii. i (D, d) :
KAI eiceAGcoN haAin eic KA<t)Ap- et iterum intrauit in cafarnaum
NADYM
Al H/wepojN KAI hkoycGh OTI eN post dies et auditum esf quod in
oiKco 6CTIN. domo esset.
Similar instances are Mc. v. 27, vii. 25, xi. 2, xiv. 63, xv. 46,

GRAMMATICAL POINTS. II9
xvi. II, 15 (an imperative sentence ; the Bezan Latin is want
ing after xvi. 6).
But the following cases of this incomplete resolution of
the participle in the Bezan Greek are, as it appears to me, of
decisive significance.
I. Matt, xxvii. 33 f
eAGONTec.KAi bAwkan. uenientes...et dederunt.
The Old Syriac (Sin. ; Cur. wanting) and the Peshitta have :
' They-came... and-^ey-^ave!
2. John xii. 3.
AamPani AeiTpAN...KAi hAeiye accipiens libram...«^ unxit.
The true text has Xa0ovaa...r]Xei-^ev. The Sinaitic (Cur.
wanting) and the Peshitta agree as to the construction : ' 5^^-
took...and-she-poured-it on the-head of-Jesus while reclining
(Pesh. omits this clause), ««i^-she-anointed His-feetV
3. Luke viii. 8.
KAI (|)YeN KAI enoiHceN KAprroN. et cum germinasset fecit fructum.
The Syriac versions (Sin, Cur. Pesh.) have : ' And-sprang-
up and-gave (Pesh. -made) fruit.'
4. Luke ix. 6.
e2epxo/weNOi...KAi hpxonto exeuntes...tra.n5\h2L-a.t.
The true text has i^epxbfievoi Be Biijpxovro. The Syriac
versions differ. The Sinaitic has : ' And-when they had-gone-
out.,.going-about were-they.' The Curetonian has: 'And-
when they-went-out.,.and (= then)-going-about (were they).'

1 This is one of the noteworthy cases of harmonizing in Sin. The Arabic
Tatian (Hill, p. 197) has the same combination: 'Now Mary took a case of
ointment of the best nard...and opened it, and poured it upon the head of Jesus,
as He reclined at meat (Matt. xxvi. 7, cf. Mc. xiv. 3) ; and anointed His feet.' So
Cod. Fuldensis : ' Habens alabastrum...et fracto effudit super capud Ihesu recum-
bentis et unxit pedes.'
In the Old Syriac ' Griechisches Praesens historicum ist hin und wieder durch
syrisches Perfect mit oder ohne KIH wiedergegeben ' (Baethgen, p. 27). Hence
conversely the Bezan Xa/ipdvei would be a natural retranslation of the Syriac
'she took.'

120 THE SYRO-LATIN TEXT OF THE GOSPELS.
The Peshitta gives a type of sentence to which the Bezan text
seems to be assimilated: ' And-they-weftt-out...and-going-
about were-they.'
5. Mark vi. 48.
KAI eiAcoN AYTOYC... KAI. ..epxGTAi et uidetis eos. ..^A.. uenit.
The Sinaitic has: 'And-when He-saw them... He-came.'
The Peshitta again has a reading which seems to lie behind
the Bezan text: 'And-He-i-«w them... «W. ..He-came.'
6. Mark viii. 10.
KAI AYTOC ANeBH eic TO nAoiON... et ipse ascendens in nauem...^/
KAI hAGbn. uenit.
The true text is /cat evdv<i ifi0d<; [avrbs:] eh rb ttXoiov...
^Xdev. The Sinaitic has : ' A.nd-}!ie-went-up sat Him in-
the-boat...««^-they-came.' The Peshitta has: 'And-He-
went-up immediately into-the-boat...i3:?za?-He-came.'
7. Mark x. 22.
o Ae ecTYfNAceN em toytco tco ad ille contristatus in hoc uerbo
Aofco KAI AnnAGeN. ^^ abut.
The true text is 0 Be arvyvdaa<i iirl rw Xbyw dTrrjXdev.
The Sinaitic has : ' And-it-was-sad to-him about (lit. upon)
this word i2;«^-he-went-away.' The Peshitta has : ' He how
ever was-saddened at this word and-he-went-away!
In these cases two questions suggest themselves : (i) If
the resolution of the participial construction of the original
Greek is indigenous in the Latin, how are we to account for
the cases (i, 2, 5, 6, 7) of partial resolution in the Latin .'
(ii) Again, how are we to account for the cases (3, 4)
where the Latin does not suggest the resolution at all .¦'
These cases seem to point to assimilation to the Syriac
idiom, which in such sentences regularly prefixed an and to
what in the Greek is the main verb of the sentence.
Three other classes of passages in the Bezan text, akin
to those j'ust considered, claim notice here.

GRAMMATICAL POINTS. 121
(a) There are passages where we find a phenomenon the
converse of that resolution of the participle which, as we
have seen, is characteristic of the Syriac, and which is
common in the Bezan text.
Thus in Lc. ix. 7 (fjKovaev Be 'Hp&)S7y9.../cat BirjTrbpei) D
has: AKOYCAc Ae HpcoAHc.HnopeiTO (audiens autem herodes...
confundebatur). Similar readings are found in Codex Bezae
in Mc. iv, 38, v, 23, vi. 7, 13, xv. 24 (ctaypcocantcc ayton
AiAMepizoNTAi, the Latin being : cruci adfixerunt eum diuis-
erunt). These Bezan phrases may of course be explained as
simple eccentricities of the Bezan scribe. On the other hand
they would naturally arise if the Bezan scribe had before
him, or had in his mind, the Syriac phrase (e.g. ' And Herod
heard. ..and he marvelled'), and retranslated it by the Greek
words of which it would be the natural rendering.
(b) There are passages in the Bezan text in which a
participle is resolved into an indicative, passages, that is, in
which a type of phrase characteristic of the Syriac is in
troduced. Thus in Matt, ix. 29 (tot6 Tjyjraro... Xeywv) D (with i) has
Tore HVfATO...KAi cincN. The Syriac (Sin. Pesh.; Cur. wanting)
has: 'Then He-touched... <3:;2(^-said.' The Latin MSS. (except
d h, which read tunc tetigit...et dixit) have tunc tetigit... dicens.
In Matt, xxvii. 49 (iBwfiev el epxerat, 'HA,eta9 awawv avrbv)
D (with 1-209) has ei epxcTAi hAciac kai ccocei ayton. The Old
Syriac (Sin. ; Cur. wanting) has : ' If coming (is) Elias and-
saving Him.' The Old Latin authorities vary: d has et
liuerat; abcff^hlq et liber abit; ff^g^ liber ans; g^ liberare.
In Lc, xxiv. 5 (ifi4>b0wv Be yevofievwv avrwv Kai KXivovawv ra
TTpbcrwTTa eh Trjv yrjv eirrav TTpb<; avrd';) D has eN<()oBoi Ae
rCNOMCNAI eKAeiNAN TA npOCCOHA CIC THN fHN 01 A6 eiUAN npOC AYTAC.
The Old Syriac (Sin. Cur.) has: 'And-they-feared and-bowed
their-heads and-looking were-they on-the-earth from their-
fear ; saying to-them (were) those men.' The Peshitta has :
' And-they-were in-fear and-bowed their-faces on-the-earth
and-saying (were they) to-them.' The Latin texts vary, f vg
rendering the genitive absolute by a circumstantial clause

122 THE SYRO-LATIN TEXT OF THE GOSPELS.
introduced by cum, B having an ablative absolute, while c r
coincide as to the construction with the Bezan Latin (in
timore autem factae inclinauerunt uultos suos in terra, ad
illi dixerunt), which its opening words (in timore factae)
stamp as a close rendering of the Bezan Greek.
For similar cases see Matt. xvii. 26; Jn. iv. 51, ix. 25, xii. 4;
Lc. ix. 39, xiv. 29, xxiii. 36, xxiv. 44; Mc. x. 16, 35, xiv. 65.
(c) Lastly there are passages in the Bezan text where a
participial construction, Syriac rather than Greek, is intro
duced. Thus in Matt. xvii. 9 (/cat Kara0aiv6vrwv avrwv iK rov
bpovi ivereiXaro avroh) D has: kai KATABAiNONrec eK toy op[oY]c
cNCTeiAATo aytoic (Et descendentes de monte praecepit eis).
The Syriac (Cur. Pesh. ; Sin. wanting) has : ' And-while
descending from the-mountain Jesus was commanding (Pesh.
commanded) them.' The Latin texts adopt the natural
construction — ' et descendentibus illis de monte precepit eis '
(the last word being often omitted). In xxvii. 41 (ol
dpxiepeh ifi7ral^ovre<; fierd rwv ypafifiarewv Kai TTpea0vrepwv
eXeyov) D has : 01 Apxiepeic eNUAizoNTec mcta tcon tpammataicon
KAI (t)ApicAicoN AeroNTec (principes sacerdotum deludentes...
dicebant). The Sinaitic (Cur, wanting) has : ' The-chief-of
the-priests as the-scribes and-the-Pharisees mocking were at-
Him and-reproaching were Him and-saying! The Peshitta
has : ' The-chief-of the-priests mocking were with the-scribes
and-elders and-Pharisees and-saying! In Lc. i. 36 (/cat
IBov '^Xeiad0er...Kal avrrj avveiXrj(j)ev vlbv) D has: kai iAoy
eAeicABeG,,,KAi ayth cynciAhcIjyia yion (et ecc elisabet,.,et ipsa
concepit filium). The Peshitta (Sin, Cur, wanting) has :
'And-behold Elizabeth... also she (is) pregtiant-with (t^ix^^)
a-son.' In Lc. xx. 47 (ot Kareadiovaiv rd<; o'lKia'i twz/ XVP^^
Kai •7Tpo(pdaei fiaKpd TTpoaevxovrai) D has : 01 KATecGoNrec tac
oiKiAC TCON XHpcoN npo^Acei MAKpA npocEYXOMeNOi (qui comedunt
domos uiduarum occasione longa orantes). The Syriac (Sin.
Cur. Pesh.) has ' And-devouring (Pesh. those who-devouring)
the-houses of-widows in-pretence that-lengthening (are they)
their-prayers.' The double participle in the Syriac answers to

GRAMMATICAL POINTS. 1 23
the double participle in the Bezan text, the same construction
being also found in P X 1 22. The omission of and before in
pretence, involving a modification of the sense, characterises
the reading of several Old Latin MSS., which however
vary considerably in the wording of the passage, e.g., s qui
deuorant dom. uid. occasione longa orantes, q qui excusatione
longa orantes deuorant..., cff^il qui fingentes longam oratio-
nem deuorant panes uiduarum, fvg qui deuorant dom. uid.
simulantes longam orationem.
There are in the Bezan text certain other verbal construc
tions — final, temporal, and circumstantial — which yet remain
to be considered.
(i) Matt, xxiii. 15 ina hoihchtai cna npocHAYTON (true text
TTOirjaai eva rrp). The Syriac texts (Sin. Cur. Pesh.) naturally
have »__oaa^>.4«.T (that-ye-may-make). The Latin texts
(except e facere) as naturally ut faciatis.
(ii) Matt. xxv. 10 ecoc Y^'^'roYciN AfopACAi (true text
drTepxofievwv Be avrwv). The Sinaitic has : ' And-while going
(were-they) (^Atr** .t^o)' ; the Peshitta: 'And-while they-
went (J^i\r<^.' The Latin texts have: d cum uadunt; ff^
illae autem dum uadunt; fg^vg dum autem irent; cbff^h (et
ch) dum, eunt; g^ dum irent; q abeuntibus autem illis.
(iii) Matt, xxvii. i, Mc. xiv. 55 ina Ganatwcoycin ayton
(true text ware davarwaai (Matt.), et9 rb davarwaai (Mc)).
The Syriac texts (Sin. Pesh.) naturally have 'that-they-
might-put-Him-to-death,' 'that-they-might-kill-Him' (Mc.
Sin.). The Latin texts as naturally ut eum morti traderent.
It must be noticed however that in some passages the
converse phenomenon is found — Jn. xi. 1 1 aAAa nopeYOWAi toy
e^YUNicAi ayton (d ut excitem eum : true text iva i^vtrviaw
avrbv). Lc. iv. 39 kai ac^hkbn aythn nApAXpHMA cocre anactacan
aythn AiAKONeiN AYTOIC (truc text TTapaxpvjfia Be dvaardaa
BirjKovei avroh). The muddle in the Bezan Latin (ut etiam
continuo surgentem earn ministraret eis) is a significant proof
of retranslation from the Bezan Greek, Lc. v, 6 cocre ta Aiktya

124 THE SYRO-LATIN TEXT OF THE GOSPELS.
pHccecGAi (true text Biepijaaero Be rd BUrva avrwv). These
may be simply the arbitrary emendations of the Bezan scribe.
On the other hand, in regard to Lc. iv. 39, v. 6, the following
characteristic of the Syriac versions should be remembered —
' Griechische Nebensatze mit ina, oti, onoY u. dgl. werden
bisweilen durch ) aufgelost, welches mit dem ) des Zustandes
verwandt ist.... Mt. 15, 31 cocTe...GAYMACAi, ))n 't''^f21t^^)'
(Baethgen, Evangelienfrag, p. 29). If the Bezan scribe had
before him, or in his mind, the Syriac words ' and she arose,'
' a7id their nets broke,' the sense of the passage might suggest
their retranslation by a Greek idiom of which such Syriac
words were frequently the equivalent,
(iv) John vi. 61 coc oyn epNco o mc (true text elBoy; Be b
'lr]aov<;). The Old Syriac (Sin. Cur.) has : ' Jesus however
when He-knew,' the Peshitta: 'Jesus however knew.,. and-
said.' The Latin texts commonly (e.g. f q vg.) have : sciens
autem fesus; but K*, Ferrar-group, a (cognouit), ff^ (cognouit
autem), er (cognouit ergo), coincide with the Peshitta.
(v) Luke xxi. 36 ina KATAlicoGnre eK(J)YreiN---KAi CTHcecGe
(true text 'iva Kariaxvarjre iK<^vyelv...Kal aradrjvai). The
Syriac (Cur. Pesh. ; Sin. wanting) has: 'that-ye-may-be-worthy
(»_ooi.i>.i) to-escape. ..and-that-ye-may-stand (^_o^a.Bh\^a ;
Pesh., and-may- (or and-ye-shall-) stand)! Here the two futures
(' be worthy,' ' stand ') naturally follow the ' that,' The same
connexion is probably intended in the Bezan Greek (tVa
Kara^iw6rjre...Kal arrjaeade). The stabitis however, which is
found in almost all Old Latin MSS., cannot of course be
taken with the preceding ut, and the construction is therefore
abruptly broken off. The order of evolution then would
seem to be (i) the Syriac, (2) the Greek reading of D, (3) the
Old Latin. Tertullian (De Resurr., 22) and r have ' ut.,,stetis '
— the reading of the Syriac texts.
(vi) Mark v. 17 kai hapckaAoyn ayton ina AneAGn (true text
/cat rjp^avTo TrapaKoXeiv avrbv drreXdelv). The Peshitta (Sin.
Cur. wanting) has: ' And-they-began asking from-Him that-
He-would-depart! The Latin texts have ut discederet.
An important point suggested by this passage may be con-

GRAMMATICAL POINTS. 1 25
veniently considered here. In this verse we have uapckaAoyn
(D ; so 225 2^" a) = rjp^avro TTapaKaXeiv (true text). Conversely
in V. 18 Hp5AT0 nApAKAAeiN (D lat-vt-vg) = TrapeKoXei (true
text). So in Mc. vi. 7 AnecreiAeN aytoyc (D 2^*abcff''i) =
fjp^aro avrov<i drroareXXeiv (true text) ; in Mc. xiii. 5 emeN
AYTOIC (D 237 2*''akn arm) = rjp^aro Xeyeiv avroh (true text);
in Mc. xiv. 72 nplATO kAaicin (D lat-vt-vg theb arm) = eTri0aXwv
BKXaiev (true text). In these three latter passages the
Sinaitic Syriac coincides with D and its companions. In
Lc. XV. 28 (true text irapeKdXei avrbv) we have in D the
incomplete expansion HplATO ayton (the word rrapaKaXelv
being omitted at the end of the line) ; here the Bezan Latin
(rogabat eum) significantly differs from the Bezan Greek.
In the Bezan Greek then it would seem that rjp^aro is
lightly added and lightly omitted. This phenomenon is
quite intelligible if the Bezan scribe was accustomed to
Syriac renderings of the New Testament. Compare the usage
of the Curetonian as described by Baethgen (Evangelienfrag.,
p. 28) : ' Bisweilen hat der Ubersetzer, um das griechische
Tempus sinngemass wiedergeben zu konnen, zu Umschrei-
bungen greifen miissen. Lc. viii. 42 AneGNHCKeN prope erat ut
moreretur. Lc. viii. 23 eKiNAYNeYON prope erat ut mergeretur
(navis) [see above, p. 35]. Lc. ix. 33 kai ercNero eN too
AiAXCopizecGAi AYTOYC et cum inciperent discedere [so Sin.J....xxi.
30 OTAN npoBAAcociN hAh cwm incipiunt pullulare et dare \so Sin.J.
xxiv. 29 nApeBiAzoNTO coeperunt rogare [so Sin.].' The only one
of the three passages— Lc. ix. 33, xxi. 30, xxiv, 29 — in which
other authorities coincide with the Syriac reading is xxi. 30,
where e has cum coeperint mittere fructus suos, f cwn incipient
ostendere fructum. On the other hand the Greek ¦tjp^aro is
sometimes not translated in the Syriac. Take the following
examples (beside those noted above) from the Sinaitic
version of St Mark — vi. 55 (rip^avrc-TTepi^epeiv) Sin. has:
'They brought those who were sick, carrying (lit. while
carrying) them on beds'.' viii. 32 (fjp^aro imnfidv avrw)
1 D has HplANTO em tPaBBattoic ({lepeiN hantac- toyc kakcoc
exoNTAC nepie4>epON pAp aytoyc- In the insertion of this last clause D has

126 THE SYRO-LATIN TEXT OF THE GOSPELS.
Sin. has: 'But Simon Cepha, as though sparing' Him, said
to Him.' X. 28 (rjp^aro Xeyeiv b Ti.erpo<s avrw) Sin. has :
'There said to Him Cepha.' xiv. 71 (o Be rjp^aro dva-
defiari^eiv Kai bfivvvai) Sin. has : ' And cursing was he and
swearing.' Thus a lightness in adding and omitting the verb
'to begin,' as though it were a mere auxiliary verb, is
characteristic of the Old Syriac text (comp. p. 45 n,). Hence,
it would appear, it passed into the Bezan, and generally into
the Syro-Latin, text in certain passages.
(vii) Mark vi. 48 eiAcoN aytoyc Bacanizomenoyc kai eAAYNONTAC
(true text 0aa. iv toj eXavveiv). The Latin (e.g. f vg) is able
here exactly to render the Greek articular infinitive — 'in re-
migando.' The Syriac however cannot do this. The Sinaitic
therefore omits the troublesome words — ' He-saw them that-
tormented (were they) from the-fear of-the-waves.' The Pesh
itta has as literal a translation as was possible—' And- He-saw
them that-tormented (were they) while rowing.' This form
of the sentence is apparently adopted by the Bezan scribe,
who retains the two participles of the Syriac but substitutes
/cat for ,1^ (while) ; compare c in mari tribulari in tempestate
remigarites. Note 2^^ 604 iXavvovra<; Kai 0aaavi,^ofievov<;,
a b ff " i q (remigantes et laborantes), apparently a revision of
the Bezan reading in the interests of logical order.
To sum up : We find in the Bezan text a persistent
revision of verbal constructions. Some of the phenomena,
if they stood alone, might be explained on the theory of
the alliance of some Old Latin MSS. — abff^iq. In viii. 32 k has obsecrabat. In
the other passages Sin. appears to stand alone. Pesh. has in each case ' he (they)
began.' ^ This seems to have been the Tatianic reading ; for the Arabic translator has
'as if suffering,' reading JLr^«»j,"| VV*r^ instead of 00T<1m.1 vy»r^ (Sin.,
as-if sparing). I have to thank Mr F. C. Burkitt for this suggestion. The Sinai
tic reading here comes from Matt. xvi. 22 (tXeiis iroi, Kiipie), where Cur. Pesh. (Sin.
viianting) have .m'isn vyi QflM (He-(God)-spares Thee, my Lord). The word
QSjj is a formula of deprecation, as in the Peshitta of Acts x. 14, xi. 8, Rom. iii. 4,
6, 31 &c.

GRAMMATICAL POINTS. 1 27
assimilation to the Latin text. When however all the
evidence is taken into consideration, we are, I believe, led to
the conclusion that the Bezan scribe was a Syriac-speaking
Christian who, in transcribing a Greek copy of the Gospels,
in many passages assimilated the Greek text to a Syriac
idiom with which he was familiar.

SUMMARY OF FACTS AND
CONCLUSIONS,
It remains that I should summarize (i) the facts disclosed
by the preceding investigation ; (2) the conclusions to which
the facts appear to point.
The main facts are as follows :
(i) There are readings in the Syro-Latin (Greek and
Latin) authorities for the text of the Gospels which, when
examined, betray their Syriac origin. Such readings are of
different kinds, (i) Sometimes a Syriac idiom is reproduced:
see e.g. the notes on Matt, xxiii. 9 (p. 16), Lc. ix. 16 (p. 36),
Lc. xviii. 14 (p. 52) and Chapter 4 passim, (ii) Sometimes
we find a form of expression characteristic of the Syriac texts
of the New Testament intruding itself: see e.g. the notes on
Matt. xxv. 41 (p. 16), Jn. xxi. 7 (p. 26), Lc. ii. 48 (p. 29 ff.),
Lc. v. 14 (p. 85), Lc. XV. 4 (p. 46). (iii) Sometimes the
genesis of a strange reading becomes intelligible when we
seek its origin in a Syriac text: see e.g. the notes on Lc. v. 10
(p. 84), Lc. xxii. 12 (p. 56), Lc. xxiv. 33 (p. 71), Mc. i. 41
(p. 88), Mc. V. 41 (p. 109 f ), Mc. viii. 10 (p. 97)*. (iv) Some-
' I take this opportunity of correcting my mistake (as I now think) as to the
interpretation of the reading of Cod. Laudianus (E) in Acts xvii. 34 kai fY^H
TIMIA. I believe that npia is a translation of the Syriac r^OX^,*:!* (= known)
in the sense of ' a certain,' in which sense it is used in the Peshitta of Acts xvi. 12,
xviii. 23 (see Payne Smith, Thes. Syr., p. 1556; and compare Bp Lightfoot,
Ignatius, i. pp. 144, 146). I suggested (Old Syriac Element, p. 97) that this
rip,la represents an Old Syriac gloss due to assimilation to xvii. 4, 12 (Pesh.). This
may be so, but the solution which I now suggest appears to be simpler.

SUMMARY OF FACTS. 1 29
times two glosses (as it appears) in different Syro-Latin (Greek
and Latin) texts are seen to be divergent representations of
a single Syriac gloss : see notes on Matt. xxvi. 59 ff. (p. 78),
Lc. iii. 10, 12, 14 (p. 34), compare the note on Mc. v. 41
(p. 109). (v) Sometimes a Syro-Latin (Greek or Latin)
reading reproduces or implies a reading which we see to be
characteristic of a Syriac text: see e.g. the notes on Matt.
xviii. 20 (p. 8), Lc. ii. 5 (p. 28), Lc. xxii. 27 (p. 14 n.),
Mc. xii. 14 (p. 18 n.). (vi) Sometimes in a Syro-Latin text
(Greek or Latin) we light upon a Syriacised word or proper
name : see Chapter 3 passim.
(2) The Syro-Latin (Greek and Latin) texts of the
Gospels, especially the Bezan text, shew abundant signs of
harmonistic influence. The phenomena of which we have to
take account (see Chapter 2 passim) are (i) harmonistic
readings which involve coincidence with the Tatianic order of
the narrative ; (2) harmonistic readings in the Bezan text in
which there is an indication of retranslation ; (3) harmonistic
readings in which, or in the context of which, there is an
indication of Syriac influence. The arrangement of the
genealogy in the Bezan text of Lc. iii., which coin
cides with a genealogy given by Aphraat, betrays har
monistic influence other than that of Tatian (see above,
p. 81 f,), (3) An important element in the Syro-Latin texts lies
in the interpolations, longer and shorter. These are of different
kinds. They may, I believe, with fair accuracy be classified
thus : (i) Some may be described as context-supplements :
see e.g. the notes on Jn. vi. 56 (p. 21), Jn. xi. 14 (p. 24),
Lc. xxiii. 40, 42 (pp. 59, 61). (ii) Some are due to a desire
for fulness and completeness of narrative or phraseology : see
e.g. the notes on Matt. xxvi. 15 (p. 18), Lc. ii. 48 (p. 29 ff.),
Lc. XX. 34 (p. 55), Lc. xxiii. 40 ff. (p. 58 ff.). (iii) Some are
the result of assimilation to other passages of Scripture : (a) to
the language of the Old Testament (see p. 46 ff.); (b) to that of
other passages of the Gospels : see e.g. notes on Matt. xx. 28
(p. gff.), Jn, xxi, 13 (p. 27), Lc xiii. 17 (p. 42 f), Lc. xxiii. 37
C. 9

130 THE SYRO-LATIN TEXT OF THE GOSPELS.
(p. 57 f) ; (c) to that of the Acts : see note on Lc. iii. 10 (p. 34).
A case of assimilation to the language of a Pauline epistle is
presented by Lc, xxiv. 32 (p. 69 f ). (iv) A few are proba
bly derived from tradition or from non-Bibhcal literature —
[Jn,] vii. 53 — viii, ii^ Lc. vi, 5 (see p. 66), Lc. xxiii. 53 (see
p. 62 ff.), and perhaps [Mc] xvi. 9 ff. (see Old Syriac Element,
p. iSoff.). (4) As there are additions, so also in the Syro-Latin
texts there are numerous omissions of single words and of
whole clauses. Such omissions, as far as the text of Codex
Bezae is concerned, are indicated in the preceding pages by
the caret ( a)"-
1 Eusebius (II. E. iii. 39) concludes his notice of Papias with the words iKriBeirai
Si KoX &XXtjv laropiav irepl yvvatKbs, iirl iroXXais dp.apriaLS Sia^X'qBeityqs iirl rov Kvpiov,
fjv rb KaB' 'E^palovs evayyiXwv irepdxei (comp. Apost. Const, ii. 24 iripav Si ruia
TipiaprriKviav). The reading of D in [Jn.] viii. 3 em AMApreiA TYNAlKA
eiAHMMENHN will be noticed. Bp Lightfoot (Essays on Supernatural /Religion,
p. 205) suggests that the story of the man working on the Sabbath day, found in
D (Lc. vi. 4) alone, was ' derived from [the] exegetical work of Papias.'
^ Great as is the hesitation which anyone must feel in traversing an opinion
of Dr Hort, I am constrained to express my doubt as to the soundness
of his position in regard to what he terms ' Western non- interpolations ' — Matt.
xxvii. 49, Lc. xxii. igf., xxiv. 3, 6, 12, 36, 40, 51, 52 (see Introduction, pp. 175 ff.,
294 f.). Of these Matt. /. c, as it has other than Western attestation, stands apai't
from the rest. The omissions however in the last three chapters of St Luke rest
on none but Western authorities. But the value of the evidence of these
authorities seems to be reduced to a vanishing quantity, when we take account of
the phenomena of which the preceding pages have afforded many examples, viz.
(i) the extreme capriciousness of these authorities in adding words to, and
omitting words from, the text; (ii) the way in which the Western authorities
conspire in giving what is obviously a wrong reading. It will be best to examine in
detail one 'Western non-interpolation.' In Lc. xxiv. 51 the words Kai dve^ipero
els rbv oipavbv are omitted by X*Dabeffrhe [ = 1] Aug. Dr Hort (/Votes on
Select Readings, p. 73) wrote thus : 'A Western non-interpolation. Text [i.e. the
supposed interpolation] was evidently inserted from an assumption that a separa
tion from the disciples at the close of a Gospel must be the Ascension. The
Ascension apparently did not lie within the proper scope of the Gospels, as seen
in their genuine texts : its true place was at the head of the Acts of the Apostles,
as the preparation for the Day of Pentecost, and thus the beginning of the
history of the Church.' Over and above the weakness of the documentary
evidence for, and the strength of the documentary evidence against, the omission,
the following considerations appear to be pertinent : (i) There is no evidence that
the Ascension 'did not lie within the proper scope of the Gospels.' On the

SUMMARY OF FACTS. 13I
(5) Syro-Latin readings given by Codex Bezae are
found in Irenaeus (see pp. 17, 52^), Marcion (see pp. 37, 87),
and Justin (pp. 17, 48).
contrary, the language of Acts i. i, 2 seems distinctly to imply that the dvdXriii,\j/is
had a place in the rrp&ros Xbyos of St Luke. (2) The tone of Christ's instructions
vv. 48, 49 and the peculiar solemnity of the narrative vv. 50, 5 1 mark this as the
Lord's final departure. The strong impression derived from these verses is
confirmed by St Luke's words as to the Apostles — they ' returned to Jerusalem
with great joy: and were continually in the temple, blessing God.' Such a
description of their feelings and of their conduct points to their entrance on a
wholly new stage of discipleship. (3) The similarity of the language of Acts i. 12
(rbre viri(rrpeif/av eis 'lepov(raX7jfi dirb opovs rou KaXovpi.ivov 'EXatwz'os) to that of
Lc. xxiv. 50 (i^:^ayev Si airois ?us irpos BtjBavlav), 52 (viriarpe^av els
'lepov(raX-!iix) and of that of Actsii. 46 to that of Lc. xxiv. 53 cannot be overlooked.
(4) The reading of the Sinaitic Old Syriac text explains how the omission
in the Western texts may have arisen. It has: 'And-when He-blessed them,
He-was-lifted-up from-them (^^OCfUia pCL»iA»rC').' Here plainly the
Syriac has a compressed rendering of the two clauses Siiarri dir' airwv Kai
dve(pipero els rbv oipavbv, the ideas being preserved, the phraseology abbreviated.
A copyist however, assimilating the Greek to this Old Syriac text, would
naturally be led by the Syriac reading to omit the words Kai dve(j>ipero els
rbv ovpavbv.
It is natural in connexion with these supposed ' Western non-interpolations' to
consider the omission of our Lord's prayer for His enemies (Lc. xxiii. 34) in
' S^BD* 38 82 435 a* me. codd. opt.' (Dr Hort, JVotes on Select Readings, p. 67).
To these authorities for the omission we must now add the Sinaitic Syriac text.
' Its omission,' wrote Dr Hort (p. 68), ' on the hypothesis of its genuineness,
cannot be explained in any reasonable manner.' It is however a significant fact
that in the Arabic Tatian (Hill, p. 249) the Lord's prayer for His enemies is not
in the context in which it stands in Lc. xxiii. 34, but is placed just before the final
prayer of commendation— ' The rest said, Let Him be; let us see whether Elijah
cometh to deliver Him. And Jesus said. My Father, forgive them ; for they
know not what they do. And Jesus, crying again with a loud voice, said. My
Father, into thy hands I commend my spirit.' The displacement of the prayer in
Tatian would naturally lead in Western texts to its omission in Lc. xxiii. 34.
The one difficulty in the way of regarding this as a sufficient explanation of the
omission in the authorities mentioned above lies in the fact that it implies that
Cod. B is guilty of having been influenced by Tatian.
' Npte also the following passages in Irenaeus — '¦a.\S....uniuersa attrahat ad
semetipsum' (in. xvii. 6), 'omnia trahit ad se' (iv. iv. 2). The reference is
clearly to Jn. xii. 32 (see above, p. 25). The diversity in the translator's phrases
together with the fact that all Latin texts, so far as I know, which represent the
Greek reading itdvra, have omnia, makes it almost certain that the original Greek
text of Irenaeus had in these two places rrdvra.
9—2

132 THE SYRO-LATIN TEXT OF THE GOSPELS.
The facts appear to warrant the following conclusions as to
(i) the date, (2) the genesis, (3) the birthplace of the Syro-
Latin (Greek and Latin) text^ of the Gospels,
(i) Date. The Syro-Latin text was no doubt a gradual
growth. The tendencies of which it is the result were active
in the first half of the second century, Syro-Latin readings,
which occur in Codex Bezae, are found, as has just been
noticed, in Irenaeus, Marcion, and Justin. This text then
must have been taking shape and already spreading before
the middle of the second century. This early date indeed
explains some of the chief characteristics of the text. To
this subject I shall presently return. The limits of date
however, as far as the Bezan text is concerned (apart from
the alterations of later transcribers; see below, p. 135), may
be more exactly fixed, (i) The coincidences between the
Bezan text and the Diatessaron forbid our assigning the
former to an earlier date than 170 A.D. (2) We are able to
fix a terminus ad quem, as to the Bezan text of the Acts.
Irenaeus in his Third Book (c xii.) has a series of lengthy
quotations from the Acts (ii. — xv,). In these quotations
there are found very many readings which occur in the text
of Codex Bezae, Hence it is certain that Irenaeus at Lyons
had a text substantially coinciding with the Bezan text.
The Third Book of Irenaeus was written during the episco
pate of Eleutherus (A.D. 175 — 190). The Bezan text of the
Acts therefore must have come into existence early enough
in the second century to allow of its having been used in
South Gaul by Irenaeus in a book which cannot have been
written later than 190 A.D. Hence we may give 180 A.D. as
the approximate date of the Bezan text of the Acts. The
Bezan text of the Acts and the Bezan text of the Gospels

^ The phrase ' the Syro-Latin (Western) text' is of course inaccurate, if it be
taken to imply that there ever existed one normal Syro-Latin (Western) text.
Strictly speaking the phrase should be 'the Syro-Latin (Western) texts or type
of text.' But the singular text is convenient and harmless, if properly
understood.

CONCLUSIONS. 133
exhibit the same characteristics^ It is natural therefore to
^ This consideration is important in regard to the theory of Dr Blass of Halle
(Theol. Studien u. Kritiken, Jahrgang 1895, Erstes Heft, p. 86 — 119 ; comp. the
same scholar's admirable Commentary on the Acts, p. 25!!.) as to the interpo
lations found in the text of the Acts as given by Codd. DE, Old Latin and
Philoxenian Syriac. Dr Blass holds that, as far at least as the interpolations are
concerned, the text of these authorities is derived from St Luke's rough draft, the
common text from his fair copy. Any theory which claims to shew us an
apostolic writer at work must have a fascination. The question however is — Does
this theory take full account of the facts of the case ? Apart from other objections,
it must, I believe, be said that Dr Blass finds certain facts telling in his favour
because he isolates them from other facts. For we cannot separate the Bezan
text of the Acts from the Bezan text of the Gospels nor either of these from the
phenomena of the Syro-Latin N. T. texts generally. Dr Blass indeed seems
uneasily conscious that here there is a difficulty which must be faced. For
in the last paragraph of his article he writes thus : ' Der Codex [D] enthalt
ja auch noch die Evangelien, und weicht, wenn auch nicht eben im Matthaus
und Johannes, so doch im Markus und Lukas recht erheblich von dem
gewohnlichen Texte ab, Aber der Charakter der Abweichungen ist ein
anderer als in der Apostelgeschichte, und auch nicht ein einheitlicher durch-
gehender wie dort, sondern die einzelnen Stellen haben ihre besondere Art....
Sicherlich verdient D auch in diesen Evangelien sorgfaltiges Studium, aber
das Problem oder besser die Probleme sind andere als in der Apostelgeschichte,
und was besonders zu beachten, von Gemeinsamkeit zvidschen D und Zusatzen des
Syrers ist keine Rede.' It is of course true that the interpolations in the Bezan
text of the Acts are a somewhat more glaring feature than are the interpolations
in the Bezan text of the Gospels. But the difference is at most one of degree, not
of kind. Further, the alliance between D and the Old Syriac texts in the Gospels
is much closer and more significant than the alliance between D and the
Philoxenian in the Acts. Curiously enough however Dr Blass singles out one
passage of the Gospels — Mc. i. 6 — where 'D (nebst einigen Itala-Codices und
einem Vulgata- Codex) hat den echten Markus bewahrt. ' A favourable reviewer,
Dr E. Nestle, in the ' Christlichen Welt ' (for 1895, Nos. 13, 14, 15), goes a step
further. After discussing two passages of the Bezan text — Lc. xi. 1, xxii. 16 — he
asks 'Kann man noch zweifeln, dass uns dieser so lange verkannte Kodex eine
eigne, direkt auf das hebraische Urevangelium zuriickgehende Form des
Lukasevangeliums erhalten hat ? Wie Lukas daran ging, den zweiten Teil seiner
Schrift, die Apostelgeschichte, fiir Theophilus auszuarbeiten, scheint er den ersten,
das Evangelium, noch einmal revidirt zu haben.' It cannot then be seriously
maintained that the Bezan text of the Gospels differs generically from the Bezan
text of the Acts. If therefore the theory of Dr Blass is true, we must suppose that
Codex Bezae preserves for us relics of the original drafts of the Gospels ; we must,
that is, assume (i) that the writers of the Gospels, as well as the author of the Acts,
made rough drafts of their viritings; (2) that these rough drafts were all preserved;
(3) that an enterprising editor of the apostolic writings in the second century was
able to bring together these very interesting relics of the Evangelists. This

134 THE SYRO-LATIN TEXT OF THE GOSPELS.
infer that they arose about the same time. Hence the
approximate limits of date as regards the Bezan text of the
Gospels are 170 A.D, and 180 A.D. Codex Bezae exhibits the
Syro-Latin or Western text of the Gospels in a fully
developed form. The cautious verdict therefore of Dr Hort
as to the date of this text must be unreservedly accepted,
'It is probable,' he wrote (Introduction, p. 122), 'that even the
relatively latest Western readings found in distinct provinces
of Western documents, for instance in different languages,
were already in existence at a very early date of Church
history, it may be before the end of the second century.'
(2) The genesis of the Syro-Latin (Greek and Latin)
text. It will be convenient to take Codex Bezae as a type of
the documents containing this text, and to construct a
theory which will account for the peculiarities of this one
MS,
Three points claim attention.
(i) Certain passages of the Bezan text are the result of
the definite assimilation of a Greek to an Old Syriac text.
The simplest and most adequate theory is, I believe, the hypo
thesis that the Greek text of Codex D is the Greek text of a
Graeco-Syriac bilingual MS., and that therefore the 'Bezan
scribe' wrote out his Greek text with the Syriac text close at
hand. This theory satisfactorily accounts for the phenomena
of the Greek text — for the chaos into which the Greek falls at
times, for the want of uniformity in the Syriacisation. Some
times the copyist transcribed the Greek accurately enough.
Then his attention was attracted to the Syriac : he inserted a
Syriac gloss, giving his own Greek rendering of it : he
retranslated a Syriac phrase. Sometimes the Greek copy
before him was hard to decipher, or he lost his place in the
MS, which he was transcribing ; at such times he went on
writing out the Greek, reproducing it as his memory was
aided or confused by the Syriac before him. Hence his
accumulation of improbabilities, which the theory appears necessarily to involve,
is, I believe, its sufficient refutation.

CONCLUSIONS, 135
Greek at this point is retranslation from the Syriac, Greek in
vocabulary, but largely influenced by Syriac idiom,
(ii) There are phenomena in the Bezan text which seem
to witness not so much to definite assimilation to a Syriac
text as to the work of a bilingual (i.e. Syro-Greek) scribe,
familiar with the Syriac text of the Gospels and accustomed
to think in Syriac, Such a scribe would naturally introduce
into the Greek text in one place a Syriacised form of a proper
name, in another a Syriac idiom.
But at this point the question will be asked — Does this
theory of assimilation to the Syriac claim to explain every
variant from the true text which is found in Codex Bezae ?
The answer to this question is in the negative. Assimilation
to the Syriac is, I believe, the determining, dominating
influence. But doubtless other minor forces have been at
work, (a) There are some passages, not, I believe, many in
number, in which the copyist, who transcribed the MS. now in
the Cambridge University Library, ' allowing his eye to
wander to the Latin copy before him, while he wrote the
Greek, may have been influenced by the Latin in his
transcription of a word or phrase of the Greek. But these
instances of Latinisation...are accidents of the particular
transcription, and do not affect the essential character of the
text which the MS. presents' (Old Syriac Element, p. 2).
(b) It would be rash to assert that the Greek text of the MS.
as we have it was transcribed immediately from the second
century Syriacised text. I believe that the former is not
separated from the latter by many steps. But to any such
intervening transcription some changes of text would be due.
(c) The bilingual scribe who Syriacised the text had, as was
natural enough in the second century, lax views of the
faithfulness required of a transcriber. If he felt at liberty to
assimilate the Greek to a Syriac text, he would not be likely
to abstain from emending and amplifying the Greek text,
quite apart from such assimilation. To such laxity of
transcription on the part of the Bezan scribe we probably
owe, to take one example, the Bezan reading in Lc. xiii. 8

136 THE SYRO-LATIN TEXT OF THE GOSPELS.
ecoc OTOY cKAyco nepi aythn kai BaAoo KO(t)iNON KonpicoN (true
text KOTTpia). Here we have introduced what appears to be a
common agricultural phrase ; comp, Colum., de Re Rust., xi. 3
(quoted in Forcellini) : ' confecta bruma stercoratam terram
inditam cophinis obserat,' Dr Hort^ quotes Plut., Vita Pomp.,
48 avrov Be rvi Korrpiwv ko^ivov Kara Ke(j)aXrj<i rov ^v0Xov
KareaKeBaae. (iii) What account can be given of the interpolations^ in
the Bezan text .-¦ They are, I believe, due in the main to two
influences, (a) This text arose in a bilingual Church, where
the Books of the New Testament were read in Syriac as well
as in the original Greek, and where the former reacted on
the latter. But translation, especially popular translation,
insensibly passes into paraphrase, and paraphrase again into
comments Hence short glosses and interpolations would
inevitably arise. It is in a bilingual Church that we should
look for considerable licence in this direction, (b) The Syro-
Latin text was in process of formation before the second
century was far advanced. At that time the unique au
thority of the Books of the New Testament was only
beginning to be recognised. Certainly the importance
attaching to the ipsissima verba of the Books was not then
understood as it has been by later generations. In the
assemblies of the Christians the writings of the Prophets and
of the Apostles were read''. This reading was followed by
' In some MS. notes, which I have been allowed to see.
^ For Syro-Latin interpolations (other than those referred to in this Essay) see
the notes in Dr Hort's Introduction on Matt. iii. 15, xvi. i, xx. 33, xxvii. 38, Mc.
xii. 23, xvi. 3, 14, Lc. xxi. 38, xxiii. 2, 5, 48.
' So, to take one example. Sin. has in Lc. v, 7 (dare ^vBl^eaBai aird) 'And-
near were-they from their-weight to-sink.' Here the words 'near were-they... to-
sink' are a paraphrase; the phrase 'from their-weight' is a brief comment (see
above, p. 35).
¦* Compare Justin, Apol. i. 67, and the following passages from the Doctrine
of Addai (ed. Phillips) : ' [Addai] made them partakers with him in the ministry ;
they read in the Old Testament and the New, and the Prophets, and the Acts of
the Apostles ; every day they meditated on them' (p. 33). ' A large multitude of
people assembled day by day and came to the prayer of the service, and to the
reading of the Old and New Testament, of the Diatessaron' (p. 34). 'But the

CONCLUSIONS. 137
explanation and exhortation'. It would be very natural that
some of these comments should become stereotyped and
should attach themselves in some cases to the text itself^.
Such a practice would grow up and prevail both in the Greek
and in the Syrian congregations of a bilingual (Syro-Greek)
Law and the Prophets, and the Gospel, which ye read every day before the
people, and the Epistles of Paul. ..and the Acts of the Twelve Apostles...; these
Books read ye in the churches of Christ ' (p. 44). See also the ' Ancient Homily '
xvii., with Bp Lightfoot's notes (Clement, ii. p. 257).
1 Comp. Justin loc. cit. This custom the Christian Church inherited from the
Synagogue (comp. Lc. iv. 20 ff.. Acts xiii. 16 ff.). 'The reading of the Scriptures
was followed by an edifying lecture or sermon (TW'^'^^ by which the portion
which had been read was explained and applied' (Schiirer, The Jewish People,
Div. ii. vol. ii. p. 82, Eng. trans.). 'The reading was accompanied by a con
tinuous translation into the Aramaic dialect' (Schiirer, p. 81). It is not impossible
that such translation had a place in the services of a bilingual (Christian) Church.
''¦ Such probably is the history of those Christian interpolations in the Old
Testament, which among the Christians had become so firmly embedded in the
LXX. text that Justin accuses the Jews of having erased them (Dial. 297 Dff.).
Two other points may be noticed, (i) It would appear that non-Canonical
writings were sometimes read in the assemblies of the Christians. («) Such a
practice seems to be implied by the prohibition in the Doctrine of Addai (p. 44) :
' And with these read not any others, as there is not any other in which the truth
which ye hold is written, except these books, which retain ye in the faith to which
ye have been called.' (b) Dionysius of Corinth (circ. 170 — 175 A.D.), writing to
Soter, Bp of Rome, in acknowledgment of a letter from the Roman Church (Eus.
H. E. iv. 23), says that the Corinthian Christians had read the letter that day — 'the
Lord's Day' — and that they would keep it and read it from time to time, as they
did the former letter written to them by Clement. It does not then seem
improbable that such works as the ii,rfyT\aei,s of Papias were read publicly in
connexion with the Scriptural lections, and that in this way illustrations from such
books attached themselves to the text of the Gospels. (2) The 'Ancient
Homily,' commonly called the 'Second Epistle of Clement,' was apparently a
written discourse (xix). 'It was,' says Bp Lightfoot (Clement, ii. p. 197 f.),
' considered of sufficient value to be carefully preserved ; and (as we may venture
to suppose) it was read publicly to the Christian congregation at Corinth from time
to time.' If now and again a discourse of 'the president,' which followed the
reading of the Gospels, was thus preserved and 'read publicly to the Christian
congregation from time to time,' it would be very natural that a. paraphrase or a
gloss or a telling quotation from the Old Testament, contained in it, should link
itself to the passage of the Gospels which it explained or enforced.
Such an explanation of the phenomena of second century texts seems natural
and in accordance with the somewhat meagre evidence at our disposal, but of
course it does not claim to rise above a not improbable conjecture.

138 THE SYRO-LATIN TEXT OF THE GOSPELS.
Church, and would leave its mark on a text of the New
Testament, which was the outcome of the common life of that
Church. When once we realize the circumstances of place and
time under which the Bezan text arose, we see that the
interpolations, which form so striking a feature in that text,
are absolutely natural. They are, at least in most cases, due
(i) to the influence of translation and retranslation in a
bilingual Church, and (ii) to the methods of instruction which
prevailed in the Christian congregation. Probably those
derived from purely literary sources are, to say the least,
very rare.
(3) The birthplace of the Syro-Latin text.
Here again it will be convenient to narrow the question
and to consider primarily what was the birthplace of the
Bezan text.
The answer to this question must fulfil three conditions.
(i) The birthplace of the Bezan text must have been a
Church where the life of the Christian body was vigorous ;
where the study of Scripture was keenly prosecuted ; where
such traditions as that about ' the woman taken in adultery'
and that about ' the man found working on the Sabbath day,'
whether they are due to a literary or an oral source, would be
likely to find a home, (ii) It must have been a bilingual
Church, where, that is, Greek and Syriac were both spoken.
(iii) It must have been a place in constant communication
with different parts of the world, so that a text of the
New Testament current there would spread rapidly and
widely. The Church of Antioch appears to satisfy these conditions
as no other Church does.
(i) Without controversy the Church of Antioch had a
vigorous life of its own. In apostolic times it was the
metropolis of Gentile Christianity, the Church which sent
St Paul forth on his several missionary journeys, and to which
he returned on their completion. In the early years of the
second century, Ignatius, the martyr-Bishop of Antioch, with

CONCLUSIONS, 139
his force of character and his practical enthusiasm, is the
most striking personality which the fragmentary history of
the time brings before us. Later in the century, about the
time when, as we have seen reason to think, the Bezan text
arose, Theophilus (circ. A.D, 170 — 185), 'the sixth from the
Apostles ' (Eus. H. E. iv. 20), presided over this see. Theo
philus was fertile as a controversial and apologetic writer
(Eus. H. E. iv. 24, Jerome, de Vir. Illust., c 25). It is
important for our purpose to notice that he seems specially to
have occupied himself in the study of Scripture. In the three
books addressed to Autolycus arguments drawn from the
Old Testament occupy a large space. Moreover Jerome tells
us (loc. cit) that he had read some commentaries of his ' in
euangelium" and on the Proverbs of Solomon, adding
however that they appeared to him inferior to his other
works in elegance and style. Elsewhere Jerome mentions
the significant fact that Theophilus drew up a harmony of
the four Gospels I During the last years of the century
(circ. A.D. 190 — 203) the Bishop of Antioch was Serapion, a
controversial writer, some of whose works are mentioned by
Eusebius (H.E. vi. 12, comp. v. 19).
Thus early and late in the second century the leaders of
the Church of Antioch were men of character and power,
whose writings occupy a conspicuous position in the Chris
tian literature of the second century.
(ii) Antioch was a bilingual city. 'Antioch,' writes
Renan (Les Apdtres, p. 217; Eng. trans., p. 181 f), 'from its
' Compare Jerome, Prol. in Comm. in Matth. : ' Et Theophili Antiochenae
urbis Episcopi commentarios.' There is extant a Latin commentary bearing the
name of Theophilus of Antioch, the genuineness of which has been maintained by
Zahn, but denied by Harnack (see the convenient summary of the arguments in
Dr Sanday's paper, Studia Biblica, i. p. 89 ff.). There seems to be little room for
doubt that the arguments of the latter scholar are decisive.
2 Ep. ad Algesiam, Qu. vi. : ' Qui quatuor euangelistarum in unum opus
dicta compingens ingenii sui nobis monumenta dimisit. ' Was this a Greek
version of the Diatessaron, the orthodoxy of which was guaranteed by the name of
Theophilus? Had it been preserved, it would doubtless have cleared up many
points, which are now obscure, as to the relation of the Diatessaron to the Syro-
Latin (Greek and Latin) authorities.

140 THE SYRO-LATIN TEXT OF THE GOSPELS,
foundation, had been altogether a Greek city,... Besides the
Greek population indeed, which in no part of the East (with
the exception of Alexandria) was as numerous as here,
Antioch included in its population a considerable number of
native Syrians, speaking Syriac. These natives composed a
low class, inhabiting the suburbs of the great city, and the
populous villages which formed vast outskirts all around
it, Charandama, Ghisira, Gandigura, and Apate, names chiefly
Syriac. Marriages between the Syrians and the Greeks were
common, Seleucus having formerly made naturalization a
legal obligation binding on every stranger establishing
himself in the city, so that Antioch, at the end of three
centuries and a half of its existence, became one of the
places in the world where race was most intermingled with
race,' Thus at Antioch many of the Christians, especially those
of lower social rank — and it was among such that Christianity
won its most signal triumphs — must have been native
Syrians, To these the Diatessaron would be brought from
the Syrian Churches further East\ At Antioch in the
intercommunion of Greek-speaking and Syriac-speaking
Christians there would be need of bilingual teachers. There
would grow up a school, if the expression be not too formal,
of bilingual scribes. Codex Bezae preserves to us, I believe,
a precious relic of their work. But its text can only be one
of many similar texts".
(iii) Lastly, Antioch was in direct communication with
all parts of the then known world. The Orontes, on which
1 We have evidence for intercourse between these Churches and Antioch ; see
the passage in the Doctrine of Addai, p. 50 : ' He [Aggai] was not able to place
the hand upon Palut. Palut himself went to Antioch, and received the hand of
the priesthood from Serapion, Bishop of Antioch.' The Doctrine of Addai is 'in
its present shape a work of the latter half of the 4th century ' (Wright, Short Hist.
of Syriac Literature, pp. 9, 43).
^ The alliance of D and the Old Latin e in certain noteworthy readings (see
above, pp. 21, 45, 47, 54 n., 83, 94; cf. no) is a remarkable fact, to which, so far
as I know, attention has not been called. But these MSS. are representatives of
kindred, not identical, recensions of the text. On the relation between D and
E (Cod. Laudianus) in the Acts, see Old Syriac Element, p. 134 ff.

CONCLUSIONS. 141
the city stood, flowed into the sea some sixteen miles
westward of the Syrian capital at the port Seleucia, Vessels
must have been continually arriving from, and sailing for,
Ostia, South Gaul, Carthage, Alexandria, Christianity and
the Christian Scriptures followed in the wake of trade.
Hence we can easily understand how an Antiochene (i.e. a
Syriacised) text of the New Testament was in the hands of
Irenaeus at Lyons and of Tertullian at Carthage, how it
passed to Alexandria, and from Alexandria to the native
Egyptian Churches ^
' On the intercourse between Syria and the West see especially Zahn, Geschichte
des Neutest. Kanons, i. p. 414 ff.
The theory that Antioch was the birthplace of the ' Western ' or Syro-Latin
text I discussed in my former volume (Old Syriac Element, p. 115 — 149). I there
quoted at length from a. review of Mr Rendel Harris' Study of Codex Bezae
which appeared in the Guardian of May 18 and May 25, 1892. I noted with
satisfaction that the writer of this review, whom I may now refer to as Dr Sanday,
on grounds independent of mine, arrived at the conclusion that the ' Western ' text
arose at Antioch. I am glad to find that this view is accepted by a writer in the
Dublin Review (July, 1894) — the Rev. H. Lucas, S. J. — who at the end of a
review of my book writes thus : ' No other place of origin will, I believe, be found
to account for the many-sided phenomena presented by the so-called (and un
fortunately so-called) "Western text" of the New Testament.'
Two views may be held as to the relation between the Old Latin text (or
texts) and the birthplace of the 'Western' text, (i) On the one hand Dr Hort
(Introduction, p. 188) wrote thus : 'On the whole we are disposed to suspect that
the ' Western ' text took its rise in North-western Syria or Asia Minor, and that
it was soon carried to Rome, and thence spread in different directions to North
Africa and most of the countries of Europe. From North-western Syria it would
easily pass through Palestine and Egypt to Ethiopia.' According to this view
Greek MSS., stamped with the characteristics of the 'Western' text, passed from
the birthplace of that text to Rome or North Africa, and there became the basis of
the Old Latin text. Thus a distinction is drawn between the birthplace of the
' Western ' and the birthplace of the Old Latin texts, (ii) On the other hand
Dr Sanday, in the review above referred to, is inclined to identify the birthplace
of the 'Western' with that of the Old Latin text. Referring to Dr Hort's words
quoted above he says : ' For "North -Western Syria" we would venture to substitute
"Antioch," because what we want is, in a strict sense, a "centre," a manufactory
where a succession of MSS. might be produced in near juxtaposition to each other,
Antioch satisfies this condition better than any other Church... Our assumption
is... that the Latin Version itself may have been made in Syria, and we will say
boldly at Antioch. '
There does not appear to be sufficient evidence to justify an absolute decision

142 THE SYRO-LATIN TEXT OF THE GOSPELS.
As regards the text of the New Testament, Antioch, we
may believe, was in the second century (as it seems to have
been in the fourth) a kind of watershed, where streams
took their rise, which, coloured afterwards by the various
soils through which they passed, flowed to the distant parts
of Christendom.
between these two views. The analogy of the Egyptian Versions, in which a distinct
'Western' element is found and which must have arisen in the native Egyptian
Churches, favours the first view. Further, a. remarkable reading in e (Mc. v. 41 ;
see above, p. no) seems to afford clear proof that the text of that MS. 'did not
spring up on Syrian soil.'

INDEX I.

MSS., VERSIONS, PATRISTIC WRITINGS.

I. Uncial MSS.
X (Cod. Sinaiticus), 5, 20, 25, 42, 74,
85, 102, 105, 124.
B (Cod. Vaticanus), 93, 131 n.
C (Cod. Ephraemi), 59, 106, non.
D (Cod. 'Beza.e:), passim.
G (Cod. Harleianus), 102.
K (Cod. Cyprius), 74.
L (Cod. Regius), 5, 8, 42.
M (Cod. Campianus), 114.
N (Cod. Purpureus), 97.
P (Cod. Guelpherbytanus A), 123.
U (Cod. Nanianus I), 63.
X (Cod. Monacensis), 30, 92, 123.
A (Cod. Sangallensis), 104.
II (Cod. Petropolitanus), 74.
* (Cod. (Purpureus) Beratinus; ed. Ba-
tiffol), 6, 9, 44.
2. Cursive MSS.
I — (118 — 131 — ) 209 (see Dr Hort,
Introduction, p. 154, and above p.
18 n.), 5, 18, 30, 57, 72n., 94n., 97,
inn., 121.
2P= (=473 (Scrivener), 565 (Gregory),
81 (Dr Hort, Introd., p, 154)), 42,
48,97, 113, 125, 126.
6P=, p. 25.
12 — 69 — 124 — 346 (Ferrar-group; see
above, p. 4n.) — 556 (Scrivener, Ad
versaria Crit. Sacra, p. I ff.), 4, 25,
30, 38, 41, 44, 57. 631 94n-> 97. 102,
113, 124.

22, p. 18.
26'^', p. 16.
28, pp. 5, 48, 72 n., 97.
29, p. 66.
38, p. 131 n.
56. p. 25.
61, p. 25.
64, pp. 92, 106.
69, p. 21.
82, p. 131 n.
88, p. 74.
122, p. 123.
130, p. 74.
157. pp- 39. 54 n., 81.
225, pp. 52, 125.
23s. P- 25.
237. P- 125.
245. P- 52-
262, p. 30.
435. P- 131 n-
604 (=700 Gregory; ed. Hoskier), pp.
5, 15, 48, 105, 126.
3. Versions.
(i) Syriac.
Sinaitic Palimpsest, passim.
Curetonian, passim.
Peshitta, passim.
Harklean (= Philoxenian), 9n., 93, 104.
Jerusalem Lectionary, 27, 102, io4n.,
107.
Arabic Tatian (ed. Ciasca; Hill, The
Earliest Life of Christ), 8, 15, 17,

144

INDEX L

i8, 20, 26, 30, 33, 39, 41 n., son.,
63. 77. 80, 81, 82, 83, 85, 92, 93, 95,
96, 97, 98, 99, 119.
Ephrem, Com. on Tatian (ed. Moe
singer; Hill, The Earliest Life of
Christ), 6, 22, 24, 25, 30, 31, 32, 39,
S3, 6on., 6in., 62 n., 83, 88, 92 n.,
96, 99, 100. (ii) Latin''-.
(i) African text:
e (Cod. Palatinus; see Old Latin
Texts, No. II.), 8, 13, 16, 21, 23,
30, 31, 36, 39, 41, 44, 4S, 46, 47,
54n., ssn., 56, 66n., 69, 71, 72n.,
77, 83, 90, 94, 96, 103, 109, no,
inn., 123, 124, 125, 140 n.
k (Cod. Bobiensis; see Old Latin
Texts, No. II.), 4, s, 6, I'jn., 48,
77. 99. 103, 107. "3. "4. "7,
125, 126.
m ('lectiones e libro de diuinis scrip-
turis siue Speculum'), 13, 26.
(2) European text :
a (Cod. Vercellensis'^), 4, 5, 7, 8, 13,
17 n., 18, 22, 23, 26, 30, 37, 38,
39, 41, 42, 46, 48. 52, 54"-, 56.
66n., 72 n., 7-8, 81, 87, 89, 90, 94,
96, 97, 102, 106, 107, 109, non.,
mn., 113, 121, 124, I2S, 126.
b (Cod. Veronensis), 4, 5, 7, 8, 13,
16, 17 n., 18, 19, 21, 23, 30, 32,
34, 37. 38, 39, 41, 42, 44, 48, 52,
S4n., 6on., 66 n., 72 n., 77, 81,
87, 89, 90, 96, 97, los, 107. 109,
113, 121, 123, I2S, 126.
h (Cod. Claromontanus), 5, 13, 16,
18, 77, 78, 81, 107, 121, 123.
i (Cod. Vindobonensis), 41, 42, 44,
62, 64 n-, 65, 57, 72 n., 94, 97,
107, 109, 113, 123, 125, 126.
1 For the classification see Bp J. Wordsworth's Edition of the Vulgate {Euatig, sec. MattJiemn,
p. xxxiii). ^ According to Bp Wordsworth (p. xxxiii) a has in Matt, a European text, in Mc. Lc. Jn. u
'mixed' text.
3 'In Mattheo uersionem ueterem exhibet...in reliquis euangeliis est Vulgate uersionis quam-
uis lectionibus ueterihus saepissime turbatus' (Bp J. Wordsworth).
4 s (Frag. Lucae Ambrosiana Mediolanensia ; not classified, see Old Latin Texts, No. li.,
p. ccxxix), 54 n., 123.

n (Fragmenta Sangallensia ; see Old
Latin Texts, No. II.), 8, 13, 99,
126.
r (r^. Cod. Usserianus), 18, 30, 31,
36, 37, 41, 54 n-, 65, 66"-, 72 n.,
74, 87, 89, 96, 99, 122, 124.
(3) Italian text:
f (Cod. Brixianus), 8, 16, 27, 36, 41,
42, 44, 46, 72 n., 78, 81, 94, 96,
los, 113, IIS, 117, 121, 123, 124,
125.
q (Cod. Monacensis ; see Old Latin
Texts, No. III.), 4, 6, 18, 30, 34,
37, 38, 39, 41, 42, 44, 48, 52,
64"., 65, 57, 66 n., 72 n., 77, 81,
87, 9°, 94, 96, inn., 113, 117,
121, 123, 126.
(4) Mixed text:
c (Cod. Colbertinus), 4, 7, 13, 16, iS,
23, 30, 31, 36, 38, 39, 4', 42, 44,
48, 50, 52, 54 n., 55"-, 56, 67, 65,
69, 71, 72 n., 75, 77, 81, 87, 90,
94, 96, 97, 105. io7> '09, I '3,
114, 121, 122, 123, 125.
ff' (Cod. Corbeiensis i), s, 7, 13, 18,
121, 123.
ff^ (Cod. Corbeiensis 2), 7, 13, 16,
17 n., 18, 22, 23, 26, 30, 37, 42,
44, 48, 52, 54 n., 55, 57, 66 n.,
72 n., 78, 81, 87, 89, 90, 94 n., 96,
97, 99, 105, 107, 109, "o, "I,
113, 121, 123, 124, 125.
g' ( = G.^ Cod. Sangermanensis i;
see Old Latin Texts, No. I.), 4, 6,
7, 8, 9, 13, 18, 26, 27, 30, 34, 37,
87, 89, 94, 113, 1x4.
g'^ (Cod. Sangermanensis 2), 13, 30,
36, 42, 87, 109, 123.
(5) 'Textus prope-Hieronymianus' I''
aur (Cod. aureus Holmiensis), 90.
1 (Cod. Rhedigerianus), 7, 19, 23, 26,

INDEX L

I4S

30, 37, 38, 41, 42, 44, 48, 52,
54 n., ssn., 56, 6g, 87, 90, 94, 95,
96, III n., 121, 123, 130 n.
(6) Latin texts of bilingual MSS. ;
d (Latin of D, Cod. Bezae'), 3, 6, 8,
18 n.", 23 n., 26*, 27 n.*, 31*, 39,
43, 45*, 46, 53 n-, 55n-, 56, 67*,
6in., 6s, 7 in., 72n.*,78n., 83,90*,
93, IOI, 102*, 104*, 105, 106*,

fuld ( = F. Cod, Noui Test. Fulden
sis), 119 n.
theo (Frag. Theotisca versionis antiq.
evang. S. Matthaei: see Tischen
dorf, Prolegom., iii., p. 1126), 13.
(iii) Egyptian.
Memphitic (or Bohairic), 6, 16, 25, 59,
63, 74 n-, 94 n-, inn.

108*, ii3n.*, IIS*, 116, "7, 118, Thebaic (or Sahidic), 16, 59, 65, 71
119*, 120*, 121*, 122*, 123*, I2S*- 86n, 97, 125.
8 (Latin of A, Cod. Sangallensis), 122.
(7) Vulgate :
vg. (Vulgate MSS. generally), 5, 6,
13, 16, 24, 26, 42, S3, 75, 94,
i°3, 105, logn., ii7f., 121, 123,
124, 126-
A (Cod. Amiatinus), 48.
D (Cod. Dublinensis: 'Book of Ar
magh'), 30, 102, 105.
E (Cod. Egertonensis, = mm (olim
Maioris Monasterii Turonensis)),
26, 27, 63, 65, 81.
S' (Cod. Epternacensis), 30, 81.
J (Cod. Foro-Juliensis), 107.
K (Cod. Karolinus), 107.
L (Cod. Lichfeldensis), 8, 30. ,
M (Cod. Mediolanensis), 67-
M"(Cod. Martini-Turonensis), 107.
Q (Cod. Kenanensis, ' Book of Kells ') ,
8, 30, 53, 65, 66 n., 78, 81, 99 n.
R (Cod. Rushworthianus, 'Gospels
of Mac Regol'), 8, 18, 30, 105.
T (Cod. Toletanus), S3, 107.
V (Cod. Vallicellanus), 107.
Y (Cod. Euang. Insulae Lindisfarn-
ensis), 81.
Z (Cod. Harleianus), 104, loj.
and (Euang. S. Andreae Auenionen-
sia), 13.
emm (Euang. S. Emmerami Ratis-
bonae), 13.
gat (Cod. Euang. S. Gatiani Turon-

(IV) Aethwpic.
I7n., 25, 26n., sgn., 59, 63, gin.,
94 n., non. (v) Armenian.
6, 21, 26, 36, 39, 72 n., 91 n., 94 n.,
inn., 125. (vi) Arabic, 86 n.
(vii) Gothic, 115.
(viii) Persian, 26 n.
4. Patristic writings.
Ambrose, 87, 109.
Aphraat, 16, 21, 22, 26, 28 f., 36 n.,
49, 6i n., 82, no.
Augustine, 56, 75, 97.
Clement (Alex.), 16, 18, 48, 66.
Clem. Hom., 7, 17.
Chrysostom, 59, 61 n.
Cyprian, 6, i8, 55 n., 56, 72 n.
Cyril (of Jerusalem), 50, 69, 62.
Dial, contra Marc, 52, 74.
Epiphanius, 37, 48, 61 n., 74n., 87, 88 n.
Eusebius, 7, 18.
Hilary, 30.

ensis), 30, 34, 53, S5, 77, 81, 105. Jerome, 73; pseudo-Jerome, 62.

' Pages marked with * give a reading of the Bezan Latin differing from that of the Bezan
Greek.

¦ C.

10

146

Ignatius, 73.
Irenaeus, 17, S2, 70 n., 74, 131 n., 132,
Justin Martyr, 17, 48.
Marcion, 37, 61 n., 87.
Origen, 19, 30, 56, 61 n., 81.
Perpetua, Acts of, 87 n.

INDEX L
Peter, Gospel of, 20, so"-, 59, 65 n..

Pilati Gesta, 58, S9, 60, 61 n.
Polyeuctes, Acts of, 60.
Tertullian, 7, 18, 7on., 87, 124.
Victor of Capua, 82.
Victorinus, 22.

INDEX II.

GENERAL.

Active voice, in paraphrase of passive
voice, 4, i6f.
Acts, Book of the, ii. 17, 74, 86 n. ; iii.
4, 107; ix. 40, no; xii. 10, 51; xv.
2g, 70 n.; xvi. 4, 70 n.; xvi. 30, 35;
xvii. 34, 128 n. Bezan text of, date,
132 ff.
Addai, the Doctrine of, i36n., 137 n.,
140 n.
Ambrose, 109.
Antioch, birthplace of Bezan and Syro-
Latin texts, 64, 73, 138 ff.
Aphraat, 21, 22, 26 n., 28, 49, 61 n.,
82, no.
Article (Greek), 112 f.
Assimilation, to context, 3, 19, 26 n.,
47, 52, 69, 61 and n., 6g, 71, gi ; to
O.T., 47 ff. ; to other passages of
Gospels, 19, 21, 27, 40, 41, 43, 44,
66, 72, 74 (cf. Chapter 3 passim); to
Acts, 34; to Pauline Epistles, 69 f.,
70 n.
Augustine, 56, 75, 97-
Baethgen, 8, 17, 33, 45 n., 46, 63, 55n',
72n., 92, 112, IIS, ngn., 124, 125.
Bede, S3-
Bert, Dr, 82.
Blass, Dr, 133 n.
Burkitt, Mr F. C, i8, 76 n., 126.
Clement, Syriac and Latin versions of
Epistle, 33.

Comparative, Syriac form of, 53.
Cureton, 55 n., 82 n.
Cyril of Jerusalem, 50 (see also Index i).
Date of Bezan and Syro-Latin texts, s i
64, 132 ff.
Double renderings, 29 f., 46, 91.
Ephrem, Commentary on Diatessaron,
28, 39, 67, 73 n., 88, 108 (see also
Index i).
Ezekiel, vii. 15 ff-, xxi. 7, p. 49 f.; xl.
6, SI-
Eusebius, H.E., 130 n., 139.
^eic, Syriac equivalent of, 23, 41 f.
Galatians, Epistle to, v. 13 — 25, 71 n.
Genesis, Book of, xxviii. 19, log.
Greek Harmony of Gospels, g7 n.
Harris, Mr J. Rendel, 4n., 7on.,86n.,
106 n., 108, n3n.
Hegesippus, s i n.
Hexapla, Syriac, 5 1 n.
Homily, the Ancient, 137 n.
Hort, Dr, 17 n., 61 n., 70 n., 104, 106 n.,
130 n., 134, 136, 141 n.
Jeremiah, Book of, iii. i, 48 f.; 1. 43,
50-
Jerome, 139 (see also Index i).
Ignatius, 51 n., 73, 138.
Toel, Book of, ii. 28 ff., 86n-

148

INDEX IL

St John, Gospel according to, ii. 3, 42 ;
V. IS, inn.; xi. 43, non. ; xii. 32,
131 n.; xvii. 3, 113 n. See Table of
Contents.
Josephus, 63, 68, 105-
Justin Martyr, 51, I36n. See also
Index I.

Lewis, Mrs, 55 n., 100 n.
Lexicographers, Syriac, 103.
Lightfoot, Bp, 33, log, 128 n., 130 n.,
137 n-
Lucas, Rev. H., 141 n.
St Luke, Gospel according to, iii. 22,
48; vi. s, 66, 130 n.; vii. 14, non.;
viii. 8, 31; viii. 29, in; ix. 16, 31;
xi. 44, 86 n.; xiii. 8, 13S; xix. 34, 42;
xix. 37, 43; xxi. 26, 49; xxii. 27,
I4ri. ; xxiii. 28, 31; xxiii. 34, 131 n. ;
xxiii. 48, 131 ; xxiv. 43, 74 ; xxiv. 51,
130 n. See Table of Contents.
Maccabees, First Book of the, xi. 67, 105.
St Mark, Gospel according to, i. 40,
88; viii. 17, 42; x. n, 48; a. 40,
17 n.; xii. 14, 18 n.; xiv. 15, 56;
xiv. 26, 72; XV. 34, 106; xvi. gff.,
130. See Table of Contents.
St Matthew, Gospel according to, v.
12, 86 n.; xiii. 48, 6; xix. 2s, 30; xx.
23, 17; xxi. g, 43; xxv. I, 30; xxvii.
16, 60. See Table of Contents.
Misreadings of Greek words in Syriac
texts, 8, I4n., I7n., 20, 72n., 97.
Nestle, Dr E., 133 n.
Noldeke, 16, 67, 102, 103.
Old Syriac Element, 2on. , 33 n., 50 n. ,
51, 70 n., 74, 86, 128 n., 130, 135,
141 n.
Omission of certain verbs (e.g. 'he
began') in Syriac texts, 45, 124 .

' One,' insertion of the word in Syriac
texts, 8; comp. 128 n.
'Our Lord' in Syriac N.T., 26.
Papias, i3on., 137 n.
Peter, Gospel of, 20, son., 60, 65 n.,
100.
Prepositions, 36, 38, 43, 71"-, "3^-
Pronouns, 5, 15, 16, 24, 72: suffixes,
2nd and 3rd person plur., confusion of,
86 n.
npa-, Syriac equivalent of, in compound
verbs, S4-
Relative in Syriac, s. 23.
Renan, 139 f-
Resch, Dr, 17 n., 48.
Resolution in Syriac N.T. of Greek
participle, 54, (>(>, 80, 83, iisff.
Robinson, Prof. A., 52 n., 88.
Sanday, Dr, 18 n., 139 n., 141 n.
Schiirer, Dr, 105, 137 n.
Scrivener, Dr, 26.
Septuagint, 48, 51 n-, 87, 109, 137 n.
Serapion, 139.
Sinaitic Syriac, harmonized passages in,
76 n., 79f., 81, g7, ngn.
Tacitus, 64.
Theophilus of Antioch, I3g.
'There' added in Syriac texts, 7f.
Trench, Archbp, 61 n.
Verbal constructions, 54, (>(>, 67, 80,
83, 165 ff-
Westcott, Bp, 57.
Wright, Prof., 82 n., 140 n.
Wordsworth, Bp J., edition of Latin
Vulgate, passim.
Zahn, Dr, 28 n., I3gn., 141 n.

CAMBRIDCJE : PRINTED BY J. & C. F. CLAY, AT THE UNIVERSITY PRESS-

~'S»-.'<«a«**;