(Henning Lehmann) Students of The Bible in 4th and PDF

Students of the Bible
in 4th and 5th Century Syria

Students of the Bible
in 4th and 5th Century Syria
seats of learning, sidelights and syriacisms
by henning Lehmann
a a r h u s u n i v e r s i t Y p r e ss | a
Students of the Bible in 4th and 5th Century Syria
© The author and Aarhus University Press 2008
Layout and cover: Jørgen Sparre
ISBN 978 87 7934 991 9
Aarhus University Press
Langelandsgade 177
DK-8200 Aarhus N
www.unipress.dk
Fax 00 45 89 42 53 80
Published with grants from Carlsbergfondet

Contents
7 Introduction
13 Hosanna
A Philological Discussion in the Old Church
23 The Spirit οf God upon the Face οf The Waters

The Sources of St. Basil’s and St. Augustine’s
Comments on Gen 1,2c
37 Some Questions Concerning the Armenian

Version of the Epistle of James
67 Severian of Gabala
New Identifications of Texts in Armenian Translation
77 An Important Text Preserved in MS Ven.

Mekh. No. 873 Dated A.D. 1299
(Eusebius of Emesa’s Commentary on Historical
Writings of the Old Testament)
95 Severian of Gabala
Fragments of the Aucher Collection in Galata MS 54
107 The Syriac Translation of the Old Testament

– as Evidenced around the Middle of the Fourth
Century (in Eusebius of Emesa)
125 Evidence of the Syriac Bible Translation in Greek

Fathers of the 4th and 5th Centuries
131 The Question of the Syrian Background of the

Early Armenian Church once again
Some Methodological Remarks
141 What Translators Veil and Reveal

Observations on two Armenian Translations of one Greek Homily
155 The Noble Art of Abbreviating
in the Light of some Texts attributed to Severian of Gabala
171 The So-Called “Absurd” Punctuation in John 1,3-4

Neglected Witnesses of the Old Church
187 What was Theodoret’s Mother Tongue?

– Is the Question Open or Closed?
217 Severian Cento No. 2 in MS Galata 54
229 Greek and Syrian

under the Aspects of some Syrian Seats of Learning
243 Indices
Biblical texts
Patristic and medieval names and sources
Modern writers
Abbreviations
ntroduction
I
1. Subject, title, and context
This book is concerned with three bishops from Syria, viz. Eusebius of Emesa
(ca. 300 – ca. 359), Severian of Gabala (? – after 408), and Theodoret of Cyr
rhus (ca. 393 – ca. 466). In their literary activity they altogether cover about
100 years – from the second quarter of the 4th century until the beginning
of the second half of the 5th century. As bishops, of course, their main re-
sponsibilities were of an ecclesiastical nature. However, we shall be more
concerned with their roles as representatives of seats of learning or school
traditions, but it should be mentioned at the very outset that they hardly
drew any sharp distinction between what belonged to School and what
belonged to Church. It should also be emphasized that the state of research
is characterized by a fairly great variation in the scholarly approach of each
of the three authors.
Their “language of office” was no doubt Greek. Eusebius and Severian
probably had Syriac as their mother tongue, whereas Theodoret’s vernacu-
lar, as will be argued below, was Greek. Generally speaking, they all have
some connection with the School of Antioch; since, however, this “school”,
as has been shown most convincingly, was not “monolithic”, and, as will
be true of any “school”, was not “identical” in form throughout the years of
350, 400, and 450, respectively, this “identification” maybe contains far more
open questions than definite answers. It is therefore a basic point that the
three authors should each be considered on the background of their time and
context in history, both theologically speaking as well as in terms of culture,
language etc.
The reason that this collection of articles dating from 1969 to 2005 (2008) has
been given the title “Students of the Bible”, is the fact that we shall mainly be
concerned with the ways in which the three bishops approached the Bible. The
state of research, however, compels us to include questions on the transmission
8 s t u d e n t s o f t h e b i bl e i n 4 t h a n d 5 t h c e n t u r y s y r i a
of their written texts, and other elements that might throw some light over their
theology and exegesis – or on very elementary questions about defining and
understanding their literary remains, all of which must be answered, before
a full evaluation of their roles in the history of learning, religion and politics
can be made.
2. State of research
The starting point of my studies into the three Syrian bishops was a collection
of homilies transmitted in Armenian. I gave a provisional account of some of
my findings in a communication at the Patristic Conference at Oxford in 1967
(published in 1970, not included here), and in my book Per Piscatores (1975)
it was shown that eight of the homilies (I-VIII) belonged to Eusebius, the rest
(IX-XIII) to Severian. I argued that it was possible in some cases to discern the
text of Eusebius’ Syriac Bible; this was an important argument for his author’s
right, but, obviously, it was not a criterion that lent itself easily to the inves-
tigator, as the texts were only transmitted in Latin and Armenian. In fact, the
very intricate question of the transmission of Eusebian texts is one of the main
reasons for using the word “sidelight” in the subtitle of this book.
The fact that Eusebius used the Syrian Bible as the basis for his exegesis
and preaching was the reason for two further considerations, that of the
evaluation of “syriacisms”, not least in the Armenian version of the Bible. I
discussed the principles of method in handling linguistic and text historical
– and other – syriacisms in an article about the Epistle of James (L 1982,1,
below pp. 37ff) and in a broader context in papers concerned with the dis-
cussion about the Syrian background of certain elements in the Armenian
church (L 1989 and 1990, below pp. 125ff and 131ff). Particular examples of
references to Eusebius as a “Christian Syrian” and of Syrian readings in both
Eusebius and Severian were presented at conferences in 1979 and 1995 and
in the jubilee volume of the Venice Mechitharists (L 1981, 1996 and 1969,
below pp. 23ff, 171ff and 13ff).
More specifically, already in Per Piscatores I pointed to the Armenian trans-
lation of Eusebius’ commentary on historical writings of the Old Testament
as an important instrument in identifying Eusebian quotations in the Greek
catenae, whereby considerable parts of Eusebius’ Greek original could be
retrieved. This was pursued in articles published in 1984 and 1987 (below
pp. 77ff and 107ff).
A couple of initiatives to publish a corpus Severianicum in Greek (connected
I n t r o d u c t i o n 9
with the names of C. Datema and K. H. Uthemann) determined to a great

extent the principal concerns of research into Severian. I tried to contribute to
the editorial preparations, mainly along two tracks: first the heuristic task of
identifying further evidence to be gained from the transmission in Armenian.
Here, especially, the exploiting of MS No. 54 of the Galata Collection was
important, but also a couple of Jerusalem manuscripts (No. 1 and No. 154)
could be helpful, not only because of the supplementary texts they contain, but
also in identifying Greek originals in the vast field of Pseudo-Chrysostomica
(L1982,2; 1986 and 2005,1, cf. below pp. 67ff, 95ff and 217ff). Second, I added
some considerations of a methodical character about the particular case where
two translations into Armenian of one homily are extant (L 1993, pp. 141ff).
The Armenian transmission could also be helpful, I argued, in identifying the
author, where we are concerned with abbreviated homilies – in Greek as in
Armenian, but often differently (L 1995, pp. 155ff).
In the 1970s, and even in the 1980s, I had not thought of going very far
into the study of Theodoret myself; and my questions concerning him were
put in very open and provisional terms (as e.g. in L 1987 and 1989, pp. 107ff
and 125ff). The reason for asking my questions was the demonstration that “the
Syrian” as referred to by Eusebius was in Syriac; yet, it could not be neglected
that the famous Old Testament scholar, A. Rahlfs (in 1915) had maintained to
have found in Theodoret a “vollständig sicheren Beweis” that “the Syrian”
was a Bible version in Greek (cf. e.g. below, p. 119).
As it seemed that very few patristic scholars contested the general assump-
tion of Theodoret’s mother tongue being Syriac, and as his use of “the Syrian”
was only approached in a rather superficial way with no definite or clear-cut
answer to the question of the language of this version, I found it necessary
to go a little deeper into Theodoret’s information about Syrian matters, his
knowledge of Syriac, and his use of “the Syrian” (L1999 and 2005,2, below
pp 187ff and 229ff).
3. Occasion, justification, and disposition of the book

It may perhaps be pretentious to publish a collection of one’s own contributions
to a certain field of research, in particular when they cover as long a span of
time as almost 40 years (1969-2005/08).
Of course, in certain cases more recent research has taken the arguments
further than I could do. This is especially true of the identification of the Greek
of Eusebius’ Commentary on Genesis, as accomplished by R. ter Haar Romeny
in his brilliant book A Syrian in Greek Dress (1997). Also Ch. Burchard’s treat-
ment of the Armenian Epistle of James deserves to be mentioned.1
However, even in my articles concerned with these questions there are
still some observations not covered by the writers just mentioned, among
other things on Eusebius’ text outside the Commentary on Genesis; and, also
the discussion of method in handling syriacisms given in my article on the
Armenian Epistle of James might still be relevant, wherefore I have chosen
to include these papers here. (For details, on which e.g. Romeny has rightly
corrected some of my observations, I refer to his book and articles.)
To this, it should be added that a secondary justification for publishing
the present collection of articles in the year 2008 is that quite a few of them
have appeared in periodicals, annuals, Festschriften etc., which might not be
easily available to the patristic scholar, generally speaking. Furthermore, one
article has so far only been published in Spanish (L 1981), two only in Danish
(L 1999 and 2005,2), and one only provisionally advertized as a contribution
to the Weitenberg Festschrift (in press) (L 2005,1).
To such “technical” considerations is added, in my own evaluation, the
idea that it might be worthwhile to assemble the studies under one aspect,
viz. that of the approaches and methods applied by students of the Bible in
4th and 5th century Syria.
This does not, of course, change their character of being modest and sin-
gular studies, the illumination of which on the main subject might often be
described as “sideways” and indirect.
I have chosen not to include any further discussion of recent research, other
than what has been mentioned above and, in particular, what is included in the
latest of the articles (L 2005,2). I have only – in connection with the process of
translating my Danish articles of 1999 and 2005 made the following additions:
material about Theodoret’s Commentary on Jeremiah (pp. 206ff); and an exten-
sion of the concluding remarks about “Greek and Syrian” in later centuries
(pp. 239f).
The studies are presented in the chronological order of their appearance
in the hope that this will furnish the reader with a first hand impression of
how the work was done. It might be appropriate however to add the hope
1 Christoph Burchard, "Zur altarmenischen Übersetzung des Jakobusbriefes", Horizonte der Christen-
heit. Festschrift für Friedrich Heyer zu seinem 85. Geburtstag, hrsg. v. M. Kohlbacher und M. Lesinski
(Oikonomia 34), Erlangen 1994, 195‑217.
I n t r o d u c t i o n 11
that this will not be too inconvenient to any reader who, for example, might
be primarily interested in just one of the figures treated of.
Finally, in this sequence of “expressions of hope” I also cherish another
hope, i.e. that I shall not be the only person to find this collection of chapters
– in direct reprint or translation – worthwhile, presented here, as it were, in
“new clothing”.
Nourishing such a hope I offer this collection of studies about “Students
of the Bible in 4th and 5th century Syria” to students of patristic, biblical and
armenological matters – with cordial thanks to the Carlsberg Foundation,
which has – in financial terms – made it possible to publish the book.
References
Below, references to the original time and place of publishing of the studies are given. Supplementary
information is added, where this is considered necessary.
1. “Hosanna – A Philological Discussion in the Old Church”, Armeniaca – Mélanges d’études armènien-
nes, Venice 1969, 165-174. (L1969)
2. “The Spirit of God upon the Face of the Waters. The Sources of St. Basil’s and St. Augustine’s
Comments on Gen 1,2c” (orig. “El Espíritu de Dios sobre las aguas. Fuentes de los comentarios de
Basilio y Agustín sobre el Génesis I,2”, Augustinus XXVI, Madrid 1981, 127*-139*). (L1981)
3. “Some Questions Concerning the Armenian Version of the Epistle of James”, Aarhus Armeniaca, eds.
Lise Bek, Henning Lehmann, and Lars Kærulf Møller (Acta Jutlandica LVII, Humanities Series 56),
Århus 1982, 57-82. (L1982,1)
4. “Severian of Gabala: New Identifications of Texts in Armenian Translation”, Classical Armenian
Culture, ed. Th. J. Samuelian (University of Pennsylvania Armenian Texts and Studies 4), University of
Pennsylvania 1982, 113-124. (L1982,2)
5. “An Important Text Preserved in MS Ven. Mekh. No. 873, dated A.D. 1299 (Eusebius of Emesa’s
Commentary on Historical Writings of the Old Testament)”, Medieval Armenian Culture, eds. T.
Samuelian & M. Stone (University of Pennsylvania Armenian Texts and Studies 6), Chico, CA 1984,
142-160. (L1984)
6. “Severian of Gabala: Fragments of the Aucher Collection in Galata MS 54”, Armenian Studies / Étu-
des Arméniennes in Memoriam Haïg Berbérian, ed. Dickran Kouymjian, Lisbon 1986, 477-487. (L1986)
7. “The Syriac Translation of the Old Testament – as Evidenced around the Middle of the Fourth Cen-
tury (in Eusebius of Emesa)”, Scandinavian Journal of the Old Testament 1, Århus 1987, 66-86. (L1987)
8. “Evidence of the Syriac Bible Translation in Greek Fathers of the 4th and 5th Centuries”, Studia Pa-
tristica XIX, ed. E. A. Livingstone, Leuven 1989, 366-371. (L1989)
9. “The Question of the Syrian Background of the Early Armenian Church once again. Some Methodo-
logical Remarks”, Studia Patristica XVIII, 4, ed. E. A. Livingstone, Kalamazoo-Leuven 1990, 255-262.
(L1990)
10. “What Translators Veil and Reveal. Observations on two Armenian Translations of one Greek Ho-
mily”, Armenian Texts – Tasks and Tools, eds. Henning Lehmann and J. J. S. Weitenberg (Acta Jutlan-
dica LXIX:1, Humanities Series 68), Århus 1993, 75-84. (L1993)
11. “The Noble Art of Abbreviating – in the Light of Some Texts Attributed to Severian of Gabala”, Pro-
ceedings of the Danish Institute at Athens I, ed. Søren Dietz, Athens 1995, 221-227. (L1995)
12. “The So-called “Absurd” Punctuation in John 1,3-4. Neglected Witnesses of the Old Church”,
Proceedings of the Vth International Congress on Armenian Linguistics, ed. D. Zakayan, Montreal 1996,
45-62. (L1996)
13. “What was Theodoret’s Mother Tongue? – Is the Question Open or Closed?” (orig. “Theodorets
modersmål – et åbent eller lukket spørgsmål?”, Ordet og livet. Festskrift til Christian Thodberg, red.
Carsten Bach-Nielsen, Troels Nørager & Peter Thyssen, Århus 1999, 43-65). (L1999)
14. “Severian Cento No. 2 in MS Galata 54”, Festschrift J. J. S. Weitenberg, ed. Theo van Lint (2005/in
print). (L2005,1)
15. “Greek and Syrian – under the Aspects of some Syrian Seats of Learning” (orig. “Det græske og det
syriske – især set fra nogle af de antiokenske skolers katedre”, Et blandet bæger. Studier tilegnet Finn
O. Hvidberg-Hansen, red. Pernille Carstens, John Møller Larsen, Dorthe Maria Kodal & Dan Enok
Sørensen, København 2005, 146-160). (L2005,2)
Studies Nos. 2, 3, 4, 5, 8, 9, 10, 12 are contributions to various conferences and workshops. In most cases,
therefore, they have been written a couple of years before the date of publication. No. 9, though, was
given as a “short communication” at the Patristic Conference at Oxford in 1983 – 7 years before the year
of publication. (It is thus “earlier” than No. 8).
For study No. 2 the text given below is the original English version as prepared for the Patristic
Conference at Oxford in 1979.
Study No. 14 is quoted as belonging to the year 2005, because that was the year it was delivered as
a contribution to a Festschrift (for J. J. S. Weitenberg), which has not yet appeared in printed form.
Studies Nos. 13 and 15 have been translated from Danish in 2008. In the case of No. 13, I have
added a paragraph on Theodoret’s Commentary on Jeremiah (below, pp. 206ff).
Apart from minor corrections (and the additions mentioned above), the studies are otherwise pre-
sented in their original shape. References to my own studies republished in this book are given in the
short form: L + year.
The original texts have been scanned by Narayana Press. As it turned out to be difficult to scan ele-
ments in Armenian, Greek, Hebrew and Syriac, and as, originally, in some cases varying transcriptions
of the alphabets of these languages had been used, it was decided to standardize these elements, which
meant that they had to be "rewritten". This work was done by my wife, Else Lehmann, who thus had
the great – or questionable (?) – pleasure of working on the manuscripts once again, as she had done
20 or 40 years ago. On technical questions good advice and assistance was given by colleagues such as
Jørgen Ledet Christiansen and Aage Pilgaard, and not least by Jørgen Friis Bak of the Data Office of the
University of Aarhus. The newly translated chapters were checked by Mary Waters Lund, and the bulk
of work in planning, printing and editing the volume was carried out by Henriette Møller and Elsebeth
Morville of Narayana Press and Katja Teilmann of the Aarhus University Press – supervised by Claes
Hvidbak and Jørgen Sparre. I owe all of them cordial thanks.
osanna
A Philological Discussion in the Old Church
H
No. 11 of the 15 Armenian homilies, edited in 1827 by J. B. Aucher under the
name of Severian of Gabala1, ends as follows: “and him, whom angels glorify
with awe in heaven, the children also praised with great joy on earth, saying
‘osanna’; in the language of the Hebrews this really means: glorification and
great praise”2. This homily, edited by Aucher from a manuscript in the mon-
astery of S. Lazzaro3, is also found in a Parisian manuscript4 with quite insig-
nificant variants5 as far as the section treated of here is concerned (as, indeed,
on the whole). What is more interesting, is that the same homily was edited
a few years ago in a considerably longer version and now under the name
of Eusebius of Emesa6. I do not intend here to expatiate upon the problem of
1 Severiani sive Seberiani Gabalorum episcopi Emesensis homiliae nunc primum editae ex antiqua versione
armena in latinum sermonem translatae per P. Jo. Baptistam Aucher, Venice 1827.
2 Ed. Aucher, p. 408: Եւ զոր հրեշտակք փառաւորեն յերկինս ահիւ, աւրհնէին եւ մանկտին ցնծալով
յերկրի, ասելով ովսաննա, այս է ըստ եբրայեցւոց բարբառոյն փառաբանութիւն իսկ նշանակեալ,
եւ մեծ գովութիւն, in Aucher’s Latin translation (op. cit., p. 409): et quem Angeli glorificant in caelo
cum timore, benedicebant et pueri exultantes in terra, dicentes Hosanna, quod secundum Hebraicam linguam
verae glorificationis est significatio, magnaeque laudis.
3 Referred to by Aucher (op. cit., p. 402 f) as ճառընտիր no. 13, p. 287. Unfortunately, Sargisean’s
catalogue of the manuscripts in the monastery of St. Lazzaro has not been accessible to me during
the preparation of the present study, but it seems to appear from Akinian’s note (Handes Amsorya,
73, 1959, col. 321f) that Aucher’s manuscript is identical with MS Ven. Mech. 212 (p. 356ff) in Sar-
gisean’s numbering (cf. according to Akinian: Sargisean, Catalogue, vol. II, p. 226).
4 MS Par. Arm. 110, fol. 314 r, col. 1 – fol. 315 r, col. 2.
5 As compared with Aucher’s text (see above, note 2), the variants are the following: յերկինս:
ի բարձունս, – ասելով: եւ ասէին, – փառաբանութիւն: փառաւորութիւն, – իսկ: om., – post
գովութիւն: add. այնմիկ որ միշտ աւրհնեալն է յաւիտեան: ամէն: (to him who is the eternally
praised for ever and ever. Amen).
6 Եւսեբեայ Եպիսկոպոսի Եմեսացւոյ ճառք – Die Reden des Bischofs Eusebius von Emesa, ed. by N.
Akinian in the Handes Amsorya, 1956-1959, on the basis of the Armenian manuscript no. 110 from
New Julfa. The homily treated of here is the last of the 13 numbers of the series (HA 73, 1959, col.
319-360).
authorship which I hope to return to on a later occasion7. Here, too, the variants
in the part of the text covering Aucher’s edition are unimportant8; however,
the continuation offers the following surprising and interesting comment on
the meaning of ‘hosanna’: “(the truth is) not, as some have thought, that it is a
word consisting of two halves, namely partly ‘os’ and partly ‘anna(j)’, but this
word has its own proper interpretation, and it is not, as (if) some one might
say that ‘os’ and ‘anna’ mean ‘(some) glory’ and ‘man’, but the word ‘osanna’,
pronounced as one word is (to be) translated by ‘greatness’. As (people say)
in our tongue: ‘greatness in the highest’, thus the language of the Hebrews
has: ‘osanna(j) in the highest’”9.
In view of the prominent place in the liturgy of the Old Church which
was given to the word ‘hos(i)anna’ on the basis of the gospels and Ps. 118
(LXX: 117), it is no wonder that the question has been put what this word
really meant. It will not be appropriate here to list all references to Ps. 118,
25, where the exegetes and preachers of the Old Church follow unreflectingly
the LXX translation of the Hebrew: ‫ הושיעה־נא‬into σῶσον δὴ, but I hope to be
able to give, if not a complete survey10, at least references to some of the most
important texts that are of any relevance to the understanding of the Armenian
evidence submitted above, and its possible place in a definite tradition.
To my knowledge the earliest evidence that is of any interest in this con-
7 In a communication delivered in Oxford at the Fifth International Conference on Patristic Studies in

September 1967 (to be published in Studia Patristica in the series Texte und Untersuchungen zur Ge-
schichte der altchristlichen Literatur) I have identified the Greek originals of the homilies 9, 11, and 12
of the series published by Akinian (hom. 9 = PG 63, 543-550; hom. 11 = PG 59, 653-664; hom. 12 = PG
48, 1081-1088) and shown that these three homilies do belong to Severian of Gabala, a fact which
will have to be taken into account when examining the authorship of the homily no. 13 treated of
here.
8 Cf. note 2: փառաւորեն յերկինս ահիւ: ահիւ փառաւորէին ի բարձունս, – եւ մանկտին ցնծալով:
3 – 1 – 2, – ովսաննա: “ովսաննա որդւոյ Դաւթի”: Ովսաննասն, – փառաբանութիւն … գովութիւն:
փառաւորութիւն իմն նշանակել (cf. Akinian’s ed., HA 73, 1959, col. 333-336, l. 223-227).
9 Ed. Akinian, HA 73, 1959, col. 335-336, l. 227-232: ո՛չ որպէս ոմանք կարծեցին, եթէ բան ինչ կէս
իցէ. այլ իմն ովսն եւ այլ իմն աննայն. այլ իւր առանձինն մեկնութիւն է բանիս եւ ոչ է ովս եւ աննա
որպէս ոք զի ասիցէ, եթէ փառք իմն եւ մարդ, այլ միասաց բանն ովսաննա մեծութիւն թարգմանի.
որպէս ի մեր լեզուս մեծութիւն ի բարձունս, յԵբարյեցւոց բարբառոյն ասի “ովսաննայ ի
բարձունս”. (I have taken the liberty of making a minor alteration in the punctuation, and regarded
ան, նա (l. 229) and ովսա նա (l. 230) as misprints or errors by the scribe. As for միասաց one might
consider the possibility of changing into: -ած, -ացեալ (or միասաց բանն into միասացական[ն]);
there can be no doubt, however, about the general meaning of the sentence).
10 Cf. F. Vigouroux’s art. “Hosanna” in Dict. de la Bible, which gives the best, if still imperfect, survey
of patristic references.
H o s a n n a 15
nection, is to be found in Clement of Alexandria’s Paidagogos, I,V,12,511, where

– after quoting the hosanna-verse in a form that combines the versions of
Matth. and John12 – he adds: “light and glory and praise with supplication to
the Lord, for this is what ‘hosanna’ means translated into Greek”13. In the criti-
cal apparatus Stählin refers to a parallel tradition in the Nicetas catena which
about John 12, 13 under the double lemma: Άναστασίου. Κλήμεντος has the
following: Τò δὲ ὡσαννὰ ἑρμηνεύεται δόξα καὶ αἷνος, ὡς εἷναι τοιοῦτον τò
λεγόμενον “χάρις καὶ δόξα τῷ υἱῷ Δαβίδ”. ὅθεν ὁ ἅγιος Λουκᾶς ὥσπερ
ἑρμηνεύων τὸ “ὡσαννὰ ἐν τοῖς ὑψίστοις” ἀντὶ τούτου “δόξα ἐν ὑψίστοις”
εἶπεν ἐν τῷ κατ’αὐτὸν εὐαγγελίῳ. τινὲς δὲ τὸ ὡσαννὰ σημαίνειν λέγουσιν
τὸ “σῶσον δή”.
Zahn comments on this text: “Da die benützte Stelle des Anastasius nicht
zu finden ist, in welcher vielleicht Clemens angeführt war, so ist der Antheil
des Clemens nicht zu bestimmen”; however, Zahn, too, refers to the above-
11 Ed. Stählin, GCS 12, p. 97.

12 Matth. 21, 8 f; John 12, 13.
13 L. cit. (see note 11): φῶς καὶ δόξα καὶ αἶνος μεθ’ ἱκετηρίας τῷ κυρίῳ· τουτὶ γὰρ ἐμφαίνει
ἑρμηνευόμενον Ἑλλάδι φωνῇ τὸ ὡσαννὰ. One could, of course, raise the question whether
Clement has intended here to give a precise translation or etymology of the word “hosanna” or
has merely wanted to reproduce the content and meaning, in a wider and vaguer sense, of the
shout with which Jesus was met on his entry into Jerusalem. In my view the former alternative is
correct, firstly because Clement’s usage definitely points in that direction – τουτὶ γὰρ ἐμφαίνει
ἑρμηνευόμενον Ἑλλάδι φωνῂ not being a formula introducing a vague and paraphrastic repro-
duction of content – and secondly because such etymologies in the service of exegesis seem to have
been widely used in the Alexandrian tradition. Cf. Ilona Opelt, art. “Etymologie”, RAC (see esp.
col. 826 f), and F. Wutz, Onomastica Sacra, TU 41,1, 1914 (see e. g. pp. 1 and 13ff about Philo, and p.
179 for the following statement about the so-called “Origenesgruppe”: … bestanden die Vorlagen
des Hieronymus aus einzelnen Listen, die alle der OGr. [“Origenesgruppe”] zugehörten. Die OGr.
selbst hat von der Exegetenschule in Alexandrien ihre Pflege und Ausgestaltung erhalten…). It
should be added, perhaps, that in the present connection the decisive feature is the occurrence of
δόξα in Clement, and that the above says nothing about Clement’s possible source(s) or about the
question whether his choice of words is due to an attempt to specify the content of the Hebrew term
by means of several Greek words (which is still something quite different from a paraphrasis of the
account of the entry, and in which case such late Judaic doxa-light-speculations as S. Aalen has dis-
cussed in his book Die Begriffe “Licht” und “Finsternis”…, Oslo, 1951 [see esp. p. 195ff], should prob-
ably be considered as the background, if not as the immediate source), or if it should be understood
as an actual multiple “etymology” (as e. g. – be it suggested with all possible reservation ‫( עז‬αἶνος,
cf. Ps. 8, 3) + ‫( זיו‬φώς) [in Biblical Aramaic e. g. Dan. 2, 31; 4, 33 – cf. the use of this word in the quota-
tion from Syr Bar 51, 10, S. Aalen, op. cit., p. 199, n. 4] + ‫( חן‬δόξα) + ‫( נא‬μεθ’ ἱκετηρίας, cf. Jerome:
interiectio deprecantis, see below)). On the possibility of an etymology of this kind, cf. Wutz., op. cit.,
pp. 530 ff, 355 ff and elsewhere, and for several multiple etymologies in Philo, cf. C. Siegfried, Die
hebräischen Worterklärungen des Philo, Magdeburg, 1863.
quoted passage from the Paidagogos as the closest parallel in the known text
material14.
Whatever Clement’s share in the latter, both texts referred to above agree
in giving δόξα and αἶνος as the translation of ὡσαννὰ and thus correspond
to the Armenian translation which we met in the short version, and to the
translation of the first part of the word which was rejected in the continuation
of the passage in question in the long version, and moreover the catena frag-
ment points to the probable source of this interpretation, namely St. Luke’s
account of the entry of Jesus into Jerusalem where we do not find the shout
of hosanna, but instead: δόξα ἐν ὑψίστοις15, even though we must agree with
Zahn that it is not possible to know if Clement already based his translation
on this synoptic parallel, or whether Anastasius was the first to do so.
As might be expected, the most learned exposition concerning “hosanna” is
to be found in Jerome, celebrated, as it will be known, as vir trilinguis. His 20th
letter to Pope Damasus16 is a learned treatise on this very subject and an answer
to Damasus’ question about it17. He begins with a sharp repudiation of Hilary
who, in his commentary on Matthew, translated “osanna” by “redemptio
domus Dauid”18 though (understood: what any Old Testament exegete ought
to know) “redemptio” is “pheduth”, and “domus” is “beth” in Hebrew. And
Jerome proceeds: “alii opinati sunt “osanna” “gloriam” dici – porro “gloria”
“chabod” appellatur –, nonnulli “gratiam”, cum “gratia” “thoda” siue “anna”
nuncupetur”19. In other words: the interpretation we have met in Clement
(gloria = δόξα) is rejected here, and as for gratia it will be remembered that in
the catena fragment the word pair χάρις καὶ δόξα was found, too20. I shall
14 Th. Zahn, Forschungen zur Geschichte des neutestamentlichen Kanons und der altkirchlichen Literatur, III,
1884, p. 52.
15 Luke 19, 38.
16 Ed. Hilberg, CSEL 54, 1910, p. 104-110.
17 Op. cit., ep. XIX, p. 103-104.
18 A corresponding interpretation is found in Ambrose, Exp. in Luc. (PL 15, 1888 D): Hosanna filio
David, hoc est, redemptorem domus David exspectatum quoque secundum carnem David filium venisse de-
clarat …, which occasions the following “Ehrenrettung” in Migne (l. cit., note 17): Ambrosium arguit
Erasmus, quasi hic cum Hilario Pictav. parum erudite doceat voce Hosanna, redemptionem domus David sig-
nificari. Sed parum arguta est ea correctio; non enim de grammatica illius vocabuli significatione agit sanctus
doctor, cum hoc unum indicat, a populo faustis ac propheticis acclamationibus adventum redemptoris domus
David esse declaratum. Cf. below, note 30 (Jansens).
19 Op. cit., p. 104.
20 Of interest, here, is the differentiation in the Onomastica literature where, according to Wutz, ‫חן‬, ‫חןה‬
in the so-called “vaticanische Gruppe” is always translated by δόξα, in the so-called “philonisch-
origenianische Gruppe” always by χάρις. Wutz, op. cit., p. 106 f.
H o s a n n a 17
not go into details in Jerome’s further exposition which he substantiates by

referring to the form of Ps. 118, 25 in all the columns of the Hexapla, but only
point to his statement that osanna is a corrupt form of the correct “osianna” (he
draws a parallel to the elision of vowels in the recitation of Latin poetry), and
that he chooses the translation saluifica or saluum fac (as it will be known from
the Vulgate), pointing out that “osi” is the actual imperative, “anna” being
an interjection expressing the mood of the worshipper: interiectio deprecantis21.
With a view to the evaluation of the next Latin text to be treated of, it should
be mentioned perhaps that here Jerome also parallels with Latin interjections
and their function as expressive of emotional moods: ut in exultando dicamus
“ua” et in admirando “papae” et in dolendo “heu” et, quando silentium uolumus
imperare, strictis dentibus spiritum coartamus et cogimus in sonandum “st”, ita et
Hebraei …22.
The next Latin witness is Augustine who discusses the question in almost
identical words in two places, namely in De doctrina Christiana23 and Tractatus
in Iohannis evangelium24. There is a close connection with Jerome’s account of
the use of interjections but for the one decisive difference that to Augustine the
whole word “hosanna” has become an interjection. The section in question of
De doctrina Christiana treats of the advantage of knowing Hebrew and Greek
in order to enable oneself to check passages where discrepant formulations by
Latin translators and interpreters cause doubt about the proper understanding.
As for the untranslated words like amen, alleluia, racha, and osanna Augustine
adds that the two first-mentioned are left untranslated propter sanctiorem auc-
toritatem, the latter two because they are untranslatable. There are in fact such
words as cannot be translated from one language into another, he says, and
concludes: Et hoc maxime interiectionibus accidit, quae verba motum animi signifi-
cant potius quam sententiae conceptae ullam particulam. Nam et haec duo talia esse
perhibentur; dicunt enim racha indignantis esse vocem, osanna laetantis.
In his comment on John 12, 12 f in Tractatus in Ioh. ev., Augustine is even
closer to Jerome’s statement about the interjections. The passage runs: Vox au-
tem obsecrantis est, Hosanna, sicut nonnulli dicunt qui hebraeam linguam nouerunt,
magis affectum iudicans25, quam rem aliquam significans; sicut sunt in lingua latina
21 L. cit., p. 109.
22 L. cit., p. 108f.
23 II, 34-35, ed. Green, CSEL 80, 1963, p. 43.
24 24. LI, 2, ed. Willems, CCSL 36, 1954, p. 440.
25 A likely conjecture would be: indicans pro iudicans.
quas interiectiones uocant, uelut cum dolentes dicimus: Heu! uel cum delectamur: Vah!
dicimus; uel cum miramur, dicimus: O, rem magnam! tunc enim, O, nihil significat,
nisi mirantis affectum. Quod ideo credendum est ita esse, quia neque graecus, neque
latinus hoc interpretari potuit, sicut illud: Qui dixerit fratri suo, Racha. Nam et haec
interiectio esse perhibetur, affectum indignantis ostendens.
As the last Latin witnesses on the discussion of hosanna must be mentioned
Eucherius of Lyons who without any doubt builds on Jerome26, and the anony-
mous Opus imperfectum in Matthaeum, handed down under the name of John
Chrysostom, but attributed by G. Morin to an Arian bishop in Northern Italy
about the year 550. The work is interesting in this connection because it knows
of three of the interpretations listed by Jerome; the author of this work, how-
ever, does not intend to undertake any scientific selection, but instead combines
the possibilities referred to into an edifying harmony. The passage is found
in the 37th homily27: Hosianna quidem interpretantur gloriam, alii redemptionem,
alii Salvifica, sive Salvum fac. Nam et gloria illi debetur, et redemptio illi convenit,
qui omnes redemit, et pretiosi sanguinis effusione salvavit.
In continuation of this review of the Latin contributions to the patristic
discussion on the meaning of hosanna, it may be appropriate to mention that
16th century scholars like Cornelius Jansens (Jansenius) the Elder (1510-1574)
and Leo de Castro (d. 1589) took the Old Church discussion as their starting-
point, not least through their attitude to Jerome. The reference to these authors
which I owe to a note in Migne’s edition of the above-cited section of Clement
of Alexandria’s Paidagogos28, is of course primarily of interest for the history
of scholarship29, but Jansens in particular has a further significance in that he
26 Eucherius, Instructiones, liber II (ed. Wotke, CSEL 31, 1894, p. 145: PL 50, col. 814): Osanna saluifica
siue saluum fac. Osia (PL: Osi) enim saluifica interpretatur, anna interiectio est deprecantis, ergo integre
dicitur Osianna; sed dum corripitur aut corrumpitur, sonat Osanna; est autem sensus: saluum fac, ut sub-
audiatur uel populum Israhel uel totum mundum. (For the last passage cf. Jerome, ep. XX, ed. Hilberg,
CSEL 54, p. 110). Ilona Opelt’s article, “Quellenstudien zu Eucherius”, Hermes 91, 1963, p. 476-483,
which to Altaner’s summary “nach Hieronymus” (Patrologie, 6. ed., 1960, p. 419) points out that Je-
rome is not Eucherius’ only source, and that where he does make use of Jerome, he shows a marked
preference for the New Testament onomasticon, does not deal with Eucherius’ osanna-interpretation.
27 PG 56, col. 838.
28 PG 8, col. 264, n. 50.
29 Even though it will probably be impossible to define what Old Church sources they are based
on, two conflicting explanations from the 11th century are also of interest in this perspective,
namely Suidas’ lexicon (ed. Adler, vol. III, 1933, p. 624): Ὡσαννα: δόξα σημαίνει. καὶ γὰρ ἄλλος
εὐαγγελιστὴς λέγει· είρηνη τῷ υἱῷ Δαβίδ, καὶ ὁ ἕτερος· δόξα τῷ υἱῷ Δαβίδ. ὥστε τὸ ὡσαννὰ τὸ
δόξα σημαίνει: παρά τισι δὲ σῶσον δὴ. οὐκ ὀρθῶς, and Theophylactus (In Matth., PG 123, col. 369
H o s a n n a 19
advances an interesting etymology for the word osanna. De Castro’s line of

argument is to emphasize the authority of Luke, the evangelist, claiming that
the linguistically gifted Luke would also know Hebrew, for which reason the
translation gloria rests on an old and linguistically sound tradition. Jansens
follows a different direction when in his principal work Concordia Evangelica
he comments on the rejection of Hilary and Ambrose by Jerome, saying that
these two commentators should not be understood as if they believed to have
translated the word “hosanna” by “redemptio”, but rather that with this word
they expressed the contents of the people’s prayer to Jesus30. He then proposes
the etymology mentioned above, saying: “quod apud omnes Evangelistas
per duas dictiones scribatur ὡς ἀννὰ31 dictio autem ὡς si Hebraice ita scri-
batur ‫ עז‬laudem significare possit: ἀννὰ autem, si scribatur ‫ חנה‬significare
gratiam …”.
Whatever the difference, recent philologists have been working on the
same problems as Jerome. Thus the hypothesis has been advanced that the
last syllable of hosanna was to be the Hebrew 1. pl. suffix, and it would then
acquire the meaning serva nos32. Dalman33 rejects this by referring partly to the
fact that this interpretation is not substantiated in Jewish literature, partly to
the impossibility of having a dative attached, as it happens in Matth. 21,9.15
(τῷ υἱῲ Δαυείδ). Zahn34 reproduces Dalman’s arguments35, adding: “Drittens
aber gibt es im Aram. ein Verbum ‫ ישע‬oder ein damit stammverwandtes gar
nicht36. Das Hiphil dieses hebr. Verbs wird in Targ. und Pesch. regelmässig
durch prq wiedergegeben”. And Zahn continues: “Das talmudische ‫הושיעהנא‬
als Name des Lulab und (mit jom) des letzten Laubhüttenfesttags, welches in
C): Τὸ δὲ Ὡσαννὰ, οἱ μὲν λέγουσιν ὃτι ὕμνον ἢ ψαλμὸν σημαίνει, οἱ δὲ, ὃ καὶ ἀκριβέστερον·
Σῶσον δή.
30 … non … eos existimasse haec omnia comprehensa in significatione vocabuli Hosanna, sed populum ejusmodi
acclamatione petiisse redemptionem (quoted from Migne, l. cit. – cf. above, note 18).
31 As it is the case in Migne’s edition of Clement, which is the occasion of the note quoted here. PG 8,
col. 264.
32 Cf. E. Kautzsch, Grammatik des Biblisch-Aramäischen, Leipzig 1884, p. 173 (with references to Sieg-
fried, Hilgenfeld, Anger, and Merx)
33 G. Dalman, Grammatik des jüdisch-palästinischen Aramäisch, 2. ed., Leipzig 1905, p. 249, n. 1
34 Th. Zahn, Einleitung in das Neue Testament I, Leipzig 1906, p. 14 f.
35 Zahn, however, refers to the old translations of the Bible and the commentators of the Old Church,
rather than to Jewish literature.
36 It must therefore be regarded as unfortunate that Kautzsch (l. cit.) vocalizes the word, as if it was
an imperative of the non-existent verb (even in his reference to Payne Smith, Thes. syr., where this
vocalization does not occur). But it may perhaps be a question of a simple misprint?
der Form ‫ ܐܘܫܥܢܐ‬als Fremdwort zu den Syrern übergegangen und auf den
Palmsonntag ubertragen worden ist (Payne Smith Thes. s. v.), kann ja nichts
anderes sein, als eine zusammengezogene Aussprache des hebr. Ausrufs,
welche die Evv und die Liturgie der Didache37 genau wiedergegeben haben”.
And Zahn concludes his examination of the problem by saying about initial
h or ᾽ (spiritus asper or lenis) that it is impossible to decide “mit wie starkem
Anhauch am Palmsonntag die Einen und die Andern in dem gemischten
Volkshaufen die erste Silbe dieses Wortes ausgerufen haben”.
If, however, we return to our starting-point, the long recension of the Ar-
menian homily, I have no doubt that its author has understood the translation
rejected by him of the first part of “osanna” from the same etymological basis
as the one we met in Cornelius Jansens, namely “os” = Hebrew ‫( עוז‬root ‫)עזז‬,
Syriac ‫( ܥܘܙ‬root ‫)ܥܙ‬. For the second part of the word, however, the anonymous
author of the etymology does not recur to any form of ‫ חנ‬or ‫( חנן‬Syriac noun
‫ܚܢܢ‬, root ‫)ܚܢ‬, as did Jansens, but offers the translation “man”, that is he refers to
Hebrew ‫אנוש‬, Aramaic ‫אנש‬, or Syriac ‫ܐܢܫ‬, ‫ܐܢܫܐ‬. How the connection between
the two parts of the word has been conceived of, we are not told, but as for the
linguistic form it should be pointed out here that the form “osannas” with a
final “s” does seem to appear in the Armenian text, namely in the beginning
of the actual explanation of the word38. When it is not found 2 and 3 lines
later (and elsewhere in the Armenian text) judging from Akinian’s edition,
it is a conceivable explanation that the scribe has retained the “s” in the first
case in the belief that it was the acc. pl. mark, and then not only forgotten to
correct this form after the construction of the sentence, but also, in the subse-
quent passages where the word “osanna” reappears, changed to the normal
and in his view “correct” form. I shall not, however, go into particulars as far
as this question is concerned. Only a closer palaeographic study of the New
Julfa-manuscript could reveal whether it might contain details that have been
overlooked, and which might make it possible to get beyond these hypotheses
on this point.
The author of the homily now replaces the rejected explanation with an-
other, namely that osanna means մեծութիւն = greatness. The question is then
whether there is also a popular (or “erudite”?) etymology behind the latter
37 Did. 10, 6.
38 Ed. Akinian, l. cit., l. 226. Incidentally, to Akinian’s critical note to l. 225, one is tempted to ask if the
indistinct ն which the ms. is supposed to have here, should not have been an ս, in which case the
form with final s would occur in two places.
H o s a n n a 21
explanation, as well as behind the one he just rejected. It seems to me the an-
swer must be affirmative, and that the Syriac word which he takes for his basis
is: ‫( ܥܘܫܢܐ‬from the root ‫ )ܥܫܢ‬to which J. Payne Smith (Margoliouth)’s Syriac-
English Dictionary gives the translations: strength, force, power, multitude;
strong current or swelling of a river; a stronghold.39 It is highly probable that
this word is easily confused with or connected with ‫ܐܘܫܥܢܐ‬, and it may be
safe to say that what Zahn said about the transition between ᾽ and h40; must
apply with all the more reason to ‫ א‬/ ‫( ܐ‬alef) and ‫ ע‬/ ‫( ܥ‬ayin), resp.41.
In my view both etymologies point to Syria as the place of origin, as they
do not necessarily imply any wide knowledge of Biblical Hebrew, but rather
some familiarity with Syriac, particularly in the case of the latter, as the word
‫ ܥܘܫܢܐ‬and the root ‫( ܥܫܢ‬in this sense42) are not found in Biblical Hebrew. The
more interesting therefore to find that the latter etymology is also substanti-
ated in Greek, and even in a text which points in the same direction as far as
the place of origin is concerned. The text in question is the pseudo-Justinian
Quaestiones et responsiones ad orthodoxos, where question no. 63 asks about the
meaning of the words hallelujah and osanna, and where the latter is explained
as μεγαλοσύνη ὑπερκειμένη43. As for the origin of this text, scholars disagree
on the problem of authorship; Harnack44 wanted to attribute it to Diodore of
Tarsus, but was sharply opposed by F. X. Funk45. After Funk’s refutation of
Harnack and after the studies of Lebon46 and Richard47 the prevailing view
seems to be that the author is Theodoret of Cyrrhus48. In any case, the names
of both authors point towards the Antiochene tradition, and so do the names
with which the homily has been connected: Eusebius of Emesa and Severian
39 Op. cit., p. 408. For text references see Payne Smith, Thes. syr., vol. II, col. 3004.
40 L. cit., cf. above, note 34.
41 Cf. moreover Wutz’s section on “Wechsel von Gutturalen”, op. cit., p. 355 ff.
42 According to Gesenius-Buhl, Hebr. u. Aram. Handwörterbuch über das AT, 44.1949, p. 626, the root ‫עשן‬
in Biblical Hebrew only has the meaning of “smoke”, corresponding to Syriac ‫ܬܢ‬. For the derivation
of Syriac ‫ ܬܢܢܐ‬from Aramaic *‫ עתן‬Gesenius-Buhl (s. v. ‫ )עשן‬refers to Hoffmann, Literarisches Central
blatt, 1882, p. 320.
43 PG 6, col. 1296, German translation in Harnack, Diodor von Tarsus. Vier pseudojustinische Schriften als
Eigentum Diodors nachgewiesen, TU 21, 4, 1901.
44 See the work, mentioned in the previous note.
45 F. X. Funk, Kirchengeschichtliche Abhandlungen und Untersuchungen, 3, 1907, p. 323 ff.
46 J. Lebon, “Restitutions à Théodoret de Cyr”, RHE 26, 1930, p. 523-550, see esp. p. 540, n. 4.
47 M. Richard, “Les citations de Théodoret conservées dans la chaîne de Nicétas sur l’évangile selon
Saint Luc”, RBibl 63, 1934, p. 88-96, see esp. p. 92, n. 1.
48 Cf. e.g. Altaner, Patrologie, 6. ed., p. 305.
of Gabala. A conclusive settlement of the authorship of the homily – whose

Severianic origin (as far as the then known short version is concerned) has
been disputed by Dürks49 and after him by Zellinger50, and whose attribution
to Eusebius appears to me to present serious difficulties – cannot be made on
the basis of the etymology treated of here or the connection with the prob-
ably Theodoretic Quaest. et resp. (the occurrence of the etymology referred to
above is not the only link between the two texts), even if both facts will have
to be considered for that purpose. What might be said, provisionally, is that
the detail which it has been attempted to elucidate here, has presented to us
a stone of the mosaic of the Antiochene school tradition, a stone which – with
all its insignificance – may be regarded as illustrative of the ways by which it
was attempted, within this tradition, to reach an understanding of the Scrip-
tures.
49 G. Dürks, De Severiano Gabalitano, Kiel 1917, p. 60-64.

50 J. Zellinger, Studien zu Severian von Gabala, Münsterische Beiträge zur Theologie, 8, Münster i. W. 1926,
p. 71.
The Spirit οf God upon the
Face οf The Waters
The Sources of St. Basil’s and St. Augustine’s
Comments on Gen 1,2c
For obvious reasons exegetes and theologians of the Old Church – like exegetes
and theologians of later centuries – attached great importance to the Mosaic
Creation Account of Genesis 1-2.
Thus also, among other exegetical questions, it was of importance to know
whether the final phrase of verse 2 in Genesis, chapter 1, should be translated:
“and the Spirit of God moved upon the face of the waters” (to quote traditional
usage within the English translations of the Bible), or whether it should be
rendered: “and a mighty wind (that) swept over the surface of the waters”
(to quote the New English Bible), and what was the content and meaning of
the phrase in either case1.
I shall begin my modest contribution to the elucidation of some authors
of the Old Church by briefly summarizing some of the views on Gen 1,2c
contained in St. Augustine’s exegetical works on Genesis. I leave aside his
extensive use of Gen 1 in e.g. the Confessiones or the City of God2. Referring to
1 For the history of exegesis the most important work is: In Principio. Interprétations des premiers
versets de la Genèse, (Études Augustiniennes), Paris 1973 (quoted: In Principio). A useful survey of
the principal trends in patristic exegesis of the Creation Account is given in E. Testa, “La creazi-
one del mondo nel pensiero dei SS. Padri”, Studii biblici franciscani, Liber annuus, XVI, 1965-66
(Jerusalem, 1966), 5-68. An older, comprehensive survey of the history of exegesis for Gen 1,2c in
particular – including exegetes of later periods – is K. Smoroński, “‘Et Spiritus Dei ferebatur super
aquas’. Inquisitio historico-exegetica in interpretationem textus Gen. 1.2c”, Biblica, 6 (Rome, 1925),
140-156.275-293.361-395. Unfortunately, Smoroński’s evaluation is misleading on one point that
will be of importance in the context of the present article, cf. below. For further literature on the
exegesis of Gen 1,1-2 and the history of exegesis I shall confine myself to referring to the extensive
bibliography in Cl. Westermann’s commentary on Genesis (Claus Westermann, Genesis, (Biblischer
Kommentar, I/1), 2. Aufl., 1976, 104ff).
2 For an extensive account of the theological impact of St. Augustine’s exegesis of the first verses of
Genesis, cf. E. Teselle, “Nature and Grace in Augustine’s Expositions of Genesis I,1-5”, Recherches
Augustiniennes, 5 (1968), 95-137, where further literature is quoted. Cf. also J. Chaix-Ruy, “La cré
ation du monde d’après Augustin”, Revue des Études Augustiniennes, XI (Paris, 1965), 85-88. For St.
the works mentioned in notes 1 and 2 I can also omit any extensive comments
on the broader theological and historical background of St. Augustine’s con-
tributions, their philosophical implications and so forth, and confine myself to
discussing the principal points of the line of development within St. Augus-
tine’s exegesis of Gen 1,2c and the question of his sources in particular. And I
shall mainly concentrate on one Augustinian source reference which has not
hitherto – to the best of my knowledge – been understood and elucidated in
a satisfactory manner.
St. Augustine’s specific exegetical works that are of interest here, are the
following (cited in chronological order): 1) De Genesi contra Manichaeos (389);
2) De Genesi ad litteram imperfectus liber (393ff); 3) De Genesi ad litteram libri
duodecim (401 – ca. 415); 4) Quaestiones in Heptateuchum (ca. 419)3.
According to the first mentioned source the Manichaeans have raised
two impertinent questions concerning Gen 1,2c, namely: Was the water the
dwelling-place of the Spirit of God? and: From where did the water in Gen 1,2
come, as nothing is told about the creation of water until later? St. Augustine’s
answer to these questions is that neither the verb superferebatur4 nor the water
should be understood in a local and material manner: Non enim per spatia lo-
corum superferebatur aquae ille Spiritus … sed per potentiam invisibilis sublimitatis
suae5 and: Non enim aqua sic appellata est hoc loco, ut haec a nobis intelligatur quam
videre iam possumus et tangere6, and this, he suggests, can be inferred from the
foregoing, where the earth is mentioned as being incomposita and invisibilis7,
i.e. the text itself reveals that it does not here deal with the material and visible
earth; hence the water should be understood in parallel with what was said
about the earth. And as the criticism involved in the Manichees’ questions
is based on the assumption that the water of Gen 1,2 is to be understood as
material water, they are thus refuted by Scripture itself.
Augustine’s understanding of verse 2, and for further literature on that particular point, cf. St. J.
Grabowski, “Spiritus Dei in Gen. 1:2 according to St. Augustine”, The Catholic Biblical Quarterly, 10
(Washington, 1948), 13-28, and Aimé Solignac, “Exegèse et Métaphysique. Genèse 1,1-3 chez saint
Augustin”, In Principio, 153-171.
3 The following editions have been used: 1) Migne’s edition, PL 34,173-220; 2) and 3) I. Zycha’s edi-
tion, CSEL 28,1, 1894; and 4) I. Fraipont’s edition, CCSL 33, 1958.
4 It should be noted that the prevailing reading of Gen 1,2c in St. Augustine is: et spiritus Dei
superferebatur super aquam as compared with et spiritus Dei ferebatur super aquas of the Vulgate.
5 PL 34,177.
6 PL 34,177f.
7 It is of importance to note that these words are the translations in St. Augustine of the LXX’s
ἀκατασκεύαστος and ἀόρατος, as against inanis et vacua in the Vulgate.
T he S pirit ο f G od upon the F ace ο f T he W aters 25
In his unfinished Commentary on Genesis St. Augustine has a fuller treat-

ment of the exegetical possibilities – three in number – that are known to him
at that time: 1) spiritus sanctus, quem in ipsa ineffabili et incommutabili trinitate
ueneramur, 2) vitalis creatura, and 3) aeris elementum8.
Formally, St. Augustine does not in the Liber imperfectus exclude any of
these possibilities, but it seems that already here he gives preference to the
interpretation that it is the Holy Spirit that is mentioned in Gen 1,2; and he
stresses the point that any other interpretation is only acceptable on condition
that it does not involve any faulty opinion concerning the Holy Spirit.
In his latest work on Genesis this preference has become quite clear. Thus
in the Quaestiones the only comment to be made on Gen 1,2 is that here (as in
four passages elsewhere in Genesis and Exodus9) the Holy Spirit is mentioned
expressis verbis.
A much more thorough-going treatment of Gen 1,2c is given in the De
Gen. ad litt. Here again, St. Augustine’s preference for the interpretation that
the verse is concerned with the Holy Spirit is clear; thus he maintains that the
very purpose of the first lines of the Holy Writ is trinitarian, viz. to describe
how God works as Creator through his Word and through his Spirit.
What is characteristic for the De Gen. ad litt. and what makes this work
his most important exegetical contribution concerning Genesis, is, that the
theological implications are worked out much more broadly and deeply than
in the other works.
We shall here be particularly concerned with one element in the descrip-
tion of how the Spirit cooperates in Creation. In parallel with a description of
how God works through his Word (aeternis atque incommutabilibus et stabilibus
rationibus coaeterni sibi verbi sui10) it is said concerning the Spirit that he works
quodam, ut ita dixerim, fotu pariter coaeterni sancti spiritus sui11. Fotus, the central
notion used here about the Holy Spirit, is explained in the following: nam et
illud, quod per graecam et latinam linguam dictum est de spiritu dei, quod superfere
batur super aquas, secundum syrae linguae intellectum, quae uicina est hebraeae – nam
hoc a quodam docto christiano Syro fertur expositum – non superferebatur, sed fouebat
potius intellegi perhibetur, nec sicut fouentur tumores ut uulnera in corpore aquis uel
frigidis uel calore congruo temperatis, sed sicut oua fouentur ab alitibus, ubi calor ille
8 CSEL 28,l,468ff.
9 The 4 other verses referred to are: Gen 6,3; 41,38 and Exod 8,19; 15,10.
10 CSEL 28,1,26
11 Ibid.
materni corporis etiam formandis pullis quodammodo adminiculatur per quendam in

suo genere dilectionis adfectum12. The image of the Spirit as a bird taking care of
its eggs or young ones is further connected with a quotation of Mt 23,37.
This element is new in De Gen. ad litt. as compared with St. Augustine’s
earlier works on Genesis, and as will have been noticed, the author himself
refers to a source for this comment, namely “a certain learned Christian Syr-
ian”. As Berthold Altaner has shown, St. Augustine’s relation to this source is
not immediate, but intermediate through the homilies on the Hexaëmeron by
St. Basil the Great13. Altaner further assumes that St. Augustine’s knowledge
of St. Basil’s homilies does not here pass through St. Ambrose as a vehicle, but
that he must have used the homily in question in Eustathius’ Latin translation,
which Altaner therefore dates to about 400 instead of the traditional dating to
about 440. Leaving this question aside we can briefly assent to Altaner’s view
that St. Augustine’s reference to a learned Christian Syrian is taken from St.
Basil. But this then leaves us with the question: who is this Syrian?
This is the question to which I pretend to be able to supplement and cor-
rect the answers given so far. Before specifying the supplementation needed I
shall for a short while indulge in mild surprise over the astonishingly wrong
answers that have been given to the question by some of my learned predeces-
sors.
Let us begin with Altaner, who claims – briefly and wrongly: “dass Basilius
hier an Ephräm den Syrer denkt, ist wohl als sicher anzunehmen”14. Earlier au-
thorities for the same opinion are e.g. Lenain de Tillemont, the famous church
historian of the early 18th century, and Jules Garnier, one of the early editors
of St. Basil’s works15. Tillemont, incidentally, also suggested the possibility
that Eusebius of Samosata might be St. Basil’s source16. This is a hypothesis,
12 CSEL 28,l,26f.
13 B. Altaner, “Eustathius, der lateinische Übersetzer der Hexaemeron-homilien Basilius des Grossen”,
ZNW, 39 (1940), 161-170 (= Kleine patristische Schriften, TU, 83 (1967), 437-447). Cf. also B. Altaner,
“Augustinus und Basilius der Grosse”, Revue Bénédictine, 60 (1950), 17-24 (= Kleine patristische
Schriften, TU, 83 (1967), 269-276), esp. 18/270.
14 Op. cit., 166/441, note 3.
15 Lenain de Tillemont, Mémoires pour servir à l’histoire ecclésiastique, vol. IX, Bruxelles 1728, 374 (cf.
515f). For Garnier, whose edition has not been accessible to me, I refer to S. Giet’s edition: Basile de
Césarée, Homélies sur l’Hexaéméron, texte grec, introduction et traduction de Stanislas Giet, SC 26, 1950,
169, note 3.
16 Loc. cit.
on which it is very difficult to give any judgment, as no texts have survived,

which can with certainty be ascribed to the Samosatene Eusebius17.
From our own day I shall just mention one author, who as late as 1961
sticks to the incorrect scholarly tradition, that makes St. Ephraem the source
utilized by St. Basil. I am referring to Carl Johann Perl, the German translator
of St. Augustine’s De Gen. ad litt. in the Paderborn translation18.
When it can be stated as definitely as I have done above, that St. Ephraem
cannot possibly be the Syrian source referred to, this is due to the fact that it
can be ascertained without any ambiguity that he understands the “Spirit of
God” of Gen 1,2c to mean the wind (or the air). This appears above all from
his commentary on Genesis19. When even K. Smoroński in his otherwise me-
ticulous and valuable survey of the history of exegesis mentioned above was
mistaken, as far as St. Ephraem is concerned, this was due to the fact that he
took a pseudo-Ephraemic text to be authentic20.
Other proposals for an identification of St. Basil’s source, besides St.
Ephraem (and Eusebius of Samosata) have been attempted. I shall quote three
suggestions, namely those, that seem to me to be the most important. Firstly,
Stanislas Giet, the editor in the Sources Chrétiennes of St. Basil’s homilies on
the Hexaëmeron, deserves to be mentioned. He suggests that Theophilus of
Antioch might be St. Basil’s Syrian source21. Admittedly, Stanislas Giet points to
a number of cases where there are striking coincidences between the apologist
of the 2nd century and St. Basil22. In the case of Gen 1,2c the coincidence, how-
ever, only amounts to the use of the notion ζωογόνησις in Theophilus, with
which can be compared St. Basil’s use of ζωογονέω, ζωογονία and ζωτική
δύναμις in the explanation of Gen 1,2c. Even allowing for a rather extensive
use of Theophilus’ Ad Autolycum in St. Basil, this is hardly a sufficient basis
for identifying the Syrian author, to whom St. Basil refers, with Theophilus,
17 Cf. e.g. Bardenhewer, IV, 388.

18 Aurelius Augustinus, Über den Wortlaut der Genesis… zum erstenmal in deutscher Sprache von Carl
Johann Perl (2 vols.), Paderborn 1961-1964. See esp. vol. I, 245.
19 Sancti Ephraemi Syri in Genesim et in Exodum commentarii, ed. R. M. Tonneau, (CSCO 152-153, Script.
Syr. 71-72), Louvain 1955. See esp. vol. 152, 11f (text), and vol. 153, 7f (translation). Cf. T. Jansma,
“Ephraems Beschreibung des ersten Tages der Schöpfung”, OCP, 37 (1971), 295-316.
20 Cf. K. Smoroński, op. cit., esp. 282ff. The commentary in question is in reality by Jacob of Edessa,
which dates it to the 7th century. Cf. e.g. I. Ortiz de Urbina, Patrologia Syriaca, Rome 1965, 61 and
179f.
21 Op. cit., 54 and 169, note 3.
22 Op. cit., 52ff.
as that would require, either that St. Basil had spun out the bird imagery of
the mere notion ζωογόνησις, or that Theophilus should have done so in an
unknown source.
A second suggestion is that a certain Aphraates might be St. Basil’s source.
This is not a new suggestion, as it goes back at least to the 8th-9th century,
namely to the Syrian exegete Ishō bar Nūn. He knows about the identification
of St. Basil’s source with St. Ephraem, a possibility which he rejects in order
to substitute it with the reference to Aphraates. The passage in question from
Ishō bar Nūn’s Selected Questions on the Pentateuch should be quoted: “As for
the certain Syrian from whom Basil said that he had heard (this interpreta-
tion), some (expositors) in their ignorance say that he was Mar Ephraem, the
teacher. In this they are wrong because Mar Ephraem interprets the word, like
the blessed Interpreter, (as referring) to the spirit of the air. Now the Syrian,
from whom (Basil) heard it, is Aphraates. For (Basil) came to him and he had
a conversation with him …”23.
Ernest G. Clarke, the editor of Ishō bar Nūn’s Selected Questions, rightly
notes that “it is impossible to decide whether IbN, in mentioning Aphraates,
was referring to Aphraates the Persian Sage or the Aphraates who is described
by Theodore(t) of Cyrrhus in his Historia Ecclesiastica (IV, 25-26) and in his
Historia Monachorum (VIII) or still another with this name but unknown to
us”24; and Antoine Guillaumont calls attention to the fact that the exegetic
point in question is not evidenced in the Demonstrationes by Aphraates, the
Persian Sage, where the closest parallel to be found is the use of the verb ‫רחף‬
to describe the relation of the Spirit to the baptismal water25.
Thus it can hardly be maintained that the reference to Aphraates has
brought the discussion to a conclusion. We must therefore turn to a third
suggestion, which is to my mind the best and most important that has been
put forward in recent years. I am referring to Robert Devreesse’s proposal that
23 Question 7 (f. 5r,16-6r,14). I quote from the translation in: Lucas Van Rompay, “Išo bar Nun and
Išo’dad of Merv: New Data for the Study of the Interdependence of their Exegetical Works”, Ori-
entalia Lovaniensia Periodica, 8 (1977), 229-249 (see esp. 232f). For some of my references concerning
the Syrian material I am indebted to Dr. Van Rompay, to whom my cordial thanks should here be
extended.
24 The Selected Questions of Ishō bar Nūn on the Pentateuch, ed. and transl. from MS Cambridge Add. 2017
with a study of the relationship of Ishō’dādh of Merv, Theodore bar Kōnī and Ishō bar Nūn on Genesis, ed.
by Ernest G. Clarke, (Studia Post-Biblica, 5), Leiden, 1962, 82.
25 Antoine Guillaumont, “Genèse 1,1-2 selon les commentateurs syriaques”, In Principio, 115-132, see
esp. 129 with note 94.
Diodore of Tarsus might be St. Basil’s source26. Eduard Schweizer already saw
a connection between St. Basil and Diodore, but he did not suppose the reason
to be that St. Basil should be dependent on Diodore. Whether he supposes a
common source or thinks that Diodore builds upon St. Basil, is not quite clear27.
That there is an evident relation between Diodore and St. Basil, is also noted
by P. Agaësse and A. Solignac, who published their French translation of St.
Augustine’s De Gen. ad litt. in 1972. Incidentally, a quotation of P. Agaësse’s
and A. Solignac’s cautious remarks upon the question of sources might form
a reasonable rounding off of this survey of recent – and older – comments on
St. Basil’s Syrian source (which, by the way, does not at all claim to be exhaus-
tive): “le texte original de Basile laisse entendre qu’il doit ses renseignements
à une explication de vive voix plutôt qu’à un texte écrit … L’informateur de
Basile peut être tout simplement un Syrien éclairé qu’il aurait connu dans sa
Cappadoce natale, ou à l’occasion de ses voyages”28. As appears, the question
of source is left open in Agaësse’s and Solignac’s final formulation. They give
no precise reason for their doubts as to Devreesse’s suggestion that Diodore
is St. Basil’s source, but seem to hint at the point that St. Basil’s description of
his source does not look quite natural in that case29, and I tend to agree with
them, since St. Basil describes his source as a man “who is as far from worldly
wisdom, as he is near the insight in what is true”30, and even if the Hexaëmeron
homilies were preached before the personal friendship between St. Basil and
Diodore was established31, the formula quoted seems to be a weak descrip-
26 Cf. R. Devreesse, Les anciens commentateurs grecs de l’Octateuque et des Rois, (Studi e Testi, 201), Città
del Vaticano, 1959, 156.
27 E. Schweizer, “Diodor von Tarsus als Exeget”, ZNW, 40 (1941), 33-75, see esp. 49f.
28 La Genèse au sens littéral en douze livres. Traduction, introduction et notes par P. Agaësse et A. Solignac,
(Bibliothèque Augustinienne, Oeuvres De Saint Augustin, vols. 48-49), Paris, 1972, see esp. 590-593. Cf.
also Aimé Solignac, op. cit. (see note 2), 161.
29 This is also E. Schweizer’s view, loc. cit., note 81.
30 Ἐρῶ σοι οὐκ ἐμαυτοῦ λόγον, ἀλλὰ Σύρου ἀνδρὸς σοφίας κοσμικῆς τοσοῦτον ἀφεστηκότος,
ὅσον ἐγγὺς ἦν τῆς τῶν ἀληθινῶν ἐπιστήμης, ed. S. Giet, 168; in Eustathius’ Latin translation the
passage runs: dicam tibi non meum sermonem sed viri cuiusdam genere Syri qui tantum aberat a sapientia
saeculari quantum verae doctrinae proximus habebatur, cf. Eustathius, Ancienne version latine des neuf
homélies sur l’Hexaéméron de Basile de Césarée, ed. E. Amand de Mendieta et S.Y. Rudberg, (TU, 66),
1958, 26. This, no doubt, is the formula that lies behind St. Augustine’s expression: a quodam docto
christiano Syro.
31 The Hexaëmeron homilies are normally dated before 370 (cf. e.g. Bardenhewer, III, 148), whereas
the bonds of friendship between St. Basil and Diodore seem to have been established during the lat-
ter’s stay in Armenia in 372 (cf. Bardenhewer, III, 305).
tion of the supposedly well-known and recognized orthodoxy of Diodore’s

dogmatic position.
Now, my central point is that the search for a source can be taken one step
further, in so far as it is possible to identify a text which is at any rate Diodore’s
principal source. In fact, he draws very much of his information from this
source, which he quotes verbally in certain passages, in a more paraphrastic
manner elsewhere. For St. Basil, too, it seems probable to me that the text in
question is his principal source, but it is hard to decide whether he has used
it directly or through an intermediary link that might for example belong to
an oral tradition. Both his quotation formula and the minor differences that
exist between St. Basil and his source might be cited in support of the latter
view.
The text in question is the so-called Commentary on the Octateuch by Eusebius
of Emesa. It may be added here, that it is no wonder that this text has not been
taken into consideration by commentators on St. Augustine and St. Basil, as
the passage in question is not represented among the Greek catena fragments
that are attributed to Eusebius32. It is to be found, however, in the Armenian
version that exists under the name of Cyril of Alexandria in two Armenian
manuscripts33. Theoretically, the passage dealing with Gen 1,2c might have
been known to patristic and biblical scholars writing later than 1938, since at
that time a Latin translation of the passage was included in Almo Zanolli’s
book on the Armenian transmission of catenae, esp. on Leviticus34.
Before turning to the actual text, two general preliminary remarks should
be made concerning the probability of Eusebius of Emesa being the common
source for Diodore and St. Basil. As far as Diodore is concerned, St. Jerome
already considered him to be dependent upon and an imitator of the style and
exegetical method, which Eusebius used35. This was not necessarily meant in a
32 Cf. R. Devreesse, op. cit., 55-103; and for a fuller edition of one branch of the catena traditions, i.e.
the Sinaitic catena, cf. Catenae Graecae in Genesim et in Exodum, I. Catena Sinaitica, ed. Françoise Petit,
(CCSG, 2), Turnhout, 1977.
33 On the text, the manuscripts, and the contributions of V. Hovhannessian and others to the identifi-
cation of the text as Eusebian, cf. Per Piscatores, 31ff. During a visit to the Mechitharists of San Laz-
zaro, Venice, in 1978 I was informed that an edition of the Armenian text based on MS Ven. Mech
873 was ready for printing.
34 A. Zanolli, Di una vetusta catena sul Levitico, perduta in greco e conservata in armeno, della sua stretta
relazione col commentario di Procopio di Gaza e dei tre codici di S. Lazzaro, che la contengono, Venice 1938,
89f.
35 Jerome, De vir. ill, 119, cf. Bardenhewer, III, 306, note 6.
positive and friendly manner in St. Jerome’s mouth; that this tie of dependence
is a historical reality, however, has been substantiated more fully, esp. by Ed.
Schweizer36.
As regards the question of dating, it is not possible to give an exact date of
Eusebius’ commentary; presumably he died in (or before) 359, and the earliest
date for any work from his hand would probably be the 330s; so the com-
mentary might very likely have been 15-30 years old, when St. Basil preached
his homilies on the Hexaëmeron. Be that as it may, St. Basil’s introductory
formula would (as suggested already) fit in better with the knowledge one
might suppose him to possess of an author of the previous generation like
Eusebius than with his relation to Diodore; and, incidentally, it would contain
a very interesting – and early – counterweight to St. Jerome’s characterization
of Eusebius as a standard-bearer of the Arian party37, which has rested as a
heavy burden on Eusebius’ dogmatic reputation ever since.
To make it possible to evaluate provisionally Eusebius’ exegetical contribu-
tion and the question of source I shall give a short paraphrase of his comment
on Gen 1,2c, as it appears in the Armenian manuscript, MS Ven. Mech. 873,
140, l. 26 – 141, l. 1338.
Eusebius first concentrates on the verb of Gen 1,2c, saying that the full
meaning of the Hebrew verb can hardly be rendered in Greek using only one
word – at any rate not ἐπιφέρω. He parallels this fact with the necessity in
some cases of using two words in Syriac and Hebrew to render one Greek
word. His examples are the words “slinger” and “archer”39.
36 Op. cit., passim.

37 Jerome, Chronicon, ed. R. Helm 1913, 236, cf. E. M. Buytaert, L’héritage littéraire d’Eusèbe d’Émèse, (Bib-
liothèque du Muséon, 24), Louvain, 1949, esp. 7. The expression used by St. Jerome is arrianae signifer
factionis.
38 My thanks are due to the Mechitharist Congregation of Venice for having placed a microfilm of the
relevant pages of the manuscript at my disposal.
39 Eusebius’ point, of course, is that in some cases when the Greek needs only one word, Hebrew or
Syriac needs two. To use his examples, the archers of Gen 49,23 are in Hebrew ‫( בעלי חצים‬a construc-
tion which is, incidentally, imitated by the LXX: κύριοι τοξευμάτων), and the slingers of Judg.
20,16 (where the LXX have σφενδονῆται) are ‫ קלע באבן‬i.e. in one case a constructus-connection of
two substantives, in the other a participle with a substantive introduced by a preposition. Zanolli
has rendered the Armenian միապէս by pariter, which is misunderstandable. Armenian միապէս
verbally means “in one way”, and normally has the meaning of “in the like manner”, “equally” etc.
Here, however, it has undoubtedly been used to translate what in the Greek original meant “using
only one word”, cf. Diodore: διὰ μίας λέξεως. On the whole, Diodore here follows Eusebius very
closely. Cf. J. Deconinck, Essai sur la chaîne de l’Octateuque, Paris, 1912, see esp. fragment No. 4 (92f)
l. 4-7.
After this general statement concerning the art and technique of transla-
tion40 the author endeavours to render the Hebrew (or Syriac) verb; and to do
so he needs four Greek verbs and the imagery of the bird keeping its young
ones warm, taking care of them and caressing them41.
Then the author refers to the question whether the phrase deals with the
Holy Spirit or the wind42. The first part of his answer is an affirmation of
the possibility of taking the verse to be about the Holy Spirit. Once again he
paraphrases the verb using four – new – verbs, this time verbs meaning: “to
nourish”, “to take care of”, “to make”, and “to guide”43, and he adds that it
might be acceptable to use the verb “to move” about the Holy Spirit – on one
condition, namely that it is used in the same way, as the verbs “to descend”
and “to ascend” are used in the Bible about God44.
40 In preceding parts of the text Eusebius has treated of this theme more fully. Part of this seems to be
paraphrased in the first lines of Diodore’s fragment (ed. cit., l. 1-3).
41 The four Armenian verbs used here are: 1. ողոքեմ, 2 գրգամ, 3. շարշեմ(–իմ), 4. շրջեմ ի վերայ
The fourth verb is the one used in Gen 1,2c in the Armenian Bible, and it is natural to consider it
a rendering of the Greek ἐπιφέρω. Below are quoted the Greek synonyms for the first three verbs
as given in the Armenian thesaurus (= G. Awetik’ean, X. Siwrmelean, M. Awgerean, Nor Bargirk’
Haykazean Lezowi, 1-2, Venice, 1836-1837) and the English translations given by Bedrossian (= M.
Bedrossian, New Dictionary Armenian-English, repr. of the 1879-edition, Beirut, n.d.):
1. Thes. arm.: κολακεύω, θωπεύω. ἀπομειλίσσομαι. θεραπεύω, ὑγιάζω. τιθασσεύω.
Bedrossian: supplicate, persuade mildly; soften, calm, appease; touch, move to pity; fawn upon,
toady, flatter.
2. Thes. arm.: θάλπω, ἐπωάζω, κλώζω.
Bedrossian: cluck, cocker.
3. Thes. arm.: κινέω. σείω, συσσείω, σαλεύω (-ομαι).
Bedrossian: move, remove, set in motion, stir, cause to move; agitate, shake, flutter, stagger; fig. ex-
cite, provoke, rouse the feelings of, move, touch; (be moved etc.).
42 It seems to me to be natural to translate the Armenian sentence as presenting two alternative inter-
pretations. What has led Zanolli to translate otherwise (viz. Nunc vero quæsierimus quod de vento, de
Sancto Spiritu prædicari?) is presumably the fact that the text must not necessarily be read as lending
itself to a formal rejection of the wind interpretation. Cf. the discussion below.
43 The four Armenian verbs used here are: 1. տածեմ, 2. խնամարկեմ, 3. գործեմ, 4. առաջնորդեմ.
Below are quoted their synonyms as given in the Thes. arm. and Bedrossian:
1. Thes. arm.: τημελέω. θάλπω. τρέφω.
Bedrossian: care, take care of, preserve; nourish, maintain, sustain, feed; cultivate; foment, stir up.
2. Thes. arm.: ἐπιμελέομαι.
Bedrossian: take care of, attend to, occupy oneself solicitously about, patronize.
3. Thes. arm.: έργάζομαι. πράσσω. ἐνεργέω, ἀπεργάζομαι.
Bedrossian: work, make, do; fashion, manufacture; commit, perpetrate; knit; twist.
4. Thes. arm.: ὁδηγέω, ἡγέομαι, κυβερνάω.
Bedrossian: conduct, lead, guide, command, head, rule, govern, direct, escort, accompany.
44 This part of Eusebius’ comment is omitted by Diodore, but is quoted by Procopius, PG 87,45.
Still concentrating on the meaning of the verb Eusebius now goes on to

say that “to move” is of course a suitable verb for the wind. But the question
is whether this interpretation can bring out the real and full meaning of the
Hebrew verb. The following passage seems to answer this question in the
affirmative, since it speaks about the familiarity and proximity of wind and
water and of their relationship in a process of heating45. This might superficially
be read as a positive statement concerning the wind interpretation. Before
accepting this reading, however, it should be noticed: 1) that the weight of
the passage might lie on the statement that this is not a process done by an
act of will, 2) that – by way of comparison – St. Basil also refers to the wind
interpretation as a possibility that cannot be excluded; only, it is “more true”
to maintain that the verse is about the Holy Spirit46, 3) that an external witness
might support the view that Eusebius’ comment – like St. Basil’s – is in reality
aimed at a rejection of the wind interpretation. I am thinking of Ishodad of
Merw, whose contribution shall be commented upon a little later on.
Finally, it is added that the designation “God’s” might be used about
the wind, because wind or air is nearer to the incorporeal than is e.g. earth
and water, and the presentation of the exegetical possibilities in Gen. 1,2c is
concluded by the statement that some people take God’s Spirit to mean the
energy that heated the nature of the waters47.
From this paraphrase of Eusebius of Emesa’s comments upon Gen. 1,2c it
will have appeared that he can be credited with the merit of being the author
of the pre-Basilian period to have given the fullest presentation of the bird
imagery and of the semantic field of the Hebrew verb as parts of the interpre-
tation that takes the verse to be about the Holy Spirit.
That his comments are used in a very direct and extensive manner by
45 Some features of this passage are quoted by Diodore, see esp. ed. cit., l.11ff.
46 Undoubtedly, there is a formal difference between St. Basil and Eusebius, not only in St. Basil’s
dismissing the wind interpretation before turning to the “truer” understanding, but also in the way
in which St. Basil pointedly underlines that the former interpretation would place the wind as one
of the four elements of the created world. Here he agrees with the central point in St. Ephraem’s
interpretation, on which, however, he is hardly dependent, as Ephraem does not – like Basil – count
four “parts of the world”, but five “natures”, presumably, as shown by T. Jansma, in order to reject
Bardesanes’ cosmogony. Cf. T. Jansma, op. cit.
47 The final passage again belongs to what is quoted by Diodore. In general, it may be added, that the
three exegetical possibilities that are mentioned by Eusebius, as well as by Diodore, St. Basil, and
St. Augustine in his De Gen. ad litt. imperf. liber, are identical in their main outline. In their general
content, therefore, they can be assumed to belong to what may be called an established fund of ex-
egetical knowledge.
34 s t u d e n t s o f t h e b i b l e i n 4 t h a n d 5 t h c e n t u r y s y r i a
Diodore is obvious, when the two texts are read in comparison48. It also seems
clear that Eusebius is very likely to be the learned Christian Syrian referred to
by St. Basil and St. Augustine.
I shall only add a few remarks on the use made elsewhere in the history
of exegesis of Eusebius’ work. In the Greek tradition, apart from the authors
mentioned already, Procopius of Gaza, esp., utilizes the Eusebian Commen-
tary49, and there is of course that special interest connected with authors such
as Diodore and Procopius that in their verbal quotations one may be able to
find fragments of Eusebius’ own original Greek text.
As was hinted at above, it may also be possible to find further testimo-
nies in the Syrian tradition, esp. in Ishodad of Merw’s commentary on the
Old Testament. In his comments on Gen. 1,2c he quotes St. Ephraem and the
Blessed Interpreter (i.e. Theodore of Mopsuestia) as witnesses for the wind
interpretation. Before that, however, he refers to “St. Basil and others” for
the Holy Spirit interpretation quoting a passage that is so closely related to
the passage where Eusebius comments upon the wind interpretation, that it
must be considered a direct quotation of Eusebius50. And also parts of what
follows in Ishodad’s commentary could be considered to be quotations from
Eusebius51.
It would thus seem that Eusebius’ Commentary on the Octateuch has taken its
natural place among other exegetical authorities of the Old Church to which
both Greek and Syrian exegetes of the Middle Ages recurred, and it would
hardly be untrue to say that there is a good deal of work ahead sorting out
such quotations.
48 Cf. above, notes 40 and 45.

49 Cf. above, note 44 (which, indeed, does not exhaust the quotations to be found in Procopius).
50 Cf. above. The passage in question is here quoted in C. van den Eynde’s translation (Commentaire
d’Išo’dad de Merv sur l’Ancien Testament, I. Genèse, traduit par Ceslas van den Eynde, (CSCO, 156,
Script. Syr., 75), Louvain, 1955, 19): Ici encore (ce sont) le bienheureux Basile et d’autres, qui ont
interprété (ces mots) du Saint-Esprit. Mais voyant que le mot était porté était absurde et qu’il n’était
pas seyant de l’appliquer au Saint-Esprit, ils eurent recours au mot couvait de l’Hébreu et du Syrien,
(disant): Si les mots son esprit couvait se rapportent à l’air, comment celui-ci, un être inanimé et in-
sensible, couve-t-il comme une poule (couve) ses poussins, et active-t-il et échauffe-t-il comme par
un effet de sa volonté, et meut-il les eaux? (Cf. MS Ven. Mech. 873,141,1. 4.6-9). Cf. also T. Jansma,
“And the Spirit of God moved upon the face of the waters – Some remarks on the Syro-Hexaplaric
reading of Gen 1.2”, Vetus Testamentum, 20 (1970), 16-24, see esp. 19. Cf. also Antoine Guillaumont,
op. cit., and Lucas Van Rompay, op. cit.
51 This is true of the following passages: op. cit., 19, l. 19-21 (~ MS, 141, l. 10); op. cit., 20, l. 7-8 (~ MS,
141, l. 12-13) (van den Eynde here refers to St. Basil, but the correspondence with Eusebius is
closer); op. cit., 20, l. 9-12 (~ MS, 141, l. 10-12).
Another question, of course, immediately suggests itself: that of Eusebius’

sources. I have made no investigations on that point, so by way of conclusion
I shall just mention that Smoroński refers to early Jewish exegetes as witnesses
of the bird imagery52. These sources are more in accordance with the interpreta-
tion met in Eusebius than is the ornithological and mythological information
on the behaviour of birds and the myth of the Cosmic Egg etc., that is met
with in a number of modern commentaries trying to come to grips with the
content and meaning of the verb used in Gen 1,2c. This does not mean that
Eusebius and the early Rabbis are necessarily right; but historically they belong
to a current of tradition which may not yet have revealed to us all the secrets
it has to tell.
52 Cf. K. Smoroński, op. cit., 146ff, esp. 151f. Cf. also W. H. McClellan, “The meaning of ruaḥ‘Elohim in
Genesis 1,2”, Biblica, 15 (Rome, 1934), 517-527, esp. 526.
Some Questions Concerning the Armenian
Version of the Epistle of James
If I may be forgiven an exordium that is perhaps slightly too rhetorical, I should

like to state initially that even if all the impressive architectural monuments
of Armenian culture and church history had disappeared or had even never
been built, we might still be in possession of one witness that would be suf-
ficient to secure for ever Armenian language and endeavours an important
place within the history of the Christian churches.
I am thinking of the Armenian version of the Bible, and I am well aware
that I am not making any new statement here, since this translation is very
often – since La Croze coined the expression about 200 years ago – referred
to as “the Queen of Versions”.
Now, even a queen may be subjected to scientific examination, and biblical
scholars who examine the Armenian version of the Scriptures, could be said to
pose questions that run to some extent parallel to questions posed by historians
of architecture, language, and literature, e.g. such questions as: What is Greek
influence? What may be defined as Syrian elements? And what is genuinely
Armenian in this literary monument, which stems in its final shape from the
fifth century, i.e. from as early a date as some of the earliest Armenian churches
now known?
In the history of textual criticism it is true that the evaluation of the signifi-
cance of the old versions has changed with the times. Thus early in the 20th
century the English scholar F. C. Burkitt for example might be quoted for a
very high estimation of the versions,1 whereas later on it may be true to say
that Greek material – especially papyri fragments of a very early date – stole
the picture. To-day a new interest in the old versions seems to have arisen in
1 Cf. the quotation below (from Metzger, see note 3).

certain quarters, whereas others give only slight attention to the witness of
the old translations.
As an example of the last-mentioned attitude may be quoted a passage from
Kurt Aland’s introduction to the 26th edition of the Novum Testamentum Graece:
“The early versions, whether in Latin, Syriac, Coptic, Armenian, Georgian, or
any other language, are frequently overrated”.2
Another well-known text-critical scholar of the present day, Bruce M.
Metzger, holds the old versions of the New Testament in higher esteem. In
the preface to his important manual on the ancient versions from 1977, after
mentioning the usefulness of the early versions for the church historian, the his-
torian of liturgy, and the philologist, he adds: “It is the textual critic, however,
for whom the early versions of the New Testament are of prime importance.
Earlier in the twentieth century F. C. Burkitt went so far as to argue that a
reading supported by the Old Latin k and the Sinaitic Syriac deserved as much
respect as one witnessed by B and ‫א‬. Although the subsequent discovery of
early Greek papyri (such as P66 and P75, which antedate B and ‫ א‬by more than a
century) has required a reassessment of Burkitt’s views, the textual critic must
still give serious attention to readings that are supported by a combination of
unrelated versional witnesses.”3
To my mind, it is by no means an accident, that two important works of
reference concerned with the old versions have been published in the 1970’es,
viz., besides Metzger’s book, the Münster manual edited by Kurt Aland,4 where
the Armenian Bible is treated of by Louis Leloir.5
I do not intend to give a survey of research into the Armenian version of
the Bible or text books concerned with it; it might be useful, however, as a
background for what I hope to point out more specifically, to call attention to
a few characteristics of the two books mentioned; and it seems to me that in
such a presentation of modern reference works, however short and incomplete,
2 Novum Testamentum Graece, post Eberhard Nestle et Erwin Nestle ed. Kurt Aland (et al.), 26. Aufl.,
Stuttgart 1979, (quotation from Introduction, p. 54*).
3 Bruce M. Metzger, The Early Versions of the New Testament. Their Origin, Transmission, and Limitations,
Oxford 1977, pp. VIIf.
4 Kurt Aland (ed.), Die alten Übersetzungen des neuen Testaments, die Kirchenväterzitate und Lektionare,
Arbeiten zur neutestamentlichen Textforschung 5, Münster 1972.
5 Op.cit., pp. 300-313.
S o m e Q u e s t i o n s C o n c e r n i n g t h e A r m e n i a n V e r s i o n o f t h e Ep i s t l e o f J a m e s 39
also Arthur Vööbus’ book on the early versions of the New Testament6 should
be mentioned, even if it is now more than 25 years old.
Since I shall be mostly concerned with Leloir’s contribution, I will not men-
tion the books in chronological order, but give a short comment on Vööbus’
and Metzger’s books first, and then turn to Leloir.
As will be known, Arthur Vööbus has been concerned especially with
problems related to the early Syrian church, not least the evidence of the
early versions of the Bible in Syriac. Thus, in a brief presentation of Vööbus’
book – and esp. the chapter on the Armenian version – the following phrase
might be picked out as a key sentence: “one cannot evade the conclusion
that the most ancient Armenian version as known to ancient Armenian writ-
ers was not made from Greek, but, beyond doubt, from Syriac manuscripts.
This Unterlage (sic) was an Old Syriac text, a type somewhat more archaic in
places than Syr-Sin and Syr-Cur and somewhat closer to the text-pattern of
the Diatesseron” (sic).7
On the basis of the conclusion – or working hypothesis – quoted, Vööbus
emphasizes the necessity of examining textual conditions in Edessa, the eccle-
siastical and theological metropolis of Northern Syria, in order to elucidate the
background and history of the Armenian Bible. As Vööbus clearly states, this is
not a new idea, and he gives a fairly broad account of the history of research,
esp. in the 20th century, evidently feeling that his own work is a pioneer work
because of its amount of Syrian material, unknown or unconsidered in earlier
generations.
It should be added that Vööbus acknowledges the results reached by St.
Lyonnet in his extremely important monograph on the origin of the Armenian
version of the Gospels and the Diatessaron,8 which run to a certain extent par-
allel to Vööbus’ own findings (even if they disagree on the question whether
the Syrian Vorlage of the Armenian Gospels was Diatessaric or not).
The reason why I do not comment upon Lyonnet’s book in this connection,
even if it is undoubtedly a highly important work on the Armenian Bible –
maybe the most important since Frédéric Macler in 1919 formulated his theory
6 Arthur Vööbus, Early Versions of the New Testament, Papers of the Estonian Theological Society in Exile,
6, Stockholm 1954.
7 Op. cit., p. 151.
8 S. Lyonnet, Les origines de la version arménienne et le Diatessaron, Biblica et Orientalia, 13, Rome 1950.
of a purely Greek background of the Armenian Bible9 – is that Lyonnet is

primarily concerned with the Gospels, and even if he must of course consider
material outside the Gospels, he has for example not one single example or
quotation derived from the Epistle of James, with which we shall mainly be
concerned here.
The absence in Lyonnet’s book of quotations from the Epistle of James may,
incidentally, have a very natural explanation in so far as the Catholic Epistles
were not accepted as canonical writings in the earliest Syrian church, so the
probability of identifying any translation into Syriac of this part of the Bible
earlier than the Peshitta may be limited.10
For a student wanting a solid and well-balanced introduction to the field of
textual criticism and the value of the old versions for this field of scholarship
Metzger’s book is undoubtedly the best. He adds no specific hypotheses of
his own, but gives a good and precise account of the present state of research.
In this respect his book may be characterized as a traditional, thorough-going
manual. What is new – and maybe one of the most recommendable features
of Metzger’s book, is that he finishes his presentation of each version with a
chapter concerned with the limitations of the language in question in render-
ing the Greek. Metzger has left the preparation of these chapters to scholars
who know the languages concerned; for the Armenian the author is Erroll
F. Rhodes of the Library Research Staff of the American Bible Society. (The
limitations of Syriac, which are of interest in our context, of course, are treated
of by Sebastian Brock of the University of Oxford.) In a way, the importance
or necessity of such chapters may be self-evident. Traditionally, however,
they are not to be found in manuals of textual criticism, so the student is too
often left with the impression that the translations can be used to determine
the choice between Greek variants or even to emend the Greek text in a more
direct and easy procedure than is in fact the case.
And, incidentally, one of the points I have to make concerning Leloir’s
contribution, is that he is not sufficiently aware of the methodological problems
that arise from the limitations and possibilities of the languages concerned.
Neither does he give full attention to the need of distinguishing between one
kind of syriacism and the other, or between readings that are chosen because
they are idiomatic Armenian or for reasons concerned with translational tech-
9 F. Macler, “ Le texte arménien de l’évangile d’après Matthieu et Marc”, Annales du Musée Guimet,
28, 1919. Cf. below, p. 54f.
10 Cf. e.g. Metzger, op. cit. p. 44.
nique on one hand, and on the other what is terminology consciously chosen
for theological reasons.
These remarks are obviously rather critical. Before I try to substantiate
the criticism, I should therefore like to point out, firstly that Leloir gives an
explicit statement of very sound methodological principles concerning some
of the points mentioned,11 secondly that the reason why the greater part of his
contribution is limited to a discussion of the text of the Epistle of James, is that
the Institut für Neutestamentliche Textforschung in Münster under Kurt Aland
has for a number of reasons concentrated on this epistle as a good sample with
which to begin the Editio Maior Critica of the New Testament. Therefore most
of the contributions in the Münster manual concentrate on that text.
To these points might be added that it seems to me to be particularly neces-
sary to take up the methodological discussion of Leloir’s presuppositions and
conclusions because his answer: ‘Probablement oui’ to the question, whether
the Vorlage of the Armenian Epistle of James is Syriac, seems to be on its way
to be generally accepted on the basis of Leloir’s reputation as a scholar familiar
with old Armenian literature.
This, for example, is the impression one may get from W. L. Richards’
stocktaking of the text critical problems in the Catholic Epistles. A short pre-
sentation of the three relevant articles from Richards’ hand might round off
this introductory survey of research (which has, evidently, not been meant
to be exhaustive).
W. L. Richards’ first article12 is mainly bibliographic; the second one13 is
concerned with the transmission of the text of the Catholic Epistles in lec-
tionaries and versions; and finally, he draws up in his third article14 a list of
manuscripts available on microfilm in the USA and Canada.
By way of introduction to his first article Richards quotes Jean Duplacy’s
remark on the Catholic Epistles: “Ni l’histoire du corpus ni celle du texte n’ont
été très étudiées”.15 This remark goes back to 1958; but Richards finds that now
the situation is changing. This verdict is based, above all, on a reference to the
11 Op. cit., p. 306.

12 W. L. Richards, “Textual Criticism on the Greek Text of the Catholic Epistles: A Bibliography”, An-
drews University Seminary Studies, 12, 1974, pp. 103-111.
13 W. L. Richards, “The Present Status of Text Critical Studies in the Catholic Epistles”, Andrews Uni-
versity Seminary Studies, 13, 1975, pp. 261-272.
14 W. L. Richards, “The New Testament Greek Manuscripts of the Catholic Epistles”, Andrews Univer-
sity Seminary Studies, 14, 1976, pp. 301-311.
15 Jean Duplacy, Où en est la critique textuelle du Nouveau Testament?, Paris 1959, p. 64.
work done by the Institut für Neutestamentliche Textforschung in Münster,

and therefore it is characteristic that in the article concerned with the old ver-
sions Richards builds mainly on the Münster manual mentioned above.
Richards’ brief summary of Leloir’s main points of view will be quoted
here before I turn to the treatment of some questions concerning Leloir’s pre-
sentation of the Armenian version of the Epistle of James: “After indicating
his conviction that a first Armenian version of Acts was based on the Syriac
and that the Pauline Epistles were more than likely based on the Syriac, Leloir
asks if we can say the same for the major Catholic Epistles (pp. 302-303). His
answer: “Probablement oui”. Examples of Syriacisms are given to support the
theory that while the Armenian text of our editions is based on the Greek,
these vestiges of the Syriac within the Armenian text point to a Syriac Vorlage
for the first Armenian version (pp. 303-304)”.16
In the following I shall turn to a discussion of eight examples touched
upon by Leloir. It should be noted beforehand that these eight instances are
not the only points in Leloir’s chapter that could be discussed; neither do they
provide a sufficient basis for deciding the question: Syrian Vorlage or not? Nor
do they permit a discussion of all the questions of method involved. They seem
to me, however, within the limited scope available here, to lend themselves
to be used as illustrations of some of the important problems and questions
of method referred to above – and others.
1. James 1,23
As far as this verse is concerned Leloir concentrates on the rendering of the verb
κατανοέω.17 The grammatical form of the Greek verb is a present participle,
rendered in the Armenian by a relative clause. In this clause is used the verb
հայիմ preceded by the participle of the verb պշնեմ,պշնում or պշուցանեմ
(պշուցեալ հայիցի), which is in Leloir’s literal translation into Latin reproduced
by stupefactus aspiciet.
Now Leloir comments upon this rendering: “‘stupefactus’ est une glose”.
And he continues by stating that whether or not this has any textual support
elsewhere, the word has a meaning – or significance of its own. In fact, “cette
addition est une finesse du traducteur, et une correction de l’imprécision de
la phrase grecque”, and his further comments consist of meditations on the
16 Richards, op. cit. (cf. note 13), p. 269.

17 Op. cit., pp. 308f.
difference between those who are made stupefacti and those who are satisfacti
through a look in the mirror.
As I hope to substantiate below, such meditations are superfluous in this
context, and it is wiser to stick to the information given by Leloir in a note, viz.
that the two verbs in combination could be translated “regarder fixement”.18
In the Armenian New Testament the Greek κατανοέω is rendered in six
different ways, as will appear from the survey in Table 1.
The verb պշնեմ, -նում, -ուցանեմ is used four times in the Armenian New
Testament, viz. those listed in Table 2.
Table 1 also shows that only in two cases does the Armenian use a combi-
nation of two verbs, namely apart from James 1,23, the neighbouring verse:
1,24, where also the grammatical construction (participle + finite verb) is the
same (հայեցեալ ետես).
Table 1
κατανοέω in the Greek New Testament – Renderings in Armenian
Rendering Texts Number of cases

հայիմ Lk. 12,24; 12,27; Acts 7,32; 11,6; Rom. 4,19 5
նշմարեմ Mt. 7,3; Lk. 6,41; Acts 7,31; 27,39 4
նայիմ Lk. 20,23; Hebr. 3,1 2
զգուշանամ Hebr. 10,24 1
պշուցեալ հայիմ Js. 1,23 1
հայեցեալ տեսանեմ Js. 1,24 1
14
Table 2
Armenian NT: պշնեմ (etc.)
Text Greek phrase Armenian phrase

Acts 1,10 ὡς ἀτενίζοντες ἦσαν մինչդեռ պշուցեալ հայէին
Acts 3,12 τί ἀτενίζετε զի էք պշուցեալ
Acts 11,6 εἰς ἥν ἀτενίσας κατενόουν յոր պշուցեալ հայէի
Js. 1,23 ἀνδρὶ κατανοοῦντι մարդոյ որ պշուցեալ հայիցի
18 L. cit., note 11.

As regards the use of պշնեմ (etc.) in the Armenian New Testament, it should
be noted that all three occurrences outside James 1,23 are found in Acts. In
two cases (Acts 1,10; 11,6) the verb is used in the participle in conjunction
with a finite form of the verb հայիմ – as is the case in James 1,23. In one of
these cases (Acts 1,10) the combination of the two verbs translates the mere
ἀτενίζω of the Greek text, in the other the Greek has the aorist participle of
ἀτενίζω in conjunction with a finite form of κατανοέω. (In the third case also,
the Greek uses ἀτενίζω).
However modest the volume of this material is, it seems to me to indicate
that in some cases the Armenian translators of the New Testament have found
the mere հայիմ too weak to translate κατανοέω. In such cases they might,
among other possibilities, choose their “synonym” for ἀτενίζω, i.e. պշնեմ
(etc.) or a combination of պշնեմ (etc.) and հայիմ.
I readily agree in calling this a “finesse du traducteur”, as does Leloir, but
I do not agree in seeing necessarily or primarily a very specified semantic
interpretation in the direction of stupefactus. To my mind meditations on the
psychological and edifying impact of certain possible, but uncommon semantic
connotations of the verb are less helpful in clarifying the semantic field of the
word and the translational technique used by the Armenian translator of the
Epistle of James, than is the comparison with other occurrences in the Arme-
nian New Testament, from which it appears that in idiomatic Armenian of the
time the primary meaning of the verb was “regarder fixement”. To this may
be added, 1) that in the translator’s mind may have been active the stylistic
ideal of variation making it desirable to choose different renderings in two
neighbouring verses (cf. Js. 1,23 and 1,24); and 2) that his wish to render what
he has understood as the emphatic element in the Greek prefix κατα-, does
not seem to be an isolated phenomenon to be registered in this verse alone
(cf. below). And to my mind it is methodologically unsound not to exhaust
such linguistic and stylistic considerations before turning to other means of
understanding the choice of expressions in a text.
2. James 2,13
Here Leloir makes three observations to demonstrate that the Armenian trans-
lator specifies or clarifies the Greek text in accordance with his linguistic,
emotional and religious understanding of the contents.19
Firstly he maintains that the translation of the Greek verb κατακαυχάω
emphasizes the feelings of confidence and serenity that may be expressed
through this verb. Secondly the preposition առ (which is understood as iden-
tical with Latin ad) is taken to be used to specify the meaning of the Greek
genetive. And thirdly the ‘article’ with դատաստան according to Leloir makes
it clear that the author is here speaking of God’s judgement.
Let us consider the three points stressed by Leloir:
re 1: It is true that κατακαυχᾶται is translated բարձրագլուխ պարծի, which
in a literal translation may be rendered erecta (or capite-alto) gloriatur, as Leloir
does. I agree that the adverb may be described as having an emphatic func-
tion; thus, the New English Bible’s choice of the verb ‘triumph’ to translate
the Greek verb and Bedrossian’s rendering of the Armenian adverb as mean-
ing ‘boldly, highly’20 both represent the meaning clearly, whereas ‘confiance’
and ‘sérénité’ are not very good paraphrases. In fact, I feel that they lead to
another interpretation of the meaning of the verse than that intended by the
Armenian translator – and the author. Probably, it would have been more to
the point to call attention to the above-mentioned possibility of a preference
for rendering the emphatic content of the Greek κατα-, here by means of an
adverb: բարձրագլուխ.
re 2: The phrase, in which Leloir comments upon the preposition ad used
in his Latin translation is not very informative; “ ad précise la valeur du génitif
κρίσεως”, it runs. To me it is difficult to see what kind and degree of “pré-
cision” is contained in the preposition; and rather than Leloir’s paraphrase
gloriatur ad + acc. it would have been preferable to translate for example:
superexultare + dat. or gloriari adversus + acc. I have deliberately chosen the
translations to be found in the Vulgate in Js. 2, 3 and Rom. 11,18. In the last-
mentioned place, incidentally, just as in Js. 2,13, κατακαυχάομαι + gen. is
rendered in the Armenian version through պարծիմ + առ + instr. This might
then seem to be the usual way of translating this locution.
re 3: That Js. 2,13 refers to God’s judgement is beyond doubt. But the role
of the distribution of articles in that respect is rather doubtful. The word for
19 Op. cit., p. 309.

20 M. Bedrossian, New Dictionary Armenian-English, 1879 (repr. Beirut, n.d.), s.v.
judgement is used twice, and the Greek has the definite article in the first place,
but no article in the second, whereas the opposite is true about the Armenian.
To me it is difficult to find any “désir de netteté” in this difference, as does
Leloir.
3. James 1,9.10
While it might be maintained that it is a question of a technical character and
a limited scope, whether Leloir’s rendering of առ + instr. through ad + acc. is a
happy choice, the importance he himself assigns to the Armenian translator’s
choice of grammatical case cannot be dismissed offhand.
Thus, for instance, in some cases he finds that the use of the accusative
instead of the ablative tends to “aboutir à une précision plus grande”.21
His examples are Js. 1,1 and 1,9.10. For 1,1, it might be noted in pass-
ing that if the readings սփիւռս or սփիւռսդ (which exist) are chosen, this
Armenian word for διασπορά is in the plural where acc. and loc. cannot be
distinguished.
Conc. Js. 1,9 and 10, Leloir comments upon the syntactical construction
in connection with the word καυχάομαι. In the Armenian version the verb
պարծիմ is followed by i + acc., and this invites us, according to Leloir, to “ap-
profondir le sens de la situation” – in the direction, it seems, of pointing at a
durable state of mind (“… en remuant sans cesse sa dignité chétienne (sic) …
en songeant constamment à la fragilité…”) – just as in 1,1 the acc. points to a
“dispersion qui se continue”.22
In Table 3 are shown the syntactical constructions with Gr. καυχάομαι and
Arm. պարծիմ in the New Testament. From this survey can be ascertained
that in the Greek NT ἐν + dat. is the construction occurring most frequently,
and that in Armenian, ի + acc. (and the instrumental case) are the prevalent
rendering(s) of this. To my mind, therefore, this (these) rendering(s) – just as
առ + instr. for Greek gen. after κατακαυχάομαι – should be characterized as
the usual Armenian idiom of the day, which in itself does not involve any
specification of a theological, religious or psychological order.
Besides this overall assessment a number of details call for comments. I
cannot go into every detail, but I should like to draw the attention to the fol-
lowing six points.
21 Op. cit., p. 312.

22 Op. cit., pp. 312f.
1. For exegetic and contentual reasons the question could be raised whether,
e.g., in the Pauline context, 2 Cor. 10,13-17, the clauses with εἰς + acc. and
those with ἐν + dat. should be listed in the same survey (to put it somewhat
bluntly). Also the constructions in 2 Cor. with both acc. and ὑπέρ + gen. (7,14
and 9,2) might be classified as a particular group, and other points as well
might deserve a closer analysis and some reservations. Here, however, we
must confine ourselves to formal considerations.
2. When in the overall assessment above I only mentioned ի + acc. (and not
ի + acc./loc.), this is due to the fact that in my evaluation the instances with ի +
acc./loc. could in reality be taken together with the instances with ի + acc; the
reason why “loc.” is added is that in the plural (as mentioned above) Arme-
nian does not distinguish between the two cases. From the prevalent choice
of the acc., when the word is in the singular, it seems to me obvious that also
in the words in the plural it is the acc. that is used. As the starting-point of
this discussion is the fact that Leloir maintains that the choice of the acc. as
opposed to other cases makes an important difference, I have, however, felt
obliged to register in Table 3 the plural-words as being in “acc./loc”.
3. Where other possibilities than ἐν + dat./ի + acc. (or instr.) are used, it
is obvious in a number of cases that the Greek choice of construction has
determined the Armenian; cf. e.g. վասն + gen. for ὐπὲρ and περὶ + gen.; ըստ
corresponding to κατὰ, առաջի to ἐνώπιον.
4. In accordance with what was mentioned under 2, there is only one case,
where in my evaluation loc. is clearly present, viz. Js. 4,16; a natural explana-
tion of that choice (not commented upon by Leloir) is that for the Armenian
translator the ἀλαζονεία (= false pretension, boastfulness etc.) is not the reason
for boasting, but the field within which boasting takes place (cf. the rendering
in the New English Bible: “you boast and brag” – as compared with, e.g. Js.
1,9-10: “… may well be proud that… must find his pride in…”). It may be
interesting to note that Js. 4,16 is the only text in the NT where the Vulgate
does not use gloriari to translate καυχᾶσθαι; here the verb is exultare.
5. In the single case, where ի + abl. is used (1 Cor. 3,21) the translator may
possibly have connected ἐν ἀνθρώποις with μηδεῖς, so as to give the sense:
“none among men” = no man – should boast.23
6. In 2 Cor. 11,30 I have taken the զ- to be the acc. mark – representing
τὰ in the Greek, the case of the noun then being gen. In order not to exclude
23 For ի + abl. = “ablativus partitivus” cf. H. Jensen, Altarmenische Grammatik, Heidelberg 1959, § 350
(and 460).
any possibility it may be mentioned that the Armenian could be read as the
preposition զ- with loc.
Table 3
Prepositional groups etc. with καυχάομαι/պարծիմ
A. Occurrences in the New Testament
Text Greek construction Armenian construction

Rom. 2,17 ἐν + dat. ի + acc./loc.
2,23 ἐν + dat. instr.
5,2 ἐπί + dat. instr.
5,3 ἐν + dat. ի + acc./loc.
5,11 ἐν + dat. ի + acc./loc.
1. Cor. 1,29 ἐνώπιον + gen. առաջի + gen.
2,31 ἐν + dat. ի + acc.
3,21 ἐν + dat. ի + abl.
2. Cor. 5,12 ἐν + dat. առ + acc./loc.
7,14 acc./ὑπὲρ + gen. acc./վասն + gen
9,2 acc./ὑπὲρ + gen. acc./վասն + gen
10,8 περί + gen. վասն + gen.
10,13 εἰς + acc. ի + acc./loc.
10,17 ἐν + dat. ի + acc.
11,18 κατὰ + acc. ըստ + dat.
11,30 acc. acc.
12,5 ὑπὲρ + gen. վասն + gen.
12,5 ὑπὲρ + gen. վասն + gen.
12,9 ἐν + dat. ի + acc. /loc.
Gal. 6,13 ἐν + dat. instr.
6,14 ἐν + dat. ի + acc.

Text Greek construction Armenian construction

Phil. 3,3 ἐν + dat. ի + acc.
Js. 1,9 ἐν + dat. ի + acc.
1,10 ἐν + dat. ի + acc.
4,16 ἐν + dat. ի + loc.
B. Distribution
Greek construction Number of cases Armenian Number of cases

construction
εἰς + acc. 3 ի + acc./loc. 3
ἐν + dat. 19 instr. 6
ի + acc. 6
ի + acc./loc. 4
ի + loc. 1
ի + abl. 1
առ + acc./loc. 1
ἐνώπιον + gen. 1 առաջի + gen. 1
ἐπὶ + dat. 1 instr. 1
κατὰ + acc. 1 ըստ + dat. 1
περὶ + gen. 1 վասն+ gen. 1
ὑπὲρ + gen. 2 վասն +gen. 2
acc./ὑπὲρ + gen. 2 acc./վասն + gen. 2
acc. 1 acc. 1
31 31
4. James 1,3
“τὸ δοκίμιον devient “probatio confecta” (vel “concinnata”)”, says Leloir.24
But it does not. What is here rendered confecta or concinnata is the Armenian
participle յաւրինեալ. But that certainly belongs to the following, not to the
preceding. Apart from Js. 1,3 the verb յաւրինեմ is only used once in the
24 Op. cit., p. 310.

Armenian NT, viz. 2 Cor. 9,11, so we are left without much comparative ma-
terial. It should be noted, however, that in 2 Cor. 9,11 it is used to translate
κατεργάζομαι, which is otherwise in most other cases rendered by գործեմ.
To my mind there is therefore no doubt that in Js. 1,3 the Armenian translator
has used a combination of the two verbs that are most natural as translations
of the Greek verb (and, incidentally, a construction that seems dear to him:
participle of one verb + finite form of another).
To this should be added the observation hinted at already: that three of the
verses we have now considered (1,3; 1,23 and 2,13) have one thing in common:
in all of them we have to do with a Greek verb beginning with κατα-. I would
certainly take the specific features of the Armenian renderings quoted to mean
that we have to do with a careful translator, but to my mind his carefulness
should rather be described in terms of considerations of how to translate
specifying or emphatic prefixes of Greek verbs than in terms of moral utility
as in Leloir’s comments. If this understanding holds true, it would point in the
direction of the translator’s Vorlage being a Greek text; but, of course, we could
not through that statement decide whether we are speaking of the Vorlage in
the hands of a primary translator or a secondary reviser.
5. James 2,21
Here the Greek ἀνενέγκας is rendered հանեալ … պատարագ. The addi-
tion of the noun պատարագ (= sacrifice) as object of the participle հանեալ
according to Leloir sets off “l’abnégation et l’esprit de religion d’Abraham”.25
Once again it seems recommendable to me to begin the evaluation of the Ar-
menian rendering with the question of the idiom of the translators and their
translational method.
The verb ἀναφέρω does not occur very often in the New Testament. As
appears from Table 4 the number of occurrences amounts to 10. Four Armenian
verbs are used to translate it (cf. Table 4, B 1). Table 4 (A) shows that in five
cases պատարագ appears as the object; in three of them the Greek text has the
corresponding θυσὶα, leaving only one incident besides Js. 2,21, where this is
not the case (Hebr. 7,27 (2°)).
Thus the statistics for ἀναφέρω are of little avail. More help may be gained
25 Op. cit., p. 311.

from a listing of the occurrences in the NT of the verb προσφέρω, which is

used much more frequently both inside and outside sacrificial contexts.
In order to evaluate the word statistics of Table 4, there is first of all reason
to underline the fact that neither ἀναφέρω nor προσφέρω has necessarily any
sacrificial meaning, the basic meaning of the verbs being “bring up; carry back”
etc. and “bring to; present” etc., resp. Most of the instances classified under
category c in Table 4 (A) appear in non-sacrificial contexts.
On account of this basic semantic fact it will often be necessary – or at
least natural – in a sacrificial context to supply the verb with an object mean-
ing ‘sacrifice’, esp. where there is no other means of indicating the sacrificial
content. This fact is mirrored in the relatively great number of instances under
category a in Table 4 (A).
What has been said here about the Greek verbs could immediately be
transferred to Armenian մատուցանեմ, as already appears (at least to a certain
extent) from the counting of Armenian renderings of the two Greek verbs.
Two observations should be added. First it should be noted that where the
Greek has an object other than δῶρον, θυσία etc. for what is sacrificed it would
not be natural to add one of the nouns mentioned, whereas the addition of
պատարագ as a “predicative accusative object” would not be an unnatural
construction in Armenian. This, in my evaluation, accounts for the majority of
instances listed under category b in Table 4 (A), and in other cases the reason
why the Greek has no object seems to be that one of the nouns has been used a
little earlier on in the same passage, so that Greek stylistic ideals would make
it undesirable to repeat the word – as opposed to Armenian stylistic usage.
Table 4
ἀναφὲρω and προσφὲρω in the Greek NT – Renderings in Armenian
A. Occurrences of պատարագ
a. Instances with պատարագ in Armenian, synonym in Greek.
b. Instances with պատարագ in Armenian, no synonym in Greek.
c. Other instances.
1. ἀναφέρω
a. Hebr. 7,27; 13,15; 1 Pet. 2,5 3 instances
b. Hebr. 7,27; Js. 2,21 2
c. Mt. 17,1; Mk. 9,2; Lk. 24,51 (vl); Hebr. 9,28; 1 Pet. 2,24 5
10 instances
2. προσφέρω
a. Mt. 2,11; 5,23; 5,24; 8,4; (Lk. 5,14 (vl)); Acts 7,42; 21,26; Hebr. 5,1; 8,3; 15 (16) inst.
8,4; 9,9; 10,1; 10,8; 10,11; 10,12; 11,4
b. Mk. 1,44; Lk. 5,14; Hebr. 5,3; 9,14; 9,25; 9,28; 11,17 7 (6)
c. Mt. 4,24; 8,16; 9,32; 12,22; 14,35; 17,16; 18,24; 19,53; 22,19; 25,20; 24
Mk. 2,4; 10,13; Lk. 18,15; 23,14; 23,36; John 16,2; 19,29; Acts 8,18;
Hebr. 5,7; 8,3; 9,7; 10,2; 11,17; 12,7
46 instances
B. Armenian verbs rendering ἀναφέρω and προσφέρω
1. ἀναφέρω
բառնամ 26 1 instance
հանեմ 3 instances
մատուցամեմ 4
վերանամ/վերացուցանեմ 2
10 instances
2. προσφέρω
ածեմ 1 instance
ածեմ + մատուցանեմ 1 –
մատուցանեմ 43 instances
(մերձենամ)27 1 instance
46 instances
26 բառնամ is used in Hebr. 9,28 in a quotation of Is. 53,12, where Z uses վերացուցանեմ (which is
reproduced in the quotation of the same verse in 1 Pet. 2,24). The translator in Hebr. may be influ-
enced by such texts as Is. 53,4 (φέρω/բառնամ) and John 1,29 (αἴρω/բառնամ).
27 մերձենամ, of course, cannot be considered a “synonym” of προσφέρω. It reproduces the variant
reading προσεγγίσαι in Mk. 2,4.
C. Greek synonyms for պատարագ

1. In texts quoted under A
δῶρον 8 instances
θυσία 9
προσφορά 2
19 instances
2. In other texts of the NT
δῶρον 6 instances
θυσία 6
προσφορά 7
19 instances
Whether the more frequent use of պատարագ than its Greek equivalents
could be described in terms of the sacrificial meaning having come to belong
more firmly and internally to the semantic field of προσφέρω than to that
of մատուցանեմ, or the description should rather be given in grammatical-
stylistic terms as suggested above, seems to me to be a mere question of lin-
guistic terminology; but it could be added that for հանեմ which may have
been chosen in Js. 2,21 as a common equivalent of ἀναφέρω, the need of a
պատարագ to denounce the sacrificial meaning has undoubtedly asserted itself
the more forcefully, as this verb is rarely used in sacrificial contexts.
Such considerations of semantics, grammar, syntax and style, then, to my
mind, are sufficient to explain the use of պատարագ in Js. 2,21, and it is again
a principle of sound method not to seek explanations from other fields – such
as psychology, ethics a.o. – for what is sufficiently explained on a linguistic
basis.
In the examples treated of so far it will have become clear that for such
reasons of method Leloir’s comments of a theological, moral or psychological
order have appeared to me superfluous. Before turning to the next examples
where the discussion will be primarily concerned with questions of a text-
historical character, I shall by way of a transitional excursus refer to a dis-
cussion of one of the occurrences of προσφέρω, where scholars have voiced
extremely varying opinions and rather far-reaching text-historical conclusions,
mainly on the basis of the presence of պատարագ in the Armenian version. I
am referring to a discussion that took place in the 1920’es and 1930’es of the
variant readings in Jesus’ words to the leper (Mt. 8,4 par.).
Excursus: The discussion of the readings in Mt. 8,4/Mk. 1,44/Lk. 5,14

The presence, resp. absence of a Greek synonym of պատարագ, will have appeared from
our Table 4, in so far as Mk. 1,44 and Lk. 5,14 are quoted under category b, whereas Mt.
8,4 – and a variant reading in Lk. 5,14 – come under category a. For the further evaluation
it may be useful here to quote the full clause in Greek, and in the version of the Zohrab
Bible as well:
Mt.: … καὶ προσένεγκον τὸ δῶρον ὃ προσέταξεν Μωϋσῆς …

Mk.: … καὶ προσένεγκε περὶ τοῦ καθαρισμοῦ σου ἃ προσέταξεν Μωϋσῆς …
Lk.: … καὶ προσένεγκε * περὶ τοῦ καθαρισμοῦ σου καθὼς προσέταξεν Μωϋσῆς …
(* vl add. τὸ δῶρον)
Mt.: … եւ մատո զպատարագն, զոր հրամայեաց Մովսէս …

Mk.: … եւ մատո վասն սրբութեան քոյ պատարագ, զոր հրամայեաց Մովսէս …
Lk.: … եւ մատո պատարագ վասն սրբութեան քոյ, որպէս հրամայեաց Մովսէս …
Viewed in a broader perspective the discussion earlier in this century of the Vorlage of the
Armenian Bible was strongly and provocatively promoted through Frédéric Macler’s book
about the Armenian version of the gospels of Matthew and Mark, published 1919.28 Macler
emphasized that “tous les faits concordent pour établir que la traduction arménienne, telle
qu’elle est fixée, a été faite sur un original grec”.29 To Macler’s followers now came the
task of identifying more precisely the text-type of the Greek original, whereas opponents
of his view tried to strengthen the case for Syrian influence. A. Merk, for example, in
articles published in 1923, 1924, and 1926,30 substantiated the dependence of the Armenian
version upon a Greek Vorlage through philological, palaeographic and other arguments.
In his last-mentioned article, however, he appends a chapter on the “syrische Einfluss”,
and he lists correspondences between the Armenian and the Syrian versions. Thus Mk.
1,44 is registered in the following way: σου + το δωρον arm syrP (syrs vac),31 i.e. the Ar-
menian is parallel to the Peshitta, whereas the vetus syra witness, Codex Sinaiticus, is silent
here. In Merk’s classification this belongs to the “eigenartige Übertragungen”, which are
“bemerkenswerter”, i.e. such peculiar translations that are more notable, sc. than other
cases where Latin or Greek witnesses support a Syrian-Armenian reading.
28 Cf. above, note 9.

29 Op. cit., p. 643.
30 A. Merk, Recension of F. Macler, op. cit., Biblica 4, 1923, pp. 220-229; “Die Einheitlichkeit der arme-
nischen Evangelienübersetzung”, Biblica 4, 1923, pp. 356-374; “Armenische und griechische Palaeo
graphie”, Miscellanea Francesco Ehrle IV, Studi e Testi 40, 1924, pp. 1-21; “Die armenischen Evangelien
und ihre Vorlage”, Biblica 7, 1926, pp. 40-71.
31 Op. cit. (1926), p. 64.
In the long and important article on the Caesarean text by Kirsopp Lake, R. P. Blake
and Silva New, published 1928,32 Blake classifies the reading of Mk. 1,44 under the
heading: “syriasms surviving in the Armenian”.33 There are two strange things about
Blake’s listing, for one thing that he faultily refers to Syrus Sinaiticus as one of the Syrian
witnesses, and secondly that he lists the reading both as a syriacism identified by Merk
and in his supplementary list.34
Lyonnet, however, in his presentation of the Armenian version as belonging to the
Caesarean text type, denies that there is any trace of Syrian influence here.35 In his con-
tribution to Lagrange’s Introduction to the New Testament, his general view and the
importance he attaches to the relative pronoun being in the singular is specified in the
following way: “Au v. 44, l’arménien n’avait pas plus besoin d’un modèle syriaque pour
traduire προςενεγκε … ο προςεταξε Μωϋσης par մատո … պատարագ զոր հրամայեաց
Մովսէս “présente… l’offrande qu’a prescrite Moïse”. La précision “l’offrande” se retrouve
dans d’autres versions même françaises, comme celle du P. Joüon; elle prouve seulement
que l’arm. lisait sans doute ὃ avec θ 565 W et non ἃ avec les autres manuscrits”.36
Lyonnet’s evaluation found little favour in Baumstark’s eyes. We are not confronted
with “eine äusserliche Hinzusetzung des Substantivs um dem Verbum matuc’anel “le
sens religieux d’”offrir””37 zu verleihen”, he comments. And immediately after he adds:
“Vielmehr liegt eine Harmonistik zugrunde, die ohne weiteres auf das “Diatessaron” als
Quelle weisen würde”.38
It would take us too far here to quote Baumstark’s further arguments and on the
whole to extend this excursus, the purpose of which has been to illustrate what pointed
and far-reaching text-historical conclusions have been based on one – or three – examples
of idiomatic Armenian usage of մատուցանեմ պատարագ.
It is difficult not to see an element of wishful thinking in the definitions of practically
the same phenomena as either “syriacisms”, “caesareanisms” or “tatianisms”. Obviously,
my above comments are most in line with Lyonnet’s views on Mk. 1,44. Unlike Lyonnet,
32 K. Lake, R. P. Blake and S. New, “The Caesarean Text of the Gospel of Mark”, Harvard Theological
Review 21, 1928, pp. 207-404.
33 Op. cit., p. 311.
34 Op. cit., p. 312.
35 S. Lyonnet, “Un important témoin du texte césaréen de Saint Marc: la version arménienne”, Mé-
langes de l’Université Saint-Joseph, Beyrouth, 19,2, 1935, pp. 23-66.
36 M.-J. Lagrange, Introduction à l’étude du Nouveau Testament, II. Critique textuelle, Études Bibliques, Paris
1935, p. 359.
37 Quoted from the work mentioned in note 35, p. 61.
38 A. Baumstark, Recension of S. Lyonnet, op. cit. (see note 35), Oriens Christianus 3. S, 11, 1936, pp.
245-252; here quoted from p. 247.
however, I feel reluctant to assign text-historical significance to the singular. Precisely

from the view that մատուցանեմ պատարագ forms one semantic unit, and from the
observations that պատարագ does not as a noun exclusively reproduce δῶρον, but other
expressions for sacrifice as well, and that the ἃ of Mk. 1,44 should be translated “the
things which”, “what” in the sense of “the sacrifice that”, I do not find that the Armenian
should of necessity, because of its singular, be connected with the reading ὃ.
But let us turn again to Leloir’s comments on the Epistle of James. After hav-
ing considered through our first five examples the importance of founding
the evaluation of a translation such as the Armenian on a careful investigation
of the translators’ language, its idiomatic characteristics, modes of expression
and limitations in rendering the “source language”, we now turn to examples
where proper text-historical questions are involved.
6. James 2,11
As regards this verse, Leloir concentrates on the fact that the Armenian version
quotes the commandments in the following order: first: Do not kill, second:
Do not commit adultery, whereas the Greek standard text has the reverse
order.39 Leloir refers this to considerations in the Armenian translator’s mind
of what is more serious: to kill or to commit adultery, and he adds some com-
ments concerned with the definitivity of homicide as compared with adultery
exemplified through the life and fate of Mussolini.
To my mind only two observations should be made concerning this textual
difference, first what Leloir mentions parenthetically, that there are other wit-
nesses for this reading. In other words: we might remain entirely within the
history of the text itself without venturing into for example the field of moral-
izing. The second observation that takes us a little outside text history proper,
but not so far as to leave the technicalities and probabilities of a translator’s
or a copyist’s work on the biblical text, is the following: the order in the Ar-
menian version (and some other witnesses) corresponds to the usual order in
the Old Testament. Thus if we should add any supplementary reason to that
mentioned first, the variant reading might very easily find its explanation in
an attempt of adaptation to the traditional order of the commandments.
39 Op. cit., p. 309.

7. James 2,23
This example takes us to a different level of text history. Here, because of the
use of the Armenian verb կատարեմ to translate the Greek πληρόω, Leloir
refers to the arm 1-version and through that to a possible Syrian influence.40 In
this connection Leloir refers to the treatment of the notion of “Vollendung” in
J. Molitor’s book on the Grundbegriffe der Jesusüberlieferung41 where it is stated:
“Vollendung bedeutet auch in der Muttersprache Jesu als umfassender Begriff
gleichzeitig Erfüllung, Wiederherstellung und Neugestaltung”,42 part of which
phrase is quoted by Leloir in order to describe the “nuance d’achèvement
complet des termes du syriaque et d’arm 1”.43
If Molitor’s above-quoted phrase is taken as an attempt to give a semantic
analysis with theological connotations of what is meant in Aramaic, when
words meaning “end, finish, complete, fill, fulfill” etc. are used, it may be
acceptable; but when such a description is used in order to maintain that the
Greek “synonyms” are poor vehicles that can hardly transmit the “dynamic”
Semitic way of thinking, this is a line of argument that was much en vogue a
generation or more ago,44 but should be considered untenable after the pub-
lication of James Barr’s Semantics of Biblical Language.45
Turning now to the Armenian evidence, it should be mentioned first that
St. Lyonnet is the scholar to whom the merit belongs of having drawn the at-
tention to the fact that in a number of texts, particularly Armenian translations
of Syrian and Greek fathers, the verb կատարեմ is used where the Zohrab
Bible has լնում, and Lyonnet takes such quotations to be evidence of a vetus
armena gospel text (“arm 1”).46
Secondly, it should be underlined that it cannot in my view, on purely
semantic grounds, be maintained (as does Leloir) that կատարեմ is a more
“dynamic” word than is լնում. One might even ask whether the relation is not
in the reverse order, for what is more dynamic: “to finish” or “to fill”?
Thirdly, it should be noticed that whereas Molitor and Leloir treat the
40 Op. cit., p. 307.

41 Joseph Molitor, Grundbegriffe der Jesusüberlieferung im Lichte ihrer orientalischen Sprachgeschichte, Düs-
seldorf 1968.
42 Op. cit., p. 108.
43 L. cit.
44 Cf. e.g. Thorleif Boman, Das hebräische Denken im Vergleich mit dem Griechischen, 1. Aufl., Göttingen
1952, 2. Aufl., Göttingen 1954.
45 James Barr, The Semantics of Biblical Language, Oxford 1961.
46 S. Lyonnet, op. cit. (see note 8), esp. p. 29.
Table 5
Distribution of πληρόω/կատարեմ, լնում
Mt Mk Lk Joh Acts Rom 1 Cor 2 Cor Gal

A. πληρόω in Greek NT – Renderings in Armenian
լի, լի առնեմ/եմ etc. 1 2 2 1
լնում 16 2 5 13 7 2
կատարեմ 1 1 3 9 2 1 1
Total 17 3 9 15 16 6 2 1
B. կատարեմ in Armenian NT – Greek synonyms

τελέω 5 3 2 1 1 1
ἀποτελέω
ἐκτελέω 2
ἐπιτελέω -. 1 3
συντελέω 2
τελειόω 2 3 1
τέλειος 3 1 1
Subtotal 8 1 9 5 2 1 1 4 1
πληρόω 1 1 3 9 2 1 1
ἀναπληρόω 1 1
ἐκπληρόω 1
συμπληρόω 1 1
Subtotal 2 1 4 11 2 1 2
ἀποκτείνω
καταρτίζω 1 1
κατεργάζω
ποιέω 1
Subtotal 1 1 1
Total 11 2 14 5 13 3 1 5 4
C. լնում in Armenian NT – Greek synonyms

πληρόω 16 2 5 13 7 2
ἀναπληρόω 1
ἀνταναπληρόω
προσαναπληρόω 2
πληροφορέω
Subtotal 16 2 5 13 7 2 1 2
πίμπλημι 1 11 9
ἐμπίμπλημι 1 1 1
Subtotal 1 12 10 1
γεμίζω 1 2 2
ἐμπνέω 1
Subtotal 1 2 2 1
Total 17 3 19 15 18 3 1 2
Eph Phil Col l Th 2 Th 2 Tim Heb Js l Joh 2 Joh Apc Total
2 8
4 2 2 1 1 1 56
2 1 1 1 1 24
4 4 4 1 1 1 1 1 2 88
1 1 5 20
1 1
2
1 1 6
2
1 5 1 13
1 1 1 3 1 12
3 2 7 6 1 5 56
2 1 1 1 1 24
1 3
1
2
2 1 1 1 1 1 30
1 1
2
1 1
1 2
1 1 1 6
1 3 2 1 2 8 7 2 1 7 92
4 2 2 1 1 1 56
1 2
1 1
2
1 1
4 4 3 1 1 1 62
21
3
24
1 6
1
1 7
4 4 3 1 1 2 93
semantic and linguistic questions in a very broad and general perspective,

Lyonnet specifically refers to instances dealing with the fulfilling of a prophecy
or the like; i.e. cases where the Greek πληρόω and its Armenian equivalent,
whether կատարեմ or լնում, means “fulfill/be fulfilled”, “make/ come true”.
As mentioned above Lyonnet’s book is only concerned with the gospels,
where he counts 29 instances with πληρόω, in the Zohrab Bible rendered լնում
in 26 cases, կատարեմ in only 3.47
In Table 5 are first listed all the occurrences of πληρόω in the Greek NT
and the Armenian renderings (A). Further, in order to get a fuller picture of
the semantic fields of the two Armenian verbs in question, are listed all the
occurrences in the Armenian NT of կատարեմ (B) and լնում (C) – with their
Greek “synonyms”.
Of course, such a contrastive listing does not necessarily exhaust the se-
mantic scope of the words in question. However, even this modest register
illustrates the fact that լնում, in accordance with its basic meaning “fill”, in no
less than 92% of the instances where it occurs, corresponds to Greek πληρόω,
πίμπλημι and derivatives. As could be expected, a high percentage of in-
stances with կատարեմ render Greek τελέω, τελειόω and derivatives (61%);
կատարեմ, however, in no less than 33% of the instances is used to translate
πληρόω and derivatives.
Presumably, however, the most interesting fact to be read out of Table 5
is the difference between the gospels and the rest of the NT. Thus while the
relation between կատարեմ and լնում (+ լի etc.) as renderings of πληρόω in
the gospels is 5:39 – or 11%:89%, for the rest of the NT the figures are 19:25 –
or 43%:57%.
Outside the gospels there are only four instances concerned with the ful-
filling of a prophecy, viz. three in Acts (1,16; 3,18; 13,27) and Js. 2,23, in all of
which is used կատարեմ.
Therefore, in my opinion, Leloir is certainly right in pointing to the text-
historical relevance of the occurrence of կատարեմ in Js. 2,23. Our modest
counting of words, however, raises the question whether a simple “revision
hypothesis” does justice to the facts. By a simple revision hypothesis I un-
derstand a theory maintaining that the early history of the Armenian Bible is
fully explained by assuming that the primary stage was a version of strongly
47 L. cit. Lyonnet’s figures are different from mine, because he only counts synoptic parallel readings
once.
Syrian affinity which was at a secondary stage revised according to the Graeca
veritas.
In view of the complications illustrated in the distribution of կատարեմ
and լմում as renderings of πληρόω such a hypothesis might seem an over-
simplification – or to put it more cautiously: adherents to such a theory must
bear the burden of explaining why the revision was carried out so differently
in different parts of the Bible.
As long as that explanation is not given, I would not allow a single oc-
currence of a word that has been appointed in some contexts to be a vetus
armena-symptom, but can in no technical sense of the word be called a Syrian
reading and does not for any clear semantic – or other – reason take us to
Syriac rather than Greek, to bear any value of proof of “Syrian influence”, let
alone “Syrian Vorlage”.
8. James 1,21
Our last example is rather closely related to the one just treated of. Concerning
Js. 1,21 Leloir maintains that the rendering of Greek σῶσαι through Armenian
կեցուցանել is an evident syriacism, and – again on the authority of J. Moli-
tor – he also claims that the use of կեցուցանեմ is evidence of an intention
to go beyond the Greek through the use of a more dynamic expression cor-
responding to Semitic (i.e. Syriac) usage. In Leloir’s own words the argument
runs as follows: “l’emploi, et en syriaque, et en arménien, d’un verbe qui
signifie “vivifier”, alors que le grec σώζω demandait simplement “sauver”,
marque une intention de dépasser le sens du grec; car “sauver” est évidemment
beaucoup moins positif que “vivifier”; le terme sémitiques (sic) est nettement
plus dynamique”, and “1,21: σῶσαι est rendu en arménien par “vivificare”
(kec’uc’anel), ce qui est un syriacisme évident; cfr. syp: dtaḥe, ut-vivificet”48
For a general evaluation of such an argument I shall here confine myself
to referring once again to James Barr49 and adding that it is not advisable to
extend what is characteristic of the idiom of the Syriac Bible to be “Semitic”
in general. Thus for instance in the Biblical Hebrew of the OT the prevailing
verb of salvation is ‫ ישע‬not the hifil of ‫חיה‬.
For a more specific evaluation of the renderings of σώζω in the Armenian
New Testament the basic fact to be registered is that seven different verbs are
48 Op. cit., pp. 307 and 304.

49 Cf. note 45.
used. Linguistically speaking, three of them can be defined as “syriacisms”.

First of all, the verb փրկեմ is a loan word (from Syriac ‫)ܦܪܩ‬, whereas the
use of կեամ (= “to live”) and its factitive formation: կեցուցանեմ to denote
“to be saved, to save” must be characterized as a translational loan from the
corresponding use of the Syriac ‫ ܚܝܐ‬and its afel,50 and Armenian կենդանի
առնեմ that is used once, must linguistically be defined in the same way as
կեցուցանեմ.51
Now, my main point of method is that it is inadmissible to draw conclu-
sions from linguistic syriacisms to textual syriacisms. For the words under
discussion here this to my mind is demonstrated very clearly from Table 6, in
which all equivalents of Greek σώζω in the Armenian and Syriac (Peshitta)
New Testament are listed.
Table 6
σώζω in the Greek New Testament – Renderings in Syriac (Peshitta) and Armenian
Armenian ապրիմ/ կեամ/ փրկեմ բժշկեմ կենդանի մնամ զերծանեմ Total

ապրեցուցանեմ կեցուցանեմ (փրկութիւն) առնեմ
Syriac:
‫ܚܝܐ‬ 40 29 14 1 1 1 86
‫ܦܨܐ‬ 4 2 6
‫ܦܪܩ‬ 1 4 5
‫ܐܣܐ‬ 4 1 5
‫ܚܠܡ‬ 2 1 3
‫ܫܘܙܒ‬ 1 1
om. 1 1
Total 48 30 25 1 1 1 1 107
The most conspicuous statistical facts to be ascertained in Table 6 are no doubt

that in Syriac ‫ ܚܝܐ‬accounts for around 80% of all instances, and that in Arme-
nian the three numerically most important verbs: ապրիմ/ապրեցուցանեմ,
50 For linguistic terminology and a description of how loan words and other linguistic loans are in-
corporated in languages in general, see Uriel Weinreich, Languages in Contact. Findings and Problems,
The Hague 1968, esp. pp. 47ff. Cf. also Louis Deroy, L’emprunt linguistique, Bibliothèque de la Faculté
de Philosophie et Lettres de l’Université de Liège, 141, Paris 1956.
51 Cf. also below, with note 52.
կեամ/կեցուցանեմ and փրկեմ account for, resp., 45, 28 and 23% of all in-
stances; and as կեամ/կեցուցանեմ and կենդանի առնեմ accounts for only 30
of the 86 instances (or 35%), where the Peshitta has ‫ ܚܝܐ‬and there is, therefore,
no specific textual correlation between ‫ ܚܝܐ‬and կեամ/կեցուցանեմ, the above
statement of method is manifestly corroborated.
It would take us too far here to go into a detailed analysis of the Armenian
terminology of salvation. Let me conclude by emphasizing two points of a
general nature and a few illustrating details of relevance in our context.
First it might be worth underlining that by the above I have certainly
not meant to characterize the Armenian translator’s choice of renderings as
haphazard, and secondly, that it would of course be absurd to deny that con-
nections between Syria and Armenia have been of great importance.
To illustrate the first point I shall just refer to two examples: 1) in cases
where Greek σώζω is used for “healing”, it is as natural for the Syrian trans-
lator to choose ‫ ܐܣܐ‬as for the Armenian to prefer բժշկեմ or փրկեմ, so it is
no accident (but of no text-historical relevance) that these words can be seen
to correspond in Table 6; 2) when in the gospels կեցուցանեմ is used in all
instances where the formula “your faith has cured you” occurs, there can be
no doubt that the translator has made a deliberate choice, whether or not we
are able to discern what were the determinative factors in his choice – be they
psychological, stylistic, liturgical, social or of another order.
In our context the importance of Armenia’s relations to Syria are obvi-
ous from the very syriacisms on the linguistic level that formed our point of
departure, and it is of course very important to try to specify within various
cultural sectors, language, architecture, religion etc., what is the exact amount
and content of what was taken over by the Armenians from Syrian life and
culture.
We have seen, however, that it cannot be expected a priori that conclusions
from one field of investigation can be transferred directly and immediately
to another. A couple of supplementary remarks on questions concerned with
the Bible text may elucidate that a little further.
If we had confined our countings conc. σώζω to the Epistle of James, the
general impression would have been a preference for փրկեմ, as this verb
is chosen in three cases, while ապրեցուցանեմ and կեցուցանեմ occur only
once each. The lack of correspondence with the Peshitta appears from Table
7, so that on numerical grounds the conclusion as to textual affinity would
have been the opposite of that maintained by Leloir. I do not pretend to have
substantiated through this paper a “probablement non” to the question of
Syrian Vorlage, but it will have appeared that I find that the “probablement
oui” is highly premature and may very easily turn out to be untenable.
Table 7
σώζω in the Epistle of James – Renderings in Syriac (Peshitta) and Armenian
Text Syriac Armenian

1,21 ‫ܚܝܐ‬ կեցուցանեմ
2,14 ‫ܚܝܐ‬ ապրեցուցանեմ
4,12 ‫ܚܝܐ‬ փրկեմ
5,15 ‫ܚܠܡ‬ փրկեմ
5,20 ‫ܚܝܐ‬ փրկեմ
In Molitor’s and Leloir’s presentations of the material they try to bring out
the “Semitic” and Armenian (+ Georgian) formation of the causative/facti-
tive forms of “to live” through translating vivificare, lebendig machen etc. This
of course is very useful by way of illustration for those sufficiently familiar
with the languages to know what is illustrated. But to others it might very
easily lead to the conclusion that Armenian is not able to distinguish between
σώζω and ζωοποιέω. And in that case the illustration is misleading. From
Table 8 it thus appears that Armenian which is a very rich language as regards
morphology and formation of words, has a number of constructions and de-
rivatives of the basic semanteme for “life”. Especially the “compound” of the
adjective կենդանի (= living, alive) + the verb առնեմ (= make), and the closely
related adjective կենդանարար are in use to translate ζωοποιέω. Only in one
case (2 Cor. 3,6) is chosen կեցուցանեմ, just as in one case (Hebr. 7,25) σώζω
is represented through կենդանի առնեմ.52 In view of the close relationship
between the numerous words derived from կեամ, կեանք and կենդանի that
are at the translator’s disposal, it may in this case be true to say that the two
“exceptions” quoted are those that prove the “rule”, viz. that σώζω is carefully
distinguished from ζωοποιέω.
A full presentation of Armenian terminology of salvation would, of course,
among other things include a listing of other Greek “synonyms” for the Ar-
menian verbs in use – corresponding to our listing in Table 5 of “fulfilling”-
52 Cf. the previous note.

terms etc.; it would also require some considerations regarding the fact that
the dispersion of translational choice as concerns the verbs is not paralleled in
the choice, where nouns are concerned; thus in all instances in the NT, where
σωτήρ and σωτηρία are used (24, resp. 46),53 the Armenian has the derivatives
of փրկեմ: փրկիչ and փրկութիւն.
Table 8
ζωοποιέω in the Greek New Testament – Renderings in Armenian
Rendering Texts Number of cases

կենդանի առնեմ John 5,21; 5,21; Rom. 4,17; Gal. 3,21 4
կենդանի եմ 1 Pet. 3,18 1
կենդանարար John 6,63; 1 Cor. 15,45 2
կենդանանամ /կենդանացուցանեմ Rom. 8,11; 1 Cor. 15,22; 15,36 3
կեցուցանեմ 2 Cor. 3,6 1
A profound study would also demand the perspective to be widened from the
“synchronic” analysis of the NT (if that may rightly be termed “synchronic”!)
to a “diachronic”, historical investigation including other translated and in-
digenous texts. The results of such an investigation might help to clarify what
elements can and must be explained within the field of semantics and history
of language, and what elements belong to the preserves of the text critic.
But such a programme of research certainly lies far beyond the limited
scope of this paper.
53 This statement is open to one interesting reservation: In 1 Thess. 5,9 εἰς περιποίησιν σωτηρίας
is translated: ի փրկութիւն կենդանութեան, i.e. formally փրկութիւն corresponds to περιποίησις,
կենդանութիւն (abstract noun to կենդանի, “alive”) to σωτηρία. In 2 Thess. 2,14 and Hebr. 10,39
περιποίησις is also rendered փրկութիւն, but in Eph. 1,14 εἰς ἀπολύτρωσιν τῆς περιποιήσεως is
rendered ի փրկութիւն նուաճութեան
Severian of Gabala
New Identifications of Texts in Armenian Translation
Severian of Gabala is one of the authors of the Old Church for whom the Ar-
menian transmission has been of greatest importance for the determination of
questions of authenticity, integrity of texts, etc. The Armenian translations of
homilies published so far can be summed up under four headings:
(a) The Aucher Collection,1

(b) The Akinian Collection,2
(c) Pseudo-Irenaeus,3 and
(d) Pseudo-Chrysostom.4
Before turning to the question of new identifications it might be useful to bear

in mind a few data concerning the transmission of the homilies known already.
First, it should be noted that the collection of ten homilies, which form the core
of Aucher’s edition, has a very broad attestation in Armenian MSS. In contrast
1 Severiani sive Seberiani Gabalorum episcopi Emesensis homiliae (ed. J. B. Aucher; San Lazzaro, Venice,
1827).
2 N. Akinian, “Die Reden des Bischofs Eusebius von Emesa”, Handes Amsorya 70-73 (1956-1959). For
the attribution of homilies 8-13 of this collection to Severian of Gabala, cf. H. J. Lehmann, “The At-
tribution of certain Pseudo-Chrysostomica to Severian of Gabala confirmed by the Armenian Tradi-
tion”, Studia Patristica 10 (ed. F. L. Cross; Texte und Untersuchungen zur Geschichte der altchristlichen
Literatur, Berlin: Akademie-Verlag), 107 (1970), 121-130; and H. J. Lehmann: Per Piscatores: Studies
in the Armenian version of a collection of homilies by Eusebius of Emesa and Severian of Gabala (Aarhus,
1975).
3 H. Jordan, Armenische Irenaeusfragmente, Texte und Untersuchungen zur Geschichte der altchristlichen
Literatur; Leipzig: J. C. Hinrichs, 36, 3 (1913), Text No. 29.
4 John Chrysostom, Interpretation of the Pauline Epistles (in Armenian; 2 vols.; San Lazzaro, Venice,
1862). The four Severianic homilies are the following: 2.694-715 (= Akinian XI, CPG 4202), 2.783-791
(CPG 4206), 2.883-891 (CPG 4195), 2.892-897 (CPG 4216). The CPG numbers quoted here and else-
where are those used in M. Geerard, Clavis Patrum Graecorum, vol. 2, Corpus Christianorum; Brepols-
Turnhout, 1974.
to that, the Akinian Collection as such is only found in one MS, New Julfa
No. 110 (Cat. No. 395), a twelfth-century MS; for Akinian’s homily No. XIII,
however, it should be noted that there exists an excerpt or a shorter version,
identical with Aucher’s homily No. XI.5 The one homily edited under the name
of Irenaeus was also published from one single MS (MS Vienna Mech. No.
2, fifteenth century), whereas the Pseudo-Chrysostomic group has a broader
attestation – together with the genuine Chrysostomic homilies in connection
with which they have been transmitted. Both the Pseudo-Irenaeus and the
Pseudo-Chrysostom belong to a later stage in the history of the Armenian
translators’ work than do the two first mentioned groups, the language of
which is pure and classical.
Before leaving the well-known editions of texts, it should be recollected that
homily No. X of Aucher’s edition belongs to Basil of Caesarea, that homilies
Nos. I, XIV, and XV of the same edition are attributable to Eusebius of Emesa,
and that Severian’s authorship of homilies Nos. XII and XIII in the Aucher
Collection has been questioned.6
Turning now to homilies that have not been published so far, it is natural
to begin with two homilies for which the main authorities are two MSS in the
Armenian Patriarchate in Jerusalem, MSS No. 1, dated A. D. 1417, and No. 154,
dated A. D. 1737, to which should be added two further nineteenth-century
5 Recently, M. van Esbroeck has published a translation of the Georgian version of this homily, which
is interesting in being much shorter than Akinian XIII without shortening as radically as is the case
in Aucher XI. See M. van Esbroeck, “Deux homélies de Sévérien de Gabala (IVe-Ve siècle) con-
servées en géorgien”, Bedi Kartlisa, Revue de kartvélologie, Paris, 36, 1978, 71-91, esp. pp. 90f.
6 Without any intention of drawing up an exhaustive list of references to the scholarly discussion of
the authenticity of these – and other – texts, the following titles should be mentioned: G. Dürks,
De Severiano Gabalitano (dissertation), Kiel 1917; J. Zellinger, Die Genesishomilien des Bischofs Severian
von Gabala. Alttestamentliche Abhandlungen 7.1, Münster i.W. 1916; J. Zellinger, Studien zu Severian
von Gabala. Münsterische Beiträge zur Theologie 8; Münster i.W. 1926; B. Marx, Severiana unter den
Spuria Chrysostomi bei Montfaucon-Migne, Orientalia Christiana Periodica 5, Rome 1939, 281-367; H.-D.
Altendorf, Untersuchungen zu Severian von Gabala (unpublished dissertation), Tübingen 1957. When I
include homilies XII and XIII of the Aucher Collection among texts for which Severianic authorship
must still be considered, this is due to the conviction that the utmost caution is required, before
negative conclusions regarding questions of authorship are drawn on the basis of internal criteria,
esp. for short texts, as mentioned below. A supplementary reason for mentioning them is that de
Aldama’s otherwise very useful and accurate manual of Pseudo-Chrysostomica (J. A. de Aldama,
Repertorium pseudochrysostomicum, Paris 1965) in the information given concerning CPG 4581 (de
Aldama No. 457), seems to confuse this homily with CPG 4247 = Aucher XII (and with CPG 4588 =
Aucher XIII, as far as his page references are concerned).
S everian of G abala – N ew I dentifications 69
MSS of the Mekhitarist library of San Lazarro, MSS No. 680/294, dated 1824-25,
and No. 1075/302, dated 1839-42.
The first of these homilies in the Jerusalem MSS bears the title: By Seweri-
anos, the Priest. Discourse on the Birth of Christ in Bethlehem in Judaea and on the
Adoration of the Wise Men. This homily is rather closely related to the Greek
homily In natale Domini nostri Iesu Christi (CPG 4657, PG 61, 763-768). The re-
lationship between the two versions will appear from the following survey:7
MS Jer. arm 1 Number of lines PG 61 Number of lines

col., line col., line
(a) 41a,48 – 42a,35 84 763,1 – 765,3 69
(b) 42a,36 – 42b,19 22 deest
(c) 42b,20 – 43a,2 31 765,4-26 23
(d) 43a,3-45 43 deest
(e) 43a, 46 – 43b,16 19 765,27-44 18
(f) 43b,17 – 44b,3 83 deest
(g) deest 765,45 – 768,12 141
Quantitatively the correspondence can be expressed in the following way: 134

out of a total of 282 lines of the Armenian homily find their parallel in 110
lines out of a total of 251 in the Greek “original”, or in other words: a little less
than half of the Armenian homily is identical with a little less than half of the
Greek text. Whether this fact could be explained by the assumption that two
excerptors have taken two-thirds each of an existing homily, combining them
differently, can hardly be decided, unless further evidence appears.
I cannot here go into a discussion of internal criteria for attributing the
homily to Severian, but it should be mentioned that there are two external
witnesses to his authorship to be added to the attribution in the Jerusalem MSS,
one in Greek and one in Armenian. We shall return below to the Armenian
evidence, which consists of two quotations within the series of fragments at-
tributed to Severian in MS Galata No. 54.8 One of these is from section (a), the
other from section (c) of the above survey. The Greek evidence is a quotation
7 For reasons of clarity I have only given references to one of the Jerusalem MSS. It should be noted,
however, that the number of variants between the texts of MS Jerusalem No. 1 and No. 154 is very
small. [Cf. below, p. 166ff.]
8 Cf. below p. 73f.
in Theodoret of Cyrrhus, who also gives the full name and title of the author:
Severian, bishop of Gabala.9 In spite of Theodoret’s evidence, which brings us
very close to Severian’s own time, some modern scholars have denied Sever-
ian’s authorship. This is true of B. Marx, J. Zellinger, and H.-D. Altendorf,
whereas W. Dürks, A. Wenger, and R. Laurentin support the attribution to
Severian.10 Those who deny it, mostly do so with reference to very general
arguments, saying that there is nothing particularly characteristic of Severian
to be found in the Greek homily.
Great caution is required before conclusions about authorship are drawn
from general observations on style and language, especially where short
homilies are concerned which may very likely have been given their form by
excerptors. Therefore, the Armenian evidence here referred to, seems to give
considerable support to the case for Severian’s authorship.11
As mentioned already, for the next homily again, MSS Jerusalem Nos. 1 and
154 are the principal witnesses. And again there are complications concerning
the relationship to the Greek tradition. Here, however, the text of the homily as
presented by the two Armenian MSS has a very close identity with one form of
a Greek homily to be found in a single MS, but – to the best of my knowledge
– never published. The Greek MS in question is a tenth-century MS in Trinity
College, Cambridge.12 Part of the Greek text has been edited, however, as ap-
proximately the first half of the homily CPG 4669: In illud: Ignem veni mittere
in terram (PG 62, 739-742) is identical with the first half of the homily found in
9 Theodoret of Cyrrhus, Eranistes, ed. Gerard H. Ettlinger, Oxford 1975, 181.

10 For the four first mentioned authors, cf. the references in note 6, above, and B. Marx, Procliana: Un-
tersuchungen über den homiletischen Nachlass des Patriarchen Proklos von Konstantinopel. Münsterische
Beiträge zur Theologie, Münster i.W., 23, 1940. For the two last mentioned authorities, cf. R. Lauren-
tin, Court traité de théologie mariale, Paris 1953.
11 It should be noted that the attribution to John Chrysostom is not only to be found in the Greek
MS material, but also in a number of quotations in Syriac, viz. in Severus of Antioch (cf. Sévère
d’Antioche, La polémique antijulianiste, ed. & trans. Robert Hespel, CSCO 244-245, Scriptores Syri
104-105, Paris and Louvain 1964, cf. esp. 245, 110.186.197. The Coptic tradition in one place contains
an attribution to John Chrysostom (cf. E. Porcher, “Analyse des manuscrits coptes 131, 1-8 de la
Bibliothèque Nationale”, Revue d’Egyptologie, 1933, 123-160, esp. pp. 124f., where also an “exégèse
de Sévérien de Gabala”, containing “allégresse au sujet de la naissance du Christ” is referred to – cf.
CPG 4282), in another to a Cappadocian bishop (cf. W. E. Crum, Theological texts from Coptic papyri,
Anecdota Oxoniensa, Semitic Series 12, Oxford 1913, 18-20), whereas the Georgian tradition seems to
put Epiphanius’ name at the head of this homily (cf. G. Peradze, “Die alt-christliche Literatur in der
georgischen Überlieferung”, Oriens Christianus 3, 1930, p. 86, note 9 [here quoted from CPG – ad
4657]). [Cf. below, pp. 155ff for the question of abbreviation technique]
12 MS Trinity Coll. Cambridge B.8.8., fol. 274v-277v.
the Greek Cambridge MS and the Armenian MSS of Jerusalem, whereas the
second half of Migne’s text is an excerpt from the homily De Pharisaeo, PG 59,
589-592.13
The exact correspondences are as follows:
PG 62, 739 init.–741, 41: Αἱ τῶν νηπίων … χλοηφορεῖν αὐτὴν ποιήσω
= MS Trinity Coll. B.8.8. fol. 274v-276r
= MS Jerusalem arm. 1, 46b, 7 – 48a, 10.
The Greek tradition – including a quotation in the catena on Luke14 – is
unanimous in attributing the homily to John Chrysostom, whereas the Ar-
menian witnesses quote Severian as the author. In the Jerusalem MSS the
author is referred to as “Severian the Priest”, i. e. the same designation as in
the homily just treated of; one of the Venice MSS says “Seberianos, bishop of
Gabala”, and the other “Severianos”, only.
The Venetian MSS present one complication, in that they add an exordium,
two thirds of which is identical with the opening paragraph of PG 60, 759-764
(CPG 4629): De remissione peccatorum. The section covered by the MSS of San
Lazzaro (where a few lines have been added, to which I have not been able to
trace any equivalent) is the following: PG 60, 759,1-20: Μίαν ἔχουσα ἡ πηγή …
οἱ ἐσκοτισμένοι φωτίζονται. (Ὄτι δὲ ταῦτα οὕτως ἔχει.)
Of course, it would be precipitate to infer Severianic paternity of PG 60,
759-764 from the occurrence of these 20 lines in the two Venetian MSS. I
cannot here go into any detailed discussion of internal or other criteria that
might be added. As possible starting points for an analysis with a view to
determining the question of authorship for this homily I might be allowed
to quote the following three items: 1) the dogmatic opponents referred to are
Arians, Eunomians, and Pneumatomachoi,15 which would fit in very well with
Severian’s theological position; 2) the way in which, in commenting upon
13 Cf. CPG ad 4669.

14 A. Cramer, Catenae Graecorum Patrum in Novum Testamentum, Oxford 1844, 2.105.
15 The latter group is here referred to as Marathonians (cf. W. Ensslin in Paulys Realencyklopädie der
klassischen Altertumswissenschaft (new edition by G. Wissowa & W. Kroll; Stuttgart, 1893ff), 14.2. col.
1430f). The three groups of heretics mentioned here are also referred to in PG 59, 569 (where the
Scythians are added) and in PG 61, 774 (in an interesting exegesis of the Parable of the Sower taking
the three groups of heretics to be those sown by the wayside, among thorns, and on stony ground,
respectively). In both cases the designation of the Pneumatomachoi is “Marathonians”, leaving
us with an interesting connecting link between the three homilies, CPG 4584, 4629, and 4660, that
might deserve further attention, also when the questions of authorship for the three homilies are
considered.
Matt. 18,18 and 18,19, the homilist collects examples from all parts of the Bible,
where λύω/δέω and συμφωνέω are used, is very reminiscent of the exegetic
(or homiletic) method of collecting “testimonia” characteristic of Severian;16 3)
finally, it should be noted that this homily appears in a collection containing
much Severianic material.17
Thus, if there are, admittedly, further questions to be solved concerning
the introduction in the Venetian MSS and concerning CPG 4629 on the whole,
before a final verdict on the question of authorship could be given, then it
should be emphasized, that, for the bulk of the homily – CPG 4669 – in the
version witnessed by the four Armenian and the one Greek MS, the unanimous
attribution to Severian in the Armenian material highly strengthens the case
for his paternity.
Leaving aside now the Armenian transmission of entire homilies – or at
any rate excerpts of such a length that they present themselves – and have
been used – as homilies, I turn to two series of fragments to be found in MS
Galata no. 54 (fourteenth century). Quantitatively these series do not furnish
us with a great amount of new material – comprising only a little more than
thirty pages in the MS; but a number of the fragments support identifications
made already, and others invite to new identifications or contain unknown
material.
The first series – given under the name of “Seberianos of Emesa” – contains
20 fragments from eight homilies of the Aucher Collection. I have treated of
these fragments elsewhere,18 so I shall confine myself here to repeating that
there are highly interesting coincidences between the excerptor of MS Galata
no. 54 and the author of the famous florilege, the “Seal of Faith”, as concerns
their quotations from Severian’s homilies.
The second series in the Galata MS is given under the name of “Seberianos,
bishop of Gabala”.19 It contains eleven fragments from six different homilies.
I shall give a brief survey of the contents of this series of fragments. I should
16 Cf. Lehmann, Per Piscatores, 292 (see note 2).

17 I am thinking of the collection of texts in MS Berlin, Phill. 1438, cf. R. E. Carter, Codices Chrysosto-
mici Graeci. Codices Germaniae, Documents, Études et Répertoires publiés par l’Institut de Recherche et
d’Histoire des Textes, 14, Paris 1968, 2.15.
18 Henning J. Lehmann, “Severian of Gabala – Fragments of the Aucher Collection in MS Galata 54”,
Haig Berberian Memorial Volume (in press). [Cf. below, pp. 95ff]
19 Only the first four letters of the name of the town are readable, but as they are Gaba(…, it is easy to
conjecture the two missing letters.
like to note at once a fact that certainly adds to the value of this collection, viz.
that for each homily quoted, both title and incipit is given.20
Below is given a list of titles, incipits, and the amount of correspondence
established; and a few remarks on the main characteristics will be added.
(1) Ի ճառէն որ ի մայր որդւոցն Զեբեդիա. յորմէ թէ վկայիցն ..ւն21 (cf. CPG
4249)
(a) 375,15 – 376,18: Եւ նա ասէ … ի կամաց նորա = Jordan 32,9 – 33,19
(2) Ի ճառէն որ ի ծնունդն Քրիստոսի. յորմէ թէ յորժամ ի ձմեռնային (cf. CPG
4657)
(b) 376, 19-25: ծնուցիչքն շաւշափէին … առանց ձեռին մարդոյ22
= MS Jerusalem arm. 1, 42a,26-34
(c) 376,25 – 377,7: Քրիստոս ծնեալ ի կուսէ … էւ ոչ ապականել
= MS Jerusalem arm. 1, 42b,20-40
(3) Ի ճառէն որ վասն ննջելոյն Յիսուսի ի նաւին. յորմէ թէ խորագնաց նաւորդք
(cf. CPG 4699, PG 64, 19-22: Οἱ πελάγιοι πλωτῆρες)
(d) 377, 9-19: ի բուն էր տէրն … դարձուցանէր
= PG 64, 21, 26-37: Ἐκάθευδεν ὁ Κύριος … ἐπιστρέφοντα
(4) Ի ճառէն յոր թէ զիարդ սա գիրս գիտէ ոչ ուսեալ. յորմէ թէ բեր դարձեալ
(cf. CPG 4201, PG 59, 643-652: Φέρε πάλιν τῶν εὐαγγελικῶν)
(e) 377, 21-24: զի ոչ զամենայնն … ընդ անհաւատս
= PG 59, 645, 58-61: Οὐ πάντα γὰρ … διαλέγεται ἀπίστοις
(f) 377, 24-33: առ հրէայս … մարտուցեալս
= PG 59, 645, 64-74: Πρὸς Ἰουδαίους … τῷ Πατρὶ μαχομένους
20 The same, incidentally, is the case for the quotations from the Aucher Collection.
21 It is difficult to read the word վկայիցն in my microfilm of the MS. I have no doubt, however, that
Babgēn’s reading of the incipit of the text is correct (Babgēn, Catalogue des manuscrits de la biblio-
thèque nationale arménienne de Galata (in Armenian; Antilias: Catholicossat arménien de Cilicie, 1961,
329); the second word, the reading of which also caused difficulties for Babgēn, seems to have
included two or three letters before the -ւն. Enough is readable, however, to make it clear that the
incipit cannot be identical with that in Jordan’s edition; this fact combined with the ascertainment
of the incipits of the other homilies in this series being rendered very correctly in MS Galata no. 54,
suggests the conclusion concerning the incompleteness of the homily in Jordan’s edition as stated
below.
22 There is no marking in the MS to keep the two fragments (b) and (c) apart, as is otherwise the case,
where more than one fragment from the same homily is quoted (usually through the excerptor’s
յետ սակաւուց or the like). Thus it is only on the basis of the Jerusalem MSS that it has been pos-
sible to ascertain that this excerpt is compounded of two fragments.
(g) 377,33 – 379,8: բայց զի կարծիցի … այլ միաբանեն միտքդ

= PG 59, 646,42 – 647,20: Ἀλλ’ ἳνα μὴ νομισθῇ … ἀλλὰ συμφωνεῖ τὰ
νοήματα
(h) 379, 9-16: յանդիմանէ զհրէիցն … ընդ հաւր մարտնչի
= PG 59, 647,72 – 648,3: Ἐλέγχει Ἰουδαίων … θεῷ μάχεται
(5) Ի ճառէն որ ի տէրունական խաչն եւ ի հոգին սուրբ եւ յերրորդութիւնն.
յորմէ թէ երէկ մեզ բանն
(cf. CPG 4196, PG 56, 499-516: Χθὲς ἡμῖν ὁ λόγος)
(i) 379, 18-23: բայց սակայն … յաղաքս ինքեան
= PG 56, 504,61 – 505,6: Πλὴν ὄταν ἀκούσῃς … ἀξία δὲ Θεοῦ δι’ ἑαυτόν
(6) Ի խաւսիցն, յոր թէ խոստուանիմ զքեզ հայր, տէր երկնի եւ երկրի. եւ ի
տեսութիւնն Դանիէլի, յորմէ թէ երէկ մեզ երանելին Ամբակում
(cf. CPG 4295, 17a, CSCO 102, 237)
(j) 379,25 – 381,21: Այլ է խոստուանութիւն մեղուցելոցն … ամենեցուն
անուանեալ
(k) 381,21 – 384,6: Դու ես որ հանէր զիս … զի նմանէ է
379,33 – 380,14: Եւ ետէ ոք հաւաստեաւ … զի զմեզ յարուսցէ
= CSCO 102, 237,21-23: Et si quis accurate … ut nos suscitet
The first fragment stems from the Pseudo-Irenaeus. Here the incipit is at
variance with that in Jordan’s edition, and I would take that to mean that the
homily is not complete, as it appears in Jordan’s edition.23
Then follow two fragments of the homily treated of above on the Birth of
Christ, giving the same incipit of the homily as in the Jerusalem MSS. As was
mentioned already, this fragmentary evidence to my mind strongly supports
the attribution to Severian.
Thirdly we meet with a fragment from a homily on Matt. 8,24. The Greek
original of the fragment quoted is found in PG 64,21; i. e. it belongs to the
homily CPG 4699. On the basis of the Greek text it has been suggested that this
homily should be attributed to Proclus.24 To my mind, however, the reasons for
this attribution are not so cogent as to weigh more heavily than the Armenian
evidence for Severianic paternity, but it should be noticed concerning the Greek
homily that once again we are confronted with a very short text, presumably
an excerpt, which makes the argument from internal criteria problematic.
In the fourth place, in the Galata MS we meet with four fragments from
23 Cf. note 21.

24 Cf. B. Marx, Procliana, 73 (see note 10).
a homily, which has with very good reason been attributed to Severian by
modern scholars.25
The fifth homily quoted was attributed to Severian already in the sev-
enteenth century by J. Sirmond, who was followed by Montfaucon. In the
twentieth century this attribution has been substantiated very fully, especially
by J. Zellinger.26
Finally, two excerpts from a sixth homily take up almost exactly as much
space as do the fragments of the five homilies mentioned until now. This quan-
titative fact is the more valuable since we are here confronted with a homily of
which only a small fragment was known beforehand. The fragment in question
has been transmitted in Syriac only, by Severus of Antioch.27 Severus and the
excerptor of MS Galata No. 54 agree, both in the attribution to Severian and
as concerns the title of the homily in question.28 As, furthermore, the contents
of the Armenian excerpts seem to me very Severianic, I find that this homily
is in reality one of those, for which Severianic authorship is least disputable.
For an overall estimation of the value of the MS Galata-quotations it should,
of course, be kept in mind that in the first series of excerpts, Aucher’s homily 1,
which is not by Severian, has been included; an inclusion of a spurious homily
in a collection such as those treated of here is thus, of course, a possibility that
should always be taken into account. On the whole, however, it seems to me
that there are so many indications of Severianic authorship for the group of six
homilies quoted by the Galata excerptor, that the appearance of a text in this
series is in itself rather a weighty positive argument for Severianic authorship
in cases where there is little or no supplementary evidence for this.
By way of rounding off it might be reasonable to point to the fact that the
registration of Severianic texts in this paper does not – as no such register
should – claim to be exhaustive. It should be noted explicitly, however, that
I have omitted references to a number of fragments that exist in florilegia,
catenae, and collections of ecclesiastical canons. The reason for that is that I
25 Cf. Marx, OCP V (see note 6), 309-314, and Altendorf, Untersuchungen, 146f.
26 Cf. Zellinger, Studien (see note 6), 27-34, and Altendorf, Untersuchungen. Cf. also A. Wenger, “Hésy-
chius de Jerusalem”, Revue des Etudes Augustiniennes 2, 1956, 461, and J. Kirchmeyer, “L’homélie
acéphale de Sévérien sur la croix dans le Sinaiticus Gr. 493”, Analecta Bollandiana, 1960, 78, 18-23.
27 Severi Antiocheni Liber contra impium Grammaticum III, 1 (ed. & trans. J. Lebon, CSCO 101-102, Scrip-
tores Syri 50-51, Paris and Louvain 1933.
28 In Lebon’s translation the title runs as follows in Severus: Ex homilia in illud: Confiteor tibi, pater
domine caeli et terrae (Matt 11:25), et de visione, quam vidit Daniel in Susa civitate, Severi Antiocheni
Liber …, CSCO 102, 237.
have not been able to go through this material in a systematic and comprehen-
sive way. As is known, parts of this material have been published already,29
whereas other elements have only been hinted at or registered in a more or less
provisional manner.30 Presumably, also, a still secret portion awaits its finder.
My humble purpose has been to point out for some texts, that exploration of a
number of Armenian MSS seems to provide a more solid basis for identifying
their instrument of origin with the stylus of Severian.
29 Cf. e. g. V. Hakopyan, Kanonagirk’ hayoc’, Erevan 1971, 2,288f; Aucher, Severiani … homiliae, xviii. Cf.
further the notes in CPG – conc. Nos. 4295, 5; 4295, 6.
30 Cf. e. g. A. Zanolli, Di una vetusta catena sul Levitico perduta in greco e conservata in armeno …, Venice,
San Lazzaro 1938, and R. W. Thomson, “The shorter recension of the Root of Faith”, REArm, 5, 1968,
250-260. Cf. further the note in CPG conc. No. 4194.
n Important Text Preserved in MS
Ven. Mekh. No. 873 Dated A.D. 1299
A
(Eusebius of Emesa’s Commentary on
Historical Writings of the Old Testament)
The collection of texts contained in MS No. 873 of the Mekhitarist library of San
Lazzaro, Venice (dated A.D. 1299),1 is clearly intended as a tool for the study
of the historical writings of the Old Testament. The commentaries found in the
manuscript fall into four groups. First come Ephraem’s commentaries on the
books of Joshua, Judges, Samuel, Kingdoms, and Chronicles (pp. 3-137). Then fol-
lows a commentary on the Pentateuch plus the historical writings mentioned,
except Chronicles (pp. 137-235), in the manuscript ascribed to Cyril of Alexan-
dria. Ephraem’s commentary on the Pentateuch makes up the third section
(pp. 235-433); and finally follows a catena on Leviticus (quoting such authors
as Origen, Apollinarius, Eusebius of Emesa, Cyril of Alexandria, Severus of
Antioch, et al.) (pp. 434-507).
For a comprehensive description – in Italian – of the manuscript one must
turn to Zanolli’s book (of 1938) on the Leviticus catena. The full title of Zanolli’s
book runs as follows in English translation: About an old catena on Leviticus,
which is lost in Greek, but preserved in Armenian; about its close relation to
Procopius of Gaza’s commentary; and about the three codices of S. Lazzaro
containing the text.2 In the Mekhitarist edition of 1980,3 to which we shall
return, a description of the manuscript is given in Armenian.
As appears from the title of Zanolli’s book, he was particularly concerned
with the assistance that can be gained from our manuscript, especially its
fourth part, the Leviticus catena, for the study of Procopius’ commentary on the
1 According to Almo Zanolli: Di una vetusta catena sul Levitico, perduta in greco e conservata in armeno,
della sua stretta relazione col commentario di Procopio di Gaza e dei tre codici di S. Lazzaro, che la conten-
gono, Venezia: Prem. Tipografia Armena, 1938, 78.98, the last part (pp. 434-507: the Leviticus catena)
may have been copied a few years later.
2 Cf. note 1.
3 Eusèbe d’Emèse: Commentaire de l’Octateuque, préparé par P. Vahan Hovhannessian, Venise: St. La-
zare 1980.
first books of the Old Testament and other Greek catenae of the early Middle
Ages.
We shall here be concerned with the second text – or group of texts – to
be found in the MS, the pseudo-Cyrillic commentary on historical writings of
the Old Testament.
The honour of having demonstrated that the Armenian translator or copy-
ist – or his Vorlage – is not justified in attributing the text to Cyril of Alexandria,
belongs to Father Vahan Hovhannessian of San Lazzaro; it was done in an
article in Bazmavēp in 1923;4 and through a lifetime, Father Vahan was con-
cerned with the text, so, undoubtedly, the Venice edition of the text, which –
as mentioned already – did not appear until 1980, i.e. after Hovhannessian’s
death, rightly bears his name on the title page.
In his first article concerned with our text (also in Bazmavēp 1923)5 Hovhan-
nessian already questioned Cyril’s authorship. He considered the possibility of
Eznik being the author, but finally – in his second and following articles6 – on
the basis of catena quotations, he reached the right conclusion: that the com-
mentary is by Eusebius of Emesa.
Almo Zanolli still voiced some doubts as to the question whether the text
could be attributed to Eusebius in its entirety.7 In my view, however, the
arguments on the basis of correspondence with catena fragments are quite
definitive. The fragments can with certainty be attributed to Eusebius and
the correspondences can to-day be expanded and substantiated much more
coherently and comprehensively than Hovhannessian and Zanolli were able
to do, so that the extent of possibly non-authentic material is very limited – if
existent at all.
The circumstances and events of Eusebius of Emesa’s life and the contents
of his work are not too well-known.8 He was born in Edessa, presumably
4 V. Hovhannessian, “Commentarius in Genesim d’Eusèbe d’Emèse”, Bazmavēp 81 (1923) 353-358 (in

Armenian); cf. Henning J. Lehmann: Per Piscatores. Studies in the Armenian version of a collection of
homilies by Eusebius of Emesa and Severian of Gabala, Aarhus 1975, 17, 31-33.
5 V. Hovhannessian: “Commentarius in Genesim de S. Cyrille patriarche d’Alexandrie”, Bazmavēp 81,
1923, 225-228 (in Armenian).
6 Cf. note 4. The article is continued in: Bazmavēp 82, 1924, 3-6, 33-36, 65-68, 225-228, and a summary
is given and a few pages of the text published in: Bazmavēp 93, 1935, 345-352 (in Armenian).
7 Zanolli, Catena sul Levitico, 17, 83-86.
8 A biographical sketch, where the sources are exploited to their utmost capacity, can be found in:
E. M. Buytaert: L’héritage littéraire d’Eusèbe d’Emèse, Bibliothèque du Muséon 24; Louvain: Bureaux du
Muséon, 1949, 43-96.
An Important Text Preserved in MS Ven. Mekh. No. 873 Dated A.D. 1299 79
around A.D. 300. He was educated both in Antioch and Alexandria, and it is
related in particular that he was trained in biblical studies by Patrophilus of
Scythopolis and Eusebius of Caesarea, presumably around 325. Some sources
say that Eusebius was not too well received as a bishop of Emesa (a town in
Phoenicia, to-day Ḥoms); the precise reasons for the upheavals are difficult
to decipher. Eusebius is said to have accompanied the Emperor, Constantius
during one – or more – of his campaigns against the Persians, but whether
this should apply to the campaigns before 350 or those of the years 357-360,
is hard to tell. According to Jerome9 Eusebius died under Constantius, which
would take us to a year before 361; in 359 Emesa is represented by another
bishop at the Council of Seleucia, so maybe Eusebius' death should be dated
even earlier than that.
On the basis of a piece of information given by Jerome10 he is remembered
as Arian – or at least semi-Arian – in theology, and this theological reputa-
tion of being semi-heretical may have earned Eusebius the ill fate of his writ-
ings encountered in the history of transmission. Today we know of only one
single text preserved in its entirety in Greek, for which Eusebian authorship
is claimed: a homily “on repentence”,11 but even here there are reasons to
doubt that he is in fact the author. I shall return to the indirect Greek transmis-
sion – in catenae; otherwise we have to turn to translations. Apart from a few
fragments in Syriac and a couple of texts in Georgian, it is the Latin and the
Armenian branches of transmission to which we owe most of our knowledge
of Eusebius. A number of homilies now form the core of Eusebius’ literary
production as known to us.12
I took as my starting point MS Ven. Mekh. No. 873. This, indeed, is by
far the most important witness for the Armenian translation; only one other
manuscript is known to contain the text in its entirety, namely MS No. 231 of
the Mekhitarist library of Vienna, and this manuscript is a nineteenth-century
copy of the manuscript of San Lazzaro. For about the last third of the text there
9 Hieronymus: De viris illustribus, ed. W. Herding, London 1924, 54; cf. Buytaert, L’héritage littéraire, 6,
94.
10 Hieronymus: Chronicon, ed. R. Helm, in: Eusebius: Werke, 7, GCS 24, J. C. Hinrichs, Leipzig 1913,
236; cf. Buytaert, L’héritage littéraire, 7, n. 9.
11 Buytaert, L’héritage littéraire, 150-156, 16*-29*.
12 See especially Eusèbe d’Emèse: Discours conservés en latin 1-2, ed. E. M. Buytaert, Spicilegium Sa-
crum Lovaniense 26-27, Louvain 1953-1957, and N. Akinian, “Die Reden des Bischofs Eusebius von
Emesa” Handes Amsorya 1956, 291-300, 385-416; 1957, 101-130, 257-267, 357-380, 513-524; 1958, 1-18,
19-22. Cf. Lehmann, Per Piscatores, 37-272.
is one further witness: a fifteenth-century manuscript of the Matenadaran in

Yerevan (MS No. 1267). Here also the author is given as Cyril; I have not had
the opportunity to examine the Yerevan manuscript myself, but judging from
the Venice edition, its variant readings do not show many important differ-
ences from the manuscript of San Lazzaro.
The Venice edition does not list indirect Armenian witnesses – such as the
quotations in the Leviticus catena of the very same manuscript as that contain-
ing the commentary, so it may be true that a certain amount of editorial work
still remains to be done. However, the most important thing to be said about
the Venice edition is that it has made this important text available to patristic
scholars, church historians and other readers.
Parts of the Armenian translation have in fact been available for a number
of years, namely in T̒.T̒oṙnian’s anthology of classical Armenian texts – here
of course under the name of Cyril of Alexandria.13 As no translation has been
available and the number of patristic scholars who know Armenian is rather
limited, the impact of the editions has so far been very modest. A translation
into a language more widely known than Armenian may very well be the
most urgent desideratum as regards this text.
In this paper I shall be concerned with an attempt to list and explain what
other tasks seem to me to be the most important to be undertaken in the wake
of the Armenian edition of Eusebius of Emesa’s commentary on historical
writings of the Old Testament.
Within the framework of a conference on Medieval Armenian Culture it
might be natural to give prominence to the question of the use of the text in
medieval Armenian literature. Since the text as described above has only been
known to a modest extent even where Armenologists are concerned, very little
has been done in the direction of searching through the works of medieval
Armenian exegetes for this purpose.
I have not myself gone into that question so far, so I would like to round off
this paragraph of my paper with an appeal to experts on medieval Armenian
exegesis to be aware of the possibility of finding quotations or allusions to
the text in Armenian authors. I should be most graceful for hints about such
findings.
For the Greek transmission of the text undoubtedly the most important
field to be examined is that of the catenae. The Latin word catena – meaning
13 T’adēos T’oŕnian: Հատընտիր ընտերցուածք ի մատենագրութեանց նախնեաց, Vienna, 1866,

386-423.
‘chain’ – was chosen to designate commentaries consisting of quotations from

earlier authors strung together as links of a chain. One of the first examples of
this genre, dating to the early sixth century, was Procopius of Gaza’s commen-
tary on the first books of the Old Testament. A number of others followed, and
as could be guessed, this traditionalist type of literature was in later centuries
often despised because of the lack of originality and spiritual activity inherent
in its making.
For one purpose, however, scholars of later centuries found the catenae
highly useful; namely, for identifying fragments of earlier exegetic works,
a number of them no longer extant in their entirety. It could be added that
exegetes seemed more liberal than other theologians in using and including
quotations of heretical authors in their works. Thus, the image of Apollinarius
might have been still more blurred, had it not been for the catenae.
Usually, from the outset, the catenist would quote the author’s name with
each quotation, and when linking fragments from the same author, he would
say “by the same”. This technique, of course, when abbreviations and additions
took place in the course of transmission, might lead to wrong attributions. Iden-
tifying the authors used by Procopius in his catenae is a particularly intricate
process since the names of authors are never given. Instead the quotations
have been combined into a continuous commentatorial text.
What has been said here may have been common knowledge; I have in-
cluded these basic facts, among other reasons, because modern catena research
has in fact had to re-evaluate a number of basic and elementary theories and
assumptions.
This re-evaluation – concerning the distribution of manuscripts in families
and branches of tradition, the attribution of fragments to their right authors,
the preferable technique of edition, etc. – is particularly the outcome of years
of meticulous work by Françoise Petit of Louvain. It must be hoped that she
can accomplish the huge task of edition; so far only one volume has been
published, namely the Sinaiticus Catena on Genesis and Exodus (published
1977).14
14 Catenae graecae in Genesim et in Exodum. I. Catena Sinaitica, ed. Françoise Petit, CCSG 2; Turnhout
1977. For articles and reports by Françoise Petit prior to 1977 reference may be made to this edition,
XI. From the years after 1977 the following articles can be mentioned from her pen: “L’édition des
chaînes exégétiques grecques sur la Genèse et l’Exode”, Le Muséon 91, 1978, 189-194; “La tradition
de Théodoret de Cyr dans les chaînes sur la Genèse”, Le Muséon 92, 1979, 281-286. In Buytaert,
L’héritage littéraire, 95*-l43*, the Greek catena fragments of Eusebius’ commentary as known in 1949
are published. For a better edition see: Robert Devreesse: Les anciens commentateurs de l’Octateuque
An impression of the mutual benefit of the study of the Armenian Eusebius

translation and research into the Greek catenae can be gained from the surveys
given in appendices I–III.
In the first survey (appendix I) are listed – in the third column – the identi-
fications of passages in the Armenian text with passages in the Catena Sinaitica
in Françoise Petit’s edition. It should be noted that the first passage of this
catena comments upon Gen 12,17, the last one upon Ex 2,18-22; hence the limit
of our survey.
Column 4 shows further identifications in Procopius as edited in Migne’s
Patrologia. This edition, however, is very unsatisfactory, even if it is better for
Genesis than for the rest of the books commented upon. In any case, identi-
fications on the basis of this edition cannot be expected to be exhaustive, but
they can give a first hint of possibilities. A full examination of Procopius must
be done on the basis of the Greek manuscripts available.
After the first survey, there follows a tally of fragments published under
Eusebius of Emesa’s name in the Catena Sinaitica edition, with an indication
of what passages find parallels in the Armenian translation (appendix II). The
numbers are: l4 out of 26 passages for Genesis, and 2 out of 10 passages for
Exodus. This register, of course, raises the question whether the Armenian text
is a translation of a complete Eusebius text. If it is, the surplus Greek passages
must either be inauthentic or derived from other works by Eusebius.
In the last list (appendix III) are recorded those fragments which are not
published under Eusebius’ name in the Greek edition, but do find parallels in
the Armenian text. In a number of cases, Françoise Petit’s notes are quoted,
from which it appears that hints occur in the Greek material pointing towards
Eusebian authorship. On the whole, this list of course illustrates the possibili-
ties of using the Armenian translation to identify anonymous passages and
passages falsely attributed to other authors in the Greek tradition.
From what has been adduced so far, I hope the usefulness of the Arme-
nian translation for the important task of determining what is in fact left of
the Greek original of Eusebius’ commentary on historical writings of the Old
Testament has become sufficiently evident. It may have appeared as well that
much of the work still remains to be done.
The next question to be considered is what material might be found in
et des Rois, Studi e Testi 201, Città del Vaticano 1959, 55-103. This, of course, is not the place for a
full discussion of modern research on OT catenae including important works by such authors as E.
Mühlenberg and G. Dorival and others.
Armenian catenae. As far as I know, research into such collections of texts and
editorial work lag behind the study of related collections such as dogmatic
florilegia, collections of canonical rules and writings, etc. Of course, a few
things have been done. As was mentioned by way of introduction, Zanolli,
for instance, was primarily concerned with the Leviticus catena to be found
in the same manuscript as the Eusebian commentary.
In the library of San Lazzaro this catena is contained in three manuscripts;
apart from No. 873: No. 352, a manuscript of the second half of the twelfth
century, and No. 740, dated 1835, in which can be found both the catena as
such (even if in a mutilated shape) and, separately, a collection of fragments of
the catena. This late manuscript also contains a catena on the Gospel of Luke
and parts of a Genesis catena.
The three manuscripts are mutually independent, it seems,15 even if there
are a number of corresponding features. The translation represented by the
three witnesses must for linguistic reasons be dated rather late; Zanolli, from
a piece of information in MS 873, assumes the translation to have been made
in Constantinople in 716 A.D. The name of the translator is given as David,
counsellor and butler (?) of the royal table, son of the priest Elia; the scribe is
Step῾anos, priest and doctor of the province of Siunik.16 As to the Vorlage of
this catena, Zanolli assumes a very close connection to Procopius’ catena.
Of course, a thorough-going investigation into the Armenian translation
and transmission of Greek catenae, and the question whether the Procopius
branch is the only one to be transmitted or the predominant one should be
pursued on a much broader scale than Zanolli’s, comprising only three manu-
scripts and a few probings in supplementary material in the library of San
Lazzaro; and due regard should be given to the new achievements hinted at
above, as far as research into the Greek catenae is concerned.
Let me just add one further observation concerning forthcoming catena
research on the Armenian tradition so far neglected: this field of research can
hardly be isolated from the field mentioned above, that of the use of exegetic
literature of the Old Church in medieval Armenian literature, here exemplified
through Eusebius of Emesa.
To illustrate this point let me refer to Vardan Arewelc῾i, the famous his-
torian, traveller and exegete of the thirteenth century, whose commentaries
on writings of the Old Testament are characterized by Vahan Inglisian as
15 Zanolli, Catena sul Levitico, 128.

16 Ibid., 2, 97.
“florilegia”,17 used by B. Outtier under the designation “chaîne scripturaire”,18

and compared by A. Zanolli with the Leviticus catena, esp. quotations from
Ephraem.19 A commentary such as that compiled by Vardan raises the ques-
tion whether he uses earlier authors directly, or indirectly through catenae,
and in the latter case, whether these have been taken over directly from Greek
Vorlagen or have been elaborated on Armenian ground.
What has been adduced so far can to some extent be said to belong to the
technicalities of the process of drawing as full a picture as possible of the trans-
mission of the text, of establishing the best text possible, and of finding such
portions of the original behind the version or versions, as can be identified. It
will have appeared that a number of questions still wait for their answer.
These answers are important and must necessarily have a prominent place
on the agenda. But of course, the text historian must admit in modesty that
considerations of this kind can only rank as prolegomena to an analysis of the
contents of the text.
What, then, does the text contain? I cannot, of course, describe that in any
detail, but I hope, through a few illustrations, to be able to demonstrate that in
the title of this paper, I was right in applying the adjective “important” to this
text, primarily preserved through the efforts of medieval Armenian scribes.
Above all, an analysis of the text will give us an improved understanding of
Eusebius’ profile as an exegete. Church historians usually divide exegetes of the
fourth and fifth centuries into “schools”, primarily the allegorists of Alexandria
and their opponents and critics in Antioch, to which may be added the Syriac-
speaking schools of Edessa and Nisibis, and, of course, Latin exegetes, more
often than not dependent on the school of Alexandria. From contemporary
sources it is known that one of the central figures of the school of Antioch,
Diodorus of Tarsus, was considered to be a pupil of Eusebius of Emesa.20 This
17 Vahan Inglisian: Die armenische Literatur, Handbuch der Orientalistik, hrsg, v. B. Spuler, 1. Abt., 7. Bd.,
156-254, Leiden, Köln 1963. Cf. esp. p. 200: Seine (Vardan’s) Kommentare zur Genesis, Jesua, zu den
Büchern der Richter, der Vier Könige, zu Psalmen und Hohenlied kann man als Florilegium be
zeichnen.
18 B. Outtier, “La version arménienne du commentaire des Psaumes de Théodoret. Premier bilan”,
REArm 12, 1977, 169-180. Cf. esp. 174: “Les chaînes scripturaires ont utilisé notre texte, en premier
lieu Vardan l’Oriental”.
19 Zanolli, Catena sul Levitico, 102.
20 Hieronymus, De virus illustribus, ed. Herding, 62; cf. Buytaert, L’héritage littéraire, 9. For Diodore’s
contribution to the exegesis of the historical writings of the Old Testament, cf. esp. J. Deconinck:
Essai sur la chaîne de l’Octateuque avec une édition des commentaires de Diodore de Tarse qui s’y trouvent
contenus, Paris 1912. A fuller picture of Diodore as an exegete is emerging with the edition of his
can now be substantiated more broadly than before, and we can, therefore, say
that we are better informed of the early phase of the school of Antioch, which
has in the past to some extent been clouded in mists of darkness.
The roads of tradition and spiritual and scholarly interdependence were not
so narrow, however, as to exclude influence and inspiration from one school
onto another. Before I turn to a couple of considerations of that topic, it may
be natural to point out in the first place, that Eusebius’ approach to the Scrip-
tures is very linguistic. Thus, in the commentary he is very much aware of the
problems of translation. He knows that the Hebrew language has a number
of characteristics, which make it difficult to render the meaning of a passage
or a word into Greek in a very literal and verbatim translation, and he often
discusses the translational choices of the various Greek versions of the Old
Testament, and the Syriac renderings as well. In fact, in one of the examples
in the list of anonymous quotations from the Greek catenae21 the “author” is
cited as “the Syrian”. This is an identification which can already be found in
Eusebius’ text. “The Syrian” is, in fact, the name of a Bible translation, which is
usually assumed, despite its name, to be a Greek version; there are, however,
indications in Eusebius, which to my mind necessitate the re-opening of the
discussion, whether this designation does instead cover an early pre-Peshitta
Syriac version. In order to illustrate the importance of the Armenian testimony,
I have chosen an instance where the Armenian deviates at a crucial point from
a Greek fragment known already.
It is to be found in a comment on Ex. 4,25 (and 26), i.e. an element of the
dramatic tale of the Lord meeting Moses “by the way in the inn”, seeking to
kill him, which is avoided by the circumcision of Moses’ son through the hand
of Zipporah, his wife, who then says: “Surely a bridegroom of blood are you
to me” (v. 25), “a bridegroom of blood for the circumcision” (v. 26). Instead
of these phrases the Septuagint has a reading which can be translated: “The
blood of circumcision of my son ‘stood’ (maybe = ‘is staunched’)”.22
In a Greek catena fragment attributed to Eusebius of Emesa we are given
the information that instead of the Septuagint rendering Aquila reads: “I have
commentary on the Psalms: Diodorus Tarsensis, Commentarius in Psalmos 1, ed. J. Olivier, CCSG 6,
Turnhout 1980. For B. Outtier’s identification of Armenian evidence for this text cf. the article men-
tioned above (note 18).
21 See Appendix III, item G 233.
22 Cf. G. Vermès, “Baptism and Jewish Exegesis: New Light from Ancient Sources”, New Testament
Studies 4, 1957-58, 308-319, see esp. 310-311.
a bridegroom of blood”, and “the Hebrew”: “He (or she) sealed the blood of
circumcision”.23
This fragment is now found in the Armenian translation with one inter-
esting deviation from the Greek, insofar as it reads “the Syrian” instead of
Aquila.24 In fact, the reading attributed to Aquila in the Greek, and to “the
Syrian” in the Armenian, corresponds to the reading of the Peshitta; and it
might be worthwhile noting that the Peshitta reading has been challenged and
discussed, since the change of one single letter would make it conform to the
Hebrew text (‫ ܐܢܬ‬instead of ‫)ܐܝܬ‬.25
I have found no further comment on the reading attributed to “the He-
brew”, apart from Field’s note26 that the reading is due to a change between an
m and an n (‫חתמ‬, ‘to seal’ instead of ‫חתן‬, ‘bridegroom’). At any rate, it should by
now be apparent that the Eusebian readings cannot be classified as trivial.
I shall not venture to give any final verdict on how much new information
can be gained from our text about the history of the versions of the Old Testa-
ment, but I should like to add that besides references to Bible versions such
as Aquila and Theodotion, the Syrian and the Hebrew, we also come across
source references such as: “A certain Hebrew says”. This, of course, means
that Eusebius had some knowledge of Jewish exegesis of his own age or of
earlier periods. The lines of tradition connecting Jewish and Christian exegesis
have been known and studied before, not least for the school of Edessa, which
was, as will be remembered, Eusebius’ birthplace; and if we consider Philo,
the Christian school to be mentioned would, of course, be that of Alexandria.
However, certain traditions of rabbinical exegesis have not been as well-known
as Philo; nor has it been possible to judge their impact on Christian interpreta-
tion of the Old Testament. Research into these lines of connection is currently
in progress, so here again our text will be welcomed for its contributions.
23 Devreesse, Anciens commentateurs, 90-91.

24 MS Ven. Mekh. No. 873, 185, 30-33, ed. Hovhannessian, 108, 434-437.
25 Cf. the note in A. E. Brooke and N. McLean, The Old Testament in Greek, Cambridge 1917, 167. In
the new Peshitta edition (The Old Testament in Syriac according to the Peshitta Version 1, [The Peshitta
Institute, Leiden 1977] 125) no variants to the ‫‘( ܐܝܬ ܠܝ‬I have’, lit. ‘there is for me’) are quoted, and
it is interesting to note that Field, in his Hexapla-edition (Origenis Hexaplorum quae supersunt, ed.
F. Field, Oxford 1875, 87-88), on the basis of his assumption that Aquila besides his knowledge of
Hebrew was well acquainted with traditions behind the Syriac Bible versions (Prolegomena XXIV)
already refused to correct a Syro-Hexaplaric ‫ ܐܝܬ‬into ‫ܐܢܬ‬, because he saw the ἔχω, presumed to
belong to Aquila, as a support for the reading ‫ܐܝܬ‬.
26 Origenis Hexaplorum, 88.
After these few references to fields and topics where the contents of Eu-
sebius’ commentary are of special interest and to one of the directions into
which considerations of his sources will take us, let me conclude this paper
with a consideration taking us the other way; namely, to a use of Eusebius,
which has hitherto appeared enigmatic to researchers and led to a number of
wrong conclusions.
I am referring to a crux in Augustine27 that has for centuries puzzled schol-
ars. In a number of Augustine’s exegetic works he discusses what is the right
understanding of Gen. 1,2c. Does the clause refer to the Holy Ghost? Should
it be translated: “And the Spirit of God moved upon the face of the waters”
(with the RSV) or “a mighty wind swept over the surface of the waters” (to
quote the NEB)? In his De genesi ad litteram Augustine refers to a source for
his final considerations as being “a certain learned Christian Syrian”. Now, it
is easy to demonstrate that Augustine has taken this reference from Basil the
Great. But to which text and which author, then, does Basil refer? Through the
centuries a number of answers have been given to this question. To my mind,
there is no doubt that the right answer is: Eusebius of Emesa’s Commentary
on Genesis, and so, through Basil’s characterization of the Syrian author as
being orthodox28 it could be maintained that the Armenian scribes have helped
not only to solve the literary crux here sketched and answer the question of
Augustine’s and Basil’s ultimate source, but also to relieve Eusebius’ reputation
of being semi-heretical, as Basil’s testimony should be considered as weighty,
at least, as that of Jerome.
27 For a fuller discussion of Augustine’s interpretation of Gen. 1,2c, referred to in the following, with
references to sources and secondary literature cf. Henning J. Lehmann, “El Espfritu de Dios sobre
las aguas. Fuentes de los comentarios de Basilio y Agustín sobre el Génesis 1,2”, Augustinus 26
(1981) 127*-139*. [Cf. above, pp. 23ff].
28 Basil’s exact phrasing is: Ἐρῶ σοι οὐκ ἐμαυτοῦ λόγον, ἀλλὰ Σύρου ἀνδρὸς σοφίας κοσμικῆς
τοσοῦτον ἀφεστηκότος, ὄσον ἐγγύς ἦν τῆς τῶν ἀληθινῶν ἐπιστήμης. Cf. Basile de Césarée: Ho-
mélies sur l’Hexaëméron, ed. Stanislas Giet; Sources Chrétiennes 26, Paris 1949, 168.
Appendix I
Correspondences between Greek catena fragments in Françoise Petit’s edition
of the Catena Sinaitica and the Armenian translation of Eusebius of Emesa’s
Commentary on Historical Writings of the Old Testament.
Column 1: MS Ven. Mekh. No. 873, page, line.
Column 2: Quotations of Biblical texts.
Column 3: Identifications in the edition of the Catena Sinaitica (F. Petit’s num-
bering).
Column 4: Further identifications in Procopius (PG 87, column, paragraph,
line).
Column 5: Non-identified elements.
Column 6: Remarks.
1 2 3 4 5 6
162,29-31 Gen 12,17
162,31-35 G2
162,35-38 G 1,1-3
162,38 – 163,4 329,D2-7
163,4-7 329,D10-13
163,7 G 1,4
163,7-8 329,D14 – 332,A1
163,8 G 1,4-5
163,8-11 G 9,2-4
163,11-14 x Conc. Gen 14,18.20
163,14-16 Gen 15,2
163,16-17 G 16:2-3
163,17-19 Gen 15,8-9
163,19-31 G 22
163,31 – 164,7 G 20,2-12 Arm. a little amplified
164,7-9 Gen 15,15f
164,9-11 G 37
164,11-18 341,D4-10 Cf. G 37, note b
164,18-22 G 31a
164,23-25 344,B1-4 Cf. G 31, note *.
164,25-31 G 31b Arm. a little amplified

1 2 3 4 5 6
164,31-32 Gen 17,5
164,32-34 x Conc. Gen 17,5
164,34-37 Gen 17,17-19
164,37 – 165,8 Parallels with G 83
165,8-10 Gen 17,14
165,10-12 G 64,1-3 Arm. a little amplified
165,12-14 Gen 18,19
165,15-20 x Conc. Gen 18,19
165,21 Gen 18,21
165,22-26 x Conc. Gen 18,21 (not G87)
165,26-27 Gen 18,27
165,27-29 x Conc. Gen 18,27 (not G 93)
165,29-38 369,D1-371/372,A6
165,38 – 166,8 G 116 Arm. a little amplified
166,8-9 Gen 20,2
166,9-15 G 129
166,15-28 x Conc. Gen 20,3ff (cit. vss. 6.16)
166,28-31 G 139
166,31 – 167,2 x
167,2-6 Gen 20,17f
167,6-12 x
167,12-14 Gen 21,14
167,14-23 G 151
167,23-24 G 150,1-2
167,24-26 x (Cf. G 147,1)
167,26-29 G 147,3-6
167,30-31 x Cf. PG 87,384,21
167,31-32 G 150,2-4
167,32-35 x Cf. PG 87,386,8f
167,36-37 x Cf. PG 87,388,A9-12
167,38 Gen 21,22
167,38 – 168,2 x
168,2-8 388,C16-D7
1 2 3 4 5 6
168,9-12 x
168,13-16 G 177
168,16-21 G 189 = Proc. 1.1-6
168,22 x Conc. Gen 22,1
168,22-24 Gen 22,12
168,24-29 G 183
168,29-32 x (Cf. G 185–not identical!)
168,32-34 Gen 23,4
168,34 – 169,3 x Conc. Gen 23,4 6 (cf. 194)
169,3-5 Gen 23,15
169,5-9 x Conc. Gen 23,15
169,9-10 Gen 24,2
169,11-13 G 209
169,13-14 x
169,14-16 x
169,16-21 G 211
169,21-32 395/396,A27-34
169,32-33 Gen 24,5
169,33-35 x Conc. Gen 24,5
169,35-37 403/404,C9-10
169,37-38 Gen 24,49
169,38 x Conc. Gen 24,49
170,1-2 Gen 24,50
170,2-3 G 233
170,3 Gen 24,63
170,3-6 401/402,B12-17
170,6-8 x (Cf. G 16)
170,8 Gen 25,22
170,8-9 407/408,A14-16 (Devreesse, p. 75)
170,9-10 Gen 25,26
170,10-16 407/408,C10-15 (Devreesse, p. 75)
170,16-17 Gen 25,27
170,17-20 409/410,Al-5
1 2 3 4 5 6
170,20-21 Gen 25,28
170,21-28 G 261 = Proc.
170,28-33 409/410,A21-28(31) . (Cf. G 262, note a)
170,33-34 Gen 25,31
170,34 – 171,4 411/412,A3-13
171,4-6 G 266 Arm. a little amplified
171,7 Gen 26,31
180,21 Ex 1,12
180,22-24 E 11
180,24-26 Ex l,20f
180,26-29 E 16 = Proc.
180,30-38 513/514,B17-21
180,38 – 181,2 Ex l,20f
181,2 x Conc. Ex l,20f
181,2-4 x Conc. Ex 1,22
181,5-25 x Conc. Ex 2,1ff
181,25 Ex 2,14
181,26-28 E 34 = Proc.
181,28-31 x Conc. Ex 2,14
181,31 Ex 2,24f
Appendix II
Passages published in the Catena Sinaitica under Eusebius of Emesa’s name.
Parallels in the Armenian translation marked with *. Numbering of Greek
fragments according to Françoise Petit’s edition.
Genesis Exodus
Fragment No. Fragment No.
2 * 11 *
9 * 15
20 * 16 *
22 * 21
64 * 22
87 23
91 33
92 36
93 43
99 44
100
116 *
118
119
139 *
150 *
151 *
153
177 *
183 *
185
189 *
192
209 *
210
211 *
Total 26 14 10 2
Appendix III
Passages published in the Catena Sinaitica edition under other names than Eu-
sebius of Emesa, for which parallels are found in the Armenian translation.
Greek Fragment No. Attributions in Greek Catenae

G1 The same; Anonymous; Eusebius (Bs) (1)
G 16 Anonymous; Eusebius (Len); Diodorus (2)
G 31 Origen; Anonymous (3)
G 37 Anonymous; Philo
G 129 Anonymous; Didymus
G 147 Anonymous; Eusebius (Len); Diodorus (4)
G 233 The Syrian
G 261 The same; Eusebius (Len); Anonymous
G 266 Anonymous
E 34 Anonymous
Notes (quotations of relevant passages from Françoise Petit’s notes).
1. p. 3f, note (a) Ce morceau n’est intelligible que si l’on tient compte du précédent … Les
deux sont distincts dans Sin, Mosq et Procope; ils sont liés dans Len et Bs
(avec, dans ce dernier, attributions à Eusèbe d’Emèse) …
2. p. 19, note * … La rédaction des chaînes du premier groupe pourrait bien revenir à
Eusèbe d’Emèse …
3. p. 36, note (a) Dans Len, soudé à G 30, attribué à Eusèbe d’Emèse
4. p. 141, note * … Comme pour G 16, nous pensons que la rédaction de Sin Len Mosq¹ Bs
revient à Eusèbe d’Emèse …
Severian of Gabala
Fragments of the Aucher Collection in Galata MS 54
Johannes Baptista Aucher (or Awgerean) did an excellent job when in 1827
he edited fifteen homilies attributed to Severian of Gabala, translated into
Armenian at a very early date1.
Already Aucher was aware that the testimony of tradition, as far as the
question of authorship was concerned, carried different weight for the first ten
homilies as compared with homilies XI-XV. The reason for this is that homilies
I-X were found as one collection of texts in a number of the manuscripts, on
which Aucher based his edition, whereas he found the last five scattered in
different homiliaries and other collections of texts.
Before I discuss homilies I-X and their attestation in Galata MS 54, one point
should be made concerning homilies XI-XV, viz. that one of these homilies (or
excerpts of homilies, as they should rightly be termed) has in fact been shown
to belong to Severian, two should be ascribed to Eusebius of Emesa, whereas
for the remaining two the question of authorship is undecided.
Homily XI is the one of Severianic authorship, being part of homily XIII in
the collection of homilies edited by N. Akinian under the name of Eusebius
of Emesa2. Concerning this collection the present author has shown that only
homilies I-VIII belong to Eusebius, IX-XIII to Severian3.
Aucher’s homilies XIV and XV are parts of homily V in Akinian’s collec-
1 Severiani sive Seberiani Gabalorum episcopi Emesensis homiliae nunc primum editae ex antiqua versione
armena in latinum sermonem translatae per P. Jo. Baptistam Aucher, Venice (S. Lazzaro) 1827.
2 N. Akinian, “Die Reden des Bischofs Eusebius von Emesa”, Handes Amsorya 70-73, 1956-1959. The
thirteenth and last homily is to be found in HA, 73, 1959, cols. 321-359.
3 Henning J. Lehmann, “The Attribution of certain Pseudo-Chrysostomica to Severian of Gabala
confirmed by the Armenian Tradition”, Studia Patristica, 10, Texte und Untersuchungen 107, 1970, pp.
121-130, and idem, Per Piscatores. Studies in the Armenian Version of a Collection of Homilies by Euse-
bius of Emesa and Severian of Gabala, Aarhus 1975. (For homily Akinian XIII/Aucher XI, see esp. pp.
335ff.).
tion, which is by Eusebius of Emesa. The correct attribution was suggested

already in 1921 by A. Vardanian and has been confirmed by such authors as
J. Lebon, N. Akinian, E.-M. Buytaert, and the present writer4.
For homilies XII and XIII no decisive answer has been given to the question
of authorship.
In the following the term “Aucher Collection” will refer to the series of
ten homilies (I-X), which were united in Aucher’s basic MS material. Leaving
aside homilies XI-XV and turning to the Aucher Collection “proper” it must
be emphasized that Aucher himself was already aware of the fact that homily
X is known in Greek under the name of St. Basil the Great (PG 31, 423-444)5.
Thus it cannot be assumed a priori that all of the material in the collection of ten
homilies stems from Severian’s hand. As for homily I, doubts about Severian’s
author’s rights were voiced by J. Zellinger6, and with more emphasis by E.-M.
Buytaert7; H.-D. Altendorf, finally, pointed out that the text could hardly be
by Severian8. It was not until Akinian’s edition was published, however, that
it could be shown that Aucher’s homily I is in fact part of Akinian’s homily
III9, which is undoubtedly by Eusebius10.
It seems, then, that only homilies II-IX remain genuine Severian texts in
the “Aucher Collection” proper.
Turning from the question of authenticity to the question of transmission,
the primary point to be made seems to me to be the fact that the collection
of ten homilies must date from a rather early age. At the end of this article I
shall return – rather tentatively – to this question and argue for a dating of the
ten-homily-collection to the Armenian phase of transmission. That is to say,
that one is not, in my view, likely to find a Greek Vorlage for the collection as
a whole, which of course does not exclude further identifications of the Greek
4 A. Vardanian, “Եւսեբեայ Եմեսացւոյ դասական մնացորդները, Eusebius von Emessa. Überreste in

altarmenischer Übersetzung”, HA 35, 1921, cols. 129-146, 292-297, esp. col. 132ff; J. Lebon, “Les cita-
tions patristiques grecques du ‘Sceau de la Foi’” Revue d’Histoire Ecclésiastique 25, 1929, pp. 5-32, esp.
p. 17; E.-M. Buytaert, L’héritage littéraire d’Eusèbe d’Emèse, Bibliothèque du Muséon 24, Louvain 1949;
N. Akinian, op. cit., esp. HA 71, 1957, col. 353f; cf. Lehmann, Per Piscatores, pp. 209ff.
5 Cf. Aucher, ed. cit., pp. 370f; J. Zellinger, Studien zu Severian von Gabala, Münsterische Beiträge zur
Theologie 8, 1926, p. 71.
6 Op. cit., pp. 72 and 74ff. Zellinger finally advocates Severianic authorship.
7 Op. cit., pp. 138ff.
8 H.-D. Altendorf, Untersuchungen zu Severian von Gabala (unpublished dissertation), Tübingen 1957;
see esp. pp. 41ff.
9 Ed. cit., HA 71, 1957, cols. 99-130.
10 On this identification see Lehmann, Per Piscatores, pp. 103ff.
S e v e r i a n o f G a b a l a – A u c h e r C o ll e c t i o n 97
originals of individual homilies. So far, apart from homily X, only two of the
homilies have been identified in their entirety in Greek11.
In any case the collection of ten homilies has a very broad attestation in
Armenian manuscripts. Therefore, of course, it would be desirable to have a
scholarly edition of this collection to replace Aucher’s which gives very little
information on variant readings. When it was said at the beginning of this
article that Aucher’s edition represented “an excellent job”, my point was that,
as far as my investigations of the MS material go, they show that by and large
the text in a scholarly edition would look very much like Aucher’s text.
As far as I have been able to ascertain, Galata MS 54 (which must presum-
ably be dated to the fourteenth century)12 is unique in the way in which it
brings supplementary attestation to the attribution of certain texts to Severian
of Gabala13. Quantitatively the material is not of great extent, as it comprises
only about twenty pages in the MS, but as I hope to show in the following,
it seems to me that there are important observations to be made from this
manuscript.
On the pages in question are found two series of fragments or ex-
cerpts of homilies, the first one (pp. 363-375) under the heading Երանելւոյ
Սեբերիանոսի Եմեսու եպիսկոպոսի, the second one (pp. 375-384) under the
heading Երանելւոյ Սեբերիանոսի Գաբա… եպիսկոպոսի. Even if the name
of locality for the author of the second group is not quite readable, there can
be no doubt that Gabala is meant. For the first series we have the combina-
tion of the personal name: Seberianos and the name of locality: Emesa, which
11 The two homilies identified in Greek are No. VII (the Greek original of which was known already
by Aucher) being identical with PG 56, 553-564 (cf. esp. Zellinger, op. cit., pp. 42ff), and No. IX, the
Greek original of which has been edited, partly by Zellinger (op. cit., pp. 9-21), and partly by Ch.
Martin, “Note sur l’homélie de Sévérien de Gabala in illud: Pater, transeat a me calix iste (Mt. 26, 39)”,
Le Muséon 48, 1935, pp. 313-320. A fairly large number of fragments of homily VIII have been found
in Greek catenae (cf. Zellinger, op. cit. pp. 96ff, and J. Zellinger, Die Genesishomilien des Bischofs Seve
rian von Gabala, Alttestamentliche Abhandlungen, VII, 1, Münster 1916, pp. 13ff).
12 For a more detailed description I refer to Babgēn [Kiwlēsērean], Ցուցակ ձեռագրաց Ղալաթիոյ
ազգային մատենադարանի հայոց, Catalogue of the Manuscripts of the Armenian National Library of
Galata, Antelias 1961; and to Charles Renoux’s preface to his edition of Irenaeus fragments from
Galata MS 54, Irénée de Lyon: Nouveaux fragments arméniens de l’Adversus Haereses et de l’Epideixis,
Patrologia Orientalis 39, 1, Turnhout 1978, esp. pp. 13ff.
13 My thanks are due to the Armenian Patriarchate of Istanbul, and especially to Mr. K. Pamboukdjian
for his courtesy and kindness during my visit in 1978. I am particularly grateful for having been
given the possibility to photograph the relevant pages of Galata MS 54, which is now in the archive
of the Patriarchate.
is often met with, and which may point to an early combination of texts by
Eusebius of Emesa and Severian of Gabala14.
Be that as it may, the fragments of the first series all belong to homilies of
the Aucher Collection, whereas the fragments of the second series derive from
six other homilies, some of which have so far not been identified as belonging
to Severian. For that reason it may be maintained that the second group of
fragments is the most important for future Severian research. A presentation
of the fragments of the second series in their relation to texts and fragments
known in Greek, Syriac, and Armenian would demand more space than is
available here. I have elsewhere given a report about the identifications this
group of fragments allows15.
First of all a list of concordances between the excerpts in Galata MS 54 and
Aucher’s Armenian edition and his Latin translation might be of use.
List of Excerpts in Galata Ms 54 from Severian of Gabala’s Homilies in the Aucher

Collection16
Galata MS 54, Aucher’s edition, Aucher's translation,

page17, line page, line page, line
Homily I
(a) 363a,21-28 4,1-5 5,1-5
զի եթէ իմասցուք … եթող

Quoniam si intelligamus … reliquit.
(b) 363a,28-b,22 14,26 – 16,1 15,30 – 17,2
զի միածին որդին … արարածք

Quoniam Unigenitus Filius … creatura.
14 Cf. Lehmann, Per Piscatores, pp. 147ff.

15 Cf. Henning J. Lehmann, “Severian of Gabala: New Identifications of Texts in Armenian Transla-
tion”, in Thomas J. Samuelian (ed.), Classical Armenian Culture, University of Pennsylvania Armenian
Texts and Studies 4, 1982, pp. 113-124. [Cf. above, pp. 67ff]
16 Incipits and desinits are given according to the manuscript, variant readings in Aucher’s edition in
the notes below.
17 At the beginning of new groups of texts the manuscript has two columns (here indicated by a and
b), otherwise only one.
Galata MS 54, Aucher’s edition, Aucher’s translation,

Homily II
(c) 363b,25 – 364,15 18,24 – 20,12 19,22 – 21,11
կոչի ճանապարհ … պատմելով

18
Vocatur itaque via … referens.
(d) 364,15-32 24,7-27 25,3-24
քանզի փրկիչն19 եւ տէր … բնութեանս մերում

Quoniam Salvator Dominusque … naturae nostrae.
(e) 364,32 – 366,2 34,7 – 36,10 35,6 – 37,10
այն իմն կարի տրտմեցուցանէր … ի պատիւն

Illud valde tristes reddebat … ad dignitatem.
(f) 366,3-29 36,37 – 38,31 37,34 – 39,30
գիտէք ասէ եղբայրք … տնտեսուտեամբ

Scitis, iniquit, fratres … dispensatione plenam?
Homily III
(g) 366,32 – 367,6 78,5-14 79,4-12
եւ յափշտակեաց զյափշտակեալսն … խոչեսչէ

20
et rapit direpta … vulneraret.
Homily IV
(h) 367,8-20 160,16-33 161,14-32
ես եմ ասէ տէր աստուած … զաւրութիւն է

Ego sum, ait, Dominus Deus … habent virtutem.
(i) 367,21 – 368, 8 162,32 – 164, 21 163,34 – 165,21
տեսցուք թէ եւ այլ ինչ անուն … աւրհնեալ յաւիտեանս

Videamus, an etiam aliud nomen … benedictus in aeternum.
18 Aucher: add. եւ.

19 Aucher: om. -ն.
20 Aucher: յապշտակէ.

Homily V
(j) 368,10 – 369,2 202,15 – 204,8 203,14 – 205,13
ասեն հերձուածողացն … ըստ մարմնոյս

Dicunt haereticorum … secundum carnem.
Homily VI
(k) 369,3-18 216,32 – 218,16 217,28 – 219,16
եկն ասէ21 Յիսուս … աշխատութիւն

Venit (ait) Jesus … laborem pateretur?
(l) 369,18-25 228,22-30 229,21-29
յորժամ ասիցէ հերիտիկուն … կրաւնիւքն

22 23
Quando dicit haereticus … dogma jam delet.
(m) 369,25 – 370,16 236,21 – 238,12 237,18 – 239,12
զնոյն միտս ասէ հարցին … համարհողաց

24 25
Eodem sensu interrogarunt discipuli … contemnentium.
Homily VII
(n) 370,17-24 258,13-21 259,14-22
յորոյ վերայ տեսանիցես զհոգին … տնտեսութեանն

Super quem videbis Spiritum … dispensationem.
21 Aucher: om. ասէ.

22 Aucher: ասէ.
23 Aucher: հերետիկոսն.
24 Aucher: om. ասէ.
25 Aucher: add. -ն.

Homily IX
(o) 370,26 – 372,9 336,14 – 338,33 337,14 – 339,29
աւն առ յաստուածեղէն իսկ … ի խաչէն

26
Agedum ad divina … crucem debebat?
(p) 372,9-32 340,6-32 341,1-26
յորժամ տէրն ի խաչելութիւն … խռովի

Cum proximum esset tempus … conturbatur.
(q) 372,33 – 373,28 342,26 – 344,20 343,25 – 345,16
այլ զի ցուցից … մնայ ճշմարտութիւնն

Sed ut ostenderem … manet veritas.
(r) 373,29 – 374,11 350,7-24 351,4-21
պարտ է կրկնարաւր … ի վեր եւ բարձրացեալ27

Oportet secundo … superans.
(s) 374,11 – 375,5 356,3 – 358,3 357,1 – 359,4
եւ արդ զինչ իցէ ասելն … դատապարտեալ էր

Cur ergo dicitur … condemnatum redimeret.
(t) 375,6-13 362,22-32 363,20-30
եւ զինչ կամի ցուցանել … զմիտս ճշմարտութեանն

et quid velit manifestare … sententiam veritatis confirmem.
To this presentation of Galata MS 54 as a witness to the Aucher Collection,

I shall only add here a few remarks, mainly on variant readings, on the dis-
tribution of quotations within the individual homilies, and on the possibility
of using the information thus gained in a comparison with other excerpts
of the collection with a view to elucidating the history of the formation and
transmission of the collection.
This is not the place to draw up an exhaustive list of variant readings. The
few examples given in notes 18-27 are quite illustrative for the character of
textual variants, which are mainly of the types shown there: omission of “ar-
ticle”, minor changes of verbal forms, omission of the ասէ used as a quotation
formula, and so forth.
As regards the distribution of the fragments quoted, the following three
observations seem to me the most important: 1) two thirds of the material de-
26 Aucher: om. իսկ.

27 Aucher: բարձր.
rive from homilies II and IX, 2) homilies VIII and X are not quoted, 3) homily
I is quoted.
The list of fragments in itself may give reason enough for placing particular
emphasis on these facts, but what gives them further importance, is that the
use of Severianic texts in the seventh and eighth centuries on Armenian soil
furnishes us with comparative material which may give us at least a basis for
a guess about the time and environment in which the collector or excerptor
of the Severian fragments of Galata MS 54 might be located. First of all I have
in mind the famous patristic florilege, the Seal of Faith – collected presumably
under catholicos Komitas (612-628)28; in addition quotations in the Contra
Phantasticos by John of Ōjun, who was catholicos about 100 years later (718-729)
and was given the surname “the Philosopher”, are of interest.
Before commenting upon parallels and discrepancies between these au-
thorities a list of the quotations in question should be given29:
Quotations from Homilies I-IX of the Aucher Collection in Galata Ms 54, Seal of Faith30
and John of Ōjun31
Galata MS 54 Seal of Faith John of Ōjun

Homily I
(a) 4,1-5
(b) 14,26 – 16,1
Homily II
(c) 18,24 – 20,12
(d) 24,7-27 24, 12-27
26,13-23
28,2 – 32, 22
32,36 – 34,34
(e) 34,7 – 36,10
(f) 36,37 – 38,31
28 Cf. Lebon, op. cit., p. 5.

29 Pages and lines are given according to Aucher’s edition.
30 Quoted from Lebon, pp. 28f.
31 Quoted from Aucher; pp. XVIII, 1, 24/25, and 340/341; and Zellinger, pp. 22 and 79.
Galata MS 54 Seal of Faith John of Ōjun

Homily III
(g) 78,5-14
Homily IV
158,10-15
(h) 160,16-33
(i) 162,32 – 164,21
Homily V
(j) 202,15 – 204,8 202,19 – 204,8
Homily VI
(k) 216,32 – 218,16 216,18 – 218,7
(l) 228,22-30
(m) 236,21 – 238,12 236,21 – 238,12
Homily VII
(n) 258,13-21
Homily IX
320,4-9
(o) 336,14 – 338,33 334,36 – 342,11
(p) 340,6-32 340,20-25
(q) 342,26 – 344,20 342,24-36
344,21 – 346,11
346,16 – 348,22
(r) 350,7-24 348,34 – 350,24
350,25 – 352,14
352,23 – 354,17
354,30 – 356,2
(s) 356,3 – 358,3
358,6-24
(t) 362,22-32
The above list shows that there are many correspondences, but no direct and
exclusive interdependence between the three authorities or any two of them,
as any of the three sources contains material that is not quoted by the other
two.
As was mentioned above, the three sources are related through the great
importance all of them attach to homilies II and IX. Hence, it is obvious that
these two texts were very well known under the name of Severian and re-
garded as important authorities in the dogmatic discussions of the Armenian
church during the seventh and eighth centuries.
The omission of homilies VIII and X in both Galata MS 54 and the Seal
of Faith may of course be accidental. However, the hypothesis immediately
suggests itself that both sources belong to a stage when homily X, which is in
fact – as mentioned above – by St. Basil the Great, was not yet connected with
the collection of Severian texts.
As regards homily VIII, it seems to me that the information given by Zel-
linger on its transmission, especially on the quotations in Greek catenae32,
might suggest that this homily belonged originally to another group of Severi-
anic homiles treating of themes concerned with the Old Testament.
The absence of quotations from homily I in the Seal of Faith, and the num-
bering in this florilegium of homilies II, IV, and VI as 1, 3, and 5 respectively33,
might suggest that the Seal of Faith is the older of the two witnesses, reflecting
a stage when the Eusebian homily had not yet been added to the collection as
its text No. I.
Also, it seems to me that the theological content of the exposé on the Human
and Divine in Christ, which might be an apposite term for the collection made
up by the excerptor of Galata MS 54, would fit better into the spiritual environ-
ment of John of Ōjun than to the Julianist theology of John Mayragomec῾i, who
may be the inspiring – or even the editorial – force behind the Seal of Faith34.
As the year 929 or 930, which is the date of the oldest manuscript referred
to by Aucher in his edition35, is the terminus ante quem for the formation of the
collection of ten homilies, the following outline of the editorial history of this
collection might be tentatively suggested:
1. An existing collection of homilies II-VII and IX is translated into Arme-
nian before 600.
32 Cf. above, note 11.

33 Cf. above, esp. the reference to Altendorf’s arguments for homily 1 in the Aucher Collection not
being by Severian (cf. note 8).
34 Cf. Lebon, p. 6.
35 Aucher, pp. 400f.
2. An excerpt is made of the Eusebian homily “on Faith” (=Akinian’s hom-

ily III)36 , and this excerpt is given the place of introductory text to the collec-
tion some time about 70037. About this text one detail should be mentioned:
fragment (b) ends immediately before the final doxology, which cannot – for
theological and historical reasons – be Eusebian38. Thus, Galata MS 54 has not
furnished us with further material on which we could base a positive assump-
tion as to the exact date of the addition of the doxology, but in so far as frag-
ment (b) is introduced with the formula “at the end of the discourse” (ի վերջս
ճառին), it seems to me to have added a literary argument to the theological
and historical reasons mentioned above – to the effect that the doxology did
not originally belong to the homily and was not yet added at the time of the
excerption, which is now witnessed by Galata MS 54.
3. At a certain time in the eighth or ninth century the Severianic homily
VIII and the Basilian homily X have been added. The last homily, which con-
centrates on baptism, might indicate that the reason for this extension of the
collection might be that it should serve as a teaching programme within the
baptismal catechesis.
This outline of a possible history of formation for the first series of excerpts
from Severianic homilies in Galata MS 54 can, of course, only be tentative and
hypothetical. It suggests, however, the importance of trying to evaluate the
Armenian collections of patristic material in terms of the background of their
origin and history in the Armenian environment, not only to elucidate the
history of the doctrine of the Armenian church, but also as a necessary part
of the procedure through which insight is gained on literary questions such
as authorship and the authenticity and integrity of texts.
36 Cf. above, notes 6-10.

37 For the character of this text as an “expositio fidei” as described by Zellinger and Altendorf, and for
Jugie’s juxtapposition of the homily and the Athanasian Creed cf. Lehmann, Per Piscatores, pp. 109ff.
38 Cf. Lehmann, Per Piscatores, pp. 65ff and 370.
he Syriac Translation of the Old Testament
– as Evidenced around the Middle of the
Fourth Century (in Eusebius of Emesa)
T
1. Quis sit ὁ Σύρος? – Montfaucon, Field, and Rahlfs
When – about 70 years ago – Alfred Rahlfs published an article under the title
“Quis sit ὁ Σύρος?”1 this title was borrowed from Frederick Field, who included
a chapter with the same title in the introduction to his edition of the Hexapla.2
It was not only in the choice of title, however, that Rahlfs built upon Field.
Also as regards the principal point of view that is maintained in the article,
viz. that references to ὁ Σύρος which occur in Patristic literature, not least in
catenae, refer to a Greek version of the Old Testament, Rahlfs could refer to
Field – and more than a hundred years further back, viz. to Montfaucon, who
already in his edition of the Hexapla had voiced the same idea.3
There are differences of opinion between the three authors, e.g. in their
evaluation of relations between “the Syrian” and the Peshitta and their ex-
planations of how and why “the Syrian” got its name. I shall not go into any
detailed discussion of such points; only, as a summing up of the main results
and points of view in earlier research I shall quote Rahlfs’ 4-point summary:
1. Der Σύρος ist nicht die Peschita …

2. Der Σύρος hat überhaupt nicht Syrisch, sondern, wie Montfaucon und
Field mit Recht annahmen, Griechisch geschrieben …
3. (dealing particularly with Judg. 12,6 in Theodoret – cf. below) … Also
dürfen wir es auf jeden Fall als sehr wahrscheinlich bezeichnen, daß der
1 Alfred Rahlfs, “Quis sit ὁ Σύρος?”, Kleine Mitteilungen II, Nachrichten von der Gesellschaft der Wissen-
schaften zu Göttingen, Phil.-hist. Klasse, 1915, 420-428.
2 Origenis Hexaplorum quae supersunt; sive veterum interpretum graecorum in totum Vetus Testamentum
fragmenta…, ed. Fridericus Field, tom. I, Oxonii 1875, LXXVII-LXXXII.
3 Hexaplorum Origenis quae supersunt, auctiora et emendatiora quam a Flaminio Nobilio, Ioanne
Drusio et tandem a Bernardo de Montfaucon … ed. … Carolus Fridericus Bahrdt, Pars I, Lipsiae et
Lubecae 1769, 31-33.
Σύρος an unserer Stelle nicht aus der Peschita, sondern aus dem hebräis-
chen Urtexte übersetzt hat.
4. Der Σύρος war in der Tat von Herkunft ein Syrer.4
The main purpose of this paper is to present evidence which in the view of the
present writer makes Rahlfs’ argument under item 2 untenable. The textual
basis for taking up this discussion and for arguing that “the Syrian” refers to
a Syriac translation, is Eusebius of Emesa’s commentary on historical writings
of the Old Testament. Therefore, in our section 2, some general information
about this text will be given. Its history of transmission will first be treated
of briefly, in connection with some information about the author, his other
works, and the possibilities of retracing further parts of the commentary in
its original language, Greek (section 2.1). For the general understanding and
evaluation of Eusebius’ commentary it is important to look at the way, in which
it deals with questions of languages and translation, in particular Syriac and
Hebrew as compared with Greek. This is done in section 2.2, and in section
2.3 the references to “the Syrian” and other versions of the Old Testament will
be counted and briefly commented upon.
One of the main arguments referred to by Montfaucon, Field, and Rahlfs,
for “the Syrian” being a Greek version was taken from its readings in Gen.
39,2f as transmitted in catenae. Eusebius’ commentary sheds new light on these
particular readings and therefore on the whole question of the language of
“the Syrian”, not only from general considerations, but from a very clear-cut
exegesis of the textual details of these verses. This is shown in section 3.
Rahlfs, however, added to the arguments of his predecessors, what he
himself calls “einen vollständig sicheren Beweis”,5 viz. of “the Syrian” having
translated from the Hebrew into Greek. This “proof”, which formed the final
basis for Rahlfs’ summing up as quoted above, was built upon a reference to
the reading of “the Syrian” in Judg. 12,6 as referred to by Theodoret of Cyr
rhus. This particular reference, and Rahlfs’ use of it, are discussed in section
4.1, and some probings into its background, i.e. Theodoret’s use of Bible ver-
sions, are submitted as an excursus in section 4.2 in order to be able further
to substantiate and profile the conclusions as to the language of “the Syrian”
and Eusebius of Emesa’s evidence about this version of the Old Testament.
Conclusions of our findings and deliberations are given in section 5.
4 Rahlfs, “Quis sit”, 426.

T h e S y r i a c T r a n sl a t i o n o f t h e Old T e s t a m e n t 109
2. Eusebius of Emesa and ὁ Σύρος

2.1. Eusebius’ Commentary on Historical Writings of the Old Testament
It is a long and well established fact that Eusebius of Emesa is one of the
writers of the Old Church, whose writings bear testimony to readings of ὁ
Σύρος. Thus e.g., in Devreesse’s edition of catena fragments to be found in
the writings of Greek fathers and exegetes,6 in the chapter about Eusebius
there are 10 references to “the Syrian”.7 Further it can be noted that Devreesse
(referring to G. Mercati) argues that Eusebius would seem to be the earliest
exegete to use ὁ Σύρος.8
For this – and other – reasons it is a regrettable fact that the history of
transmission has not been lenient with Eusebius’ works in their original lan-
guage, Greek. Fortunately, some of his works have been preserved in trans-
lations, primarily into Latin and Armenian, some of them rather early.9 For
his commentary on historical writings of the Old Testament future research
will to a great extent be dependent on the Armenian tradition. The text was
translated into Armenian in the “classical” period, i.e. presumably as early as
the 5th century A.D. However, in Armenian tradition the commentary was
for some – yet obscure – reason attributed to Cyril of Alexandria, and it was
not until 1923 that it was demonstrated by the Armenian Mechitharist scholar
V. Hovhannessian, that the text rightly belongs to Eusebius. The final printed
6 Robert Devreesse, Les anciens commentateurs grecs de l’Octateuque et des Rois. (Fragments tirés des
chaînes). Studi e Testi 201, Città del Vaticano 1959.
7 Cf Devreesse, Commentateurs grecs, 55ff. The verses referred to are the following: Gen. 2,8f; 5,3; 11,3;
17,14; 19,21ff; 24,2; 26,35; 27,27; 31,7f; Exod. 1,12.
8 Devreesse, Commentateurs grecs, 57, n. 5. Cf Giovanni Mercati, “A quale tempo risale ‘il Siro’ dei
commentatori greci della Bibbia”, Bib 20, 1945, 1-11. In this article Mercati shows that a quotation of
“the Syrian” traditionally attributed to Melito of Sardes should rightly be transferred to Eusebius of
Emesa.
9 About Eusebius’ life and works, see esp. É.M. Buytaert, L’héritage littéraire d’Eusèbe d’Émèse. Biblio-
thèque du Muséon 24, Louvain 1949. Buytaert also published two series of homilies in Latin transla-
tion: É.M. Buytaert, ed., Eusèbe d’Émèse: Discours conservés en latin. I. La collection de Troyes; II. La
collection de Sirmond. Spicilegium Sacrum Lovaniense 26 & 27. Louvain 1953, 1957. For homilies in
Armenian, see esp. Henning J. Lehmann, Per Piscatores. Studies in the Armenian version of a collection
of homilies by Eusebius of Emesa and Severian of Gabala, Aarhus 1975. A brief introduction to Eusebius’
theology is given in: P. Smulders, Eusebius van Emesa, wegbereider van de Antiocheense Christologie.
Rede uitgesproken bij het neerleggen van het ambt … aan de Katholieke Theologische Hogeschool te Amster-
dam op vrijdag 20 mei 1983, edited by the KHTA, n.d. Cf also P. Smulders, “Eusèbe d’Émèse comme
source du De Trinitate d’Hilaire de Poitiers”, Hilaire et son temps, Actes du Colloque de Poitiers 29 sept.
– 3 oct. 1968, Editions Augustiniennes, 1969, 175-212.
edition of the Armenian translation of the text – with V. Hovhannessian’s name

posthumously on the title page – did not appear until the year 1980.10
By way of conclusion of this brief introduction to the important Eusebian
text, which has, because of its history of transmission and publication, so far
attracted little scholarly interest, four points should be singled out for special
mention:
i) The commentary belongs to the genre of quaestiones in loca difficilia. In

the commentary the writings from Genesis to 2 Kings (except Ruth) are
commented upon; as is often the case in commentaries belonging to this
genre, the comments on Genesis are very rich, whereas comments on the
rest of the Pentateuch and other writings are more scanty. However, it is
characteristic of Eusebius’ commentary that also the Books of Samuel and
Kings are commented upon rather extensively.
ii) Certain indications seem to reveal that the Armenian text does not render
the Greek original in its entirety.11 It is impossible to judge how great are
the lacunae; they may be incidental and small; at any rate, it is obvious that
the Armenian text contains a great amount of material that has hitherto
been unknown in Greek.
iii) The possibility of identifying further portions of the Greek original is pri-
marily bound up with a comparison of the Armenian text with Procop-
ius.12
iv) The importance of the text for the topic treated of in this paper should be ev-
10 Eusèbe d’Émèse, I: Commentaire de l’Octateuque. Préparé par V. Hovhannessian, Mekhitariste,

Venise-St. Lazare 1980. In the present paper this edition is referred to as “Ven.ed.” – followed by
indication of page and line (line counting according to the edition). For further information about
the Armenian text and references to V. Hovhannessian’s earlier works about it, see Lehmann, Per
Piscatores, 31-33, and Henning Lehmann, “An Important Text Preserved in MS Ven. Mekh. No. 873,
dated A.D. 1299 (Eusebius of Emesa’s Commentary on Historical Writings of the Old Testament)”,
Medieval Armenian Culture, ed. by Th.J. Samuelian and Michael E. Stone, University of Pennsylvania
Armenian Texts and Studies 6, Chico 1984, 142-160 [above, pp. 77ff]. In the Venice edition (as in the
relevant catalogues) there are descriptions in Armenian of the three manuscripts on which the edi-
tion is based. By far the most important manuscript is MS Ven. Mekh. No. 873, which is described
very thoroughly in Italian by Almo Zanolli in his: Di una vetusta catena sul Levitico, perduta in greco
e conservata in armeno, della sua stretta relazione col commentario di Procopio di Gaza e dei tre codici di S.
Lazzaro, che la contengono, Venice 1938, 78-103.
11 Cf below p. 116f.
12 Cf Lehmann, “An Important Text”, esp. App. I and II, [above, pp. 88ff].
ident from the mere fact that it contains 74 references to ὁ Σύρος (out of which
only 8 find their parallels in Devreesse’s edition of catena fragments)13.
2.2. The Use of Hebrew and Syriac Material in Eusebius’ Commentary

From the very first page of the commentary it is manifest that questions of
language and translation are of great importance to the author, in particular
as regards differences between Syriac and Hebrew on one hand, and Greek
on the other. In the following some characteristic illustrations of this qual-
ity of the text shall be given. The author opens his work by considering the
question how meaning and content can be transferred from one language to
another. This cannot be done through translating word for word or syllable
for syllable, he says; and he continues: often it is necessary to use more – or
fewer – words in the translation than in the original, if one wants to render
the meaning. For “meaning” and “content” the Armenian uses such words
as զաւրութիւն and միտք (basically = “power” and “mind”, corresponding to
Greek δύναμις, resp. νοῦς (or derivatives such as: διάνοια, ἔννοια).14
After these general and fundamental statements the author applies himself
to specific problems concerning the translation of the Hebrew Old Testament.
Problems have particularly arisen from literal translations having been obscure
and therefore of inferior quality.15 The addressee of these remarks, of course,
is Aquila, and this is said quite openly a little later.16
The author refers to “erudite men” as his source(s) of this knowledge, and
correspondingly, a little further on he refers to “those who know Hebrew”
as his authorities.17 Also references to “a certain Hebrew” as authority for a
reading in Hebrew occur,18 and such references raise the question whether
Eusebius’ knowledge of Hebrew was extensive or rather modest.
There can be no doubt that he had a certain knowledge of Hebrew. This
can be seen from e.g. the following observations: In the final chapter of the
commentary, where – in a way related to passages in the introduction such as
13 Cf. above note 7 (where 10 references are given, two of which are without any parallel in the Arme-
nian text).
14 Ven.ed. 1,1-10.
15 Ven.ed. 1,10-18.
16 Ven.ed. 1,18ff.
17 Ven.ed. 3,67f.
18 Ven.ed. 18,65 (about Gen. 2,6; cf Devreesse, Commentateurs grecs, 58); 18,86 (about Gen. 2,8; cf.
Devreesse, Commentateurs grecs, 59, where, however, the reference to “a certain Hebrew” is lacking);
33, 537 (about Gen. 3,22).
those mentioned above – the author comments upon matters of principle and
method concerning linguistic phenomena, problems of Bible translations and
the like, there is a section which presents a number of elementary facts about
the Hebrew language; it is mentioned that Hebrew script is unvocalized, that
there are separate verbal forms for the masculine and the feminine gender,
and that some words are used in plural as opposed to Greek singular.19 The
role of vocalization of the Hebrew text is commented upon explicitly in the
comment on Gen. 49,5f, where the author knows that the choice between the
two translations testified in Greek versions: ταύρος or τεῖχος, is dependent
upon which vowel is used in the Hebrew word: ‫( שור‬šor or šur).20
Thus it is evident that Eusebius is not entirely without knowledge and
understanding of what is characteristic of Hebrew. Therefore the above men-
tioned references to intermediate sources and authorities can be understood
in either of the two following ways: either Eusebius himself considers his
knowledge of Hebrew imperfect, and has had to rely upon other authorities
regarding the Hebrew text, or his “modesty” is of a rhetorical kind, by which
I mean that his reluctance to boast knowledge of Hebrew can be paralleled
with the reluctance evidenced in his homilies against giving exact information
about times and places, numbers and sources.21 It might be added that the
two possibilities here mentioned do not necessarily exclude one another. The
first alternative could be modified, in so far as even an imperfect knowledge
of Hebrew might be considered sufficient for commenting upon difficult pas-
sages of the Old Testament, particularly when supplementary use of sources,
oral or written, could be made, and to the second alternative could be added
that the general rhetorical practice referred to might be considered all the
more appropriate for an author who is counted among those responsible for
writings adversus Judaeos,22 when dealing with Jewish matters.
However, even if the biographical question of Eusebius’ educational stan-
dard and his possible quality of vir trilinguis is difficult to answer, it seems
obvious to me that his working knowledge of Hebrew and his ability to find
the necessary sources, be they Greek or Jewish, have been so great as to allow
19 Ven.ed. 217,1 – 218,34.

20 Ven.ed. 91f. 227-241 (esp. 236-241).
21 On this rhetorical or homiletic practice which Eusebius shares with many homilists of the Old
Church, cf. Lehmann, Per Piscatores, 149ff, and Henning J. Lehmann, On Some Round Numbers in
Some Patristic Texts, Aarhus 1974.
22 Cf. Buytaert, L’héritage littéraire, 6 (and passim).
us to evaluate the commentary as a text in which Hebrew material is used in

a competent manner.
Two presuppositions for Eusebius’ work on the text of the Old Testament
should be made clear, and should not be confused. As will be shown in section
2.3, technically, his point of departure is the Septuagint. On a superior level,
however, the Hebrew Bible is his point of departure or basic textual norm.
This e.g. appears indirectly from his evaluation of the Syriac language and the
Syriac Bible. The value of Syriac, according to Eusebius, lies in its character of
being a language closely related to Hebrew. It would take us too far here to
consider further theological consequences of Eusebius’ approach to the Bible;23
it should be added, however, that his evaluation of the relationship between
Hebrew and Syriac has been a contributory reason why it has been natural to
clear the way for our considerations of Eusebius’ knowledge of Syriac through
the above remarks about his knowledge of Hebrew.
For plain biographical reasons it would be natural to assume that Eusebius,
being born in Edessa, would know Syriac. His above mentioned knowledge of
the close relationship between Syriac and Hebrew is demonstrated very early
in the introduction to the commentary; the Armenian word used is դրացի
(= neighbour).24 Concerning his knowledge of Syriac, reference could also be
made to the way in which, commenting upon Gen. 1,2c, Eusebius mingles
references to Hebrew and Syriac.25 His main point is here that the Hebrew
verb (‫ )רחף‬cannot be rendered through one Greek verb (particularly not the
ἐπιφέρω of the Septuagint),26 and this he sees as a parallel to the fact that the
rendering of words like “slinger” and “archer” in Greek necessitates the use
of more than one word in Syriac.27
In the comment upon Gen. 8,4 it is stated that the mountain of Ararat is
23 For a brief introduction to the discussion in the Old Church of the primary divine authority of ei-
ther the Hebrew Bible or the Septuagint, cf. Sidney Jellicoe, The Septuagint and Modern Study, Oxford
1968, esp. 160.
24 Ven.ed. 1,11f. The text runs as follows: … ի հեբրայեցւոց լեզուն եւ ի նորին դրացւոյն յասորին
25 Ven.ed. 8,108ff.
26 Eusebius uses two times four verbs in order to paraphrase ‫רחף‬. Cf. Henning J. Lehmann, “El Es-
píritu de Dios sobre las aguas. Fuentes de los comentarios de Basilio y Agustín sobre el Génesis
1,2”, Augustinus XXVI, 1981, 127-139 [above, pp. 23ff].
27 This part of Eusebius’ comments is taken over very directly by Diodore of Tarsus (unless the cor-
respondences should be explained from the assumption that the catenist has wrongly attributed a
Eusebian text to Diodore), cf Joseph Deconinck, Essai sur la chaîne de l’Octateuque, Paris 1912, esp.
92f.
called Kordus in Syriac28 – corresponding to the name used in the Peshitta; and
about Exod. 3,14 it is noted that “the Syrian” takes over directly the renderings
of YHWH’s name29 – again corresponding to the Peshitta.
There are thus a number of instances where it is obvious that the author is
very observant of renderings into Syriac, even if again it would be difficult to
state exactly what is his level of knowledge of the Syriac language, and even
if in a number of cases a reference to ասորին with the possible dual sense
of “Syriac” and “the Syrian” would leave the question undecided whether,
when the version is meant, it could be a Greek one. However, for an unbiased
reader, the references, in connection with discussions of readings and exegetic
possibilities, to what is characteristic of the Hebrew and Syriac languages
would lead to the only natural understanding that “the Hebrew” and “the
Syrian” in Eusebius is meant to be the Hebrew Bible and a translation into
Syriac, respectively. We have seen a couple of instances where the “Syriac”
readings run parallel with the Peshitta. In other instances this is not the case.
To my mind, the most natural inference from this ascertainment would be that
we are dealing with a pre-Peshitta stage of the Syriac translation of the Old
Testament, of which some elements were transferred to the Peshitta, whilst
others were eliminated in the final revision.
By way of transition to our next section where the amount of textual mate-
rial from the old versions of the Old Testament to be found in Eusebius is to
be considered, I shall refer to an interesting way of using the versions of the
Old Testament evidenced in the comments upon Gen. 4,4. Here Eusebius’ way
of commenting could be described as homiletically harmonizing, in so far as
he pleads that each of the translators (including the Septuagint, “the Three”
and “the Syrian”) brings forth a side or a part of the content and message of
the Biblical narrative.30 In more than one way this is not typical for Eusebius’
commentary. For one thing, as will be shown below, he does not use “the
Three” very often, for another, rather than harmonizing he would usually
set out “the Hebrew” and “the Syrian” against the others, sometimes “the
Hebrew” against “the Syrian”.
28 Ven. ed. 47,938-941.

29 Ven. ed. 104,214-218. The Armenian wording is: (զ)ահ յահ(ն եւ զ)շարահ յահ(ն).
30 Ven. ed. 37,658 – 38,674. The distribution of verbs to translate Hebr. ‫ שעה‬is well-known from a.o.
Procopius, cf. e.g. Field, Hexapla ad loc. It would now seem evident that Procopius here builds upon
Eusebius (e.g., they use the same verb in Aquila as against other witnesses: Procopius: ἀπεκλήθη;
Eusebius: մխիթարեցաւ, Ven. ed. 37,661).
2.3. References to Bible Versions in Eusebius’ Commentary

As his textual basis Eusebius uses some kind of Septuagint-text. I have not yet
gone into more specific investigations about the question which recension he
uses; as part of the answer to the question whether this will be discernable
through the Armenian translation, it should be mentioned that the Armenian
lemmata offer rather a great amount of variation from the Armenian “Vulgate”.
Thus the Armenian translator of Eusebius’ commentary would seem to have
translated the lemmata directly from the Greek Vorlage rather than using any
existent Armenian version, and therefore – with all necessary reservation –
conclusions about the Greek Bible text in the Vorlage may be drawn from the
Armenian. Apart from the question which recension of the Septuagint has
been used, also that of the possibility of using the Armenian translation of the
commentary in the investigations into early Armenian endeavours to render
the Bible in the vernacular is important, but neither of these questions will
be treated of here. To the basic ascertainment that the text on which Eusebius
comments is a Septuagint should only be added that this, of course, is the
reason why direct mention of “the Seventy” very rarely occurs. In a couple of
instances, however, “the Seventy” are referred to explicitly, viz. for Ps. 2,11f
and Ps. 19(18),5.31
The most conspicuous feature of the commentary, as far as quantitative
figures are concerned, is the restricted use of “the Three” over against a very
great number of references to “the Syrian” and “the Hebrew”. These numerical
facts, of course, constitute a significant part of the background for our state-
ment that Eusebius’ commentary is of importance for the study of the Syriac
Bible of the 4th century. In the following the commentary’s references to the
five relevant Bible versions are listed:32
Aquila: Gen. 1,1; 4,4; Josh. 24,29; Eccl. 3,17; Is. 5,2; Hos. 11,1.
Total: 6 references.
Theodotion: Gen. 2,23; 4,4.
31 Ven.ed. 219,62 – 220,79.

32 Figures quoted as totals give the numbers of verses referred to. It should be noted that sometimes
the author considers more than one textual problem in a verse. On the other hand some of the
references are of a very summary or even indirect character, such as readings referred to under the
heading: “The Syrian also reads like this”, or: “The Syrian does not read like this”.
Symmachus: Gen. 2,23; 4,4; 5,3; Exod. 20,7 (?).

Ὁ Σύρος: Gen. 2,8; 4,1; 4,4; 4,5; 4,7; 4,12; 4,15; 4,24; 5,3; 6,5; 6,13; 6,14; 8,4; 8,7;
8,21; 9,4f; 12,8; 17,14; 18,19; 18,27; 19,22; 22,12; 22,13; 23,6; 23,15; 24,31; 24,50;
24,63; 26,33; 26,35; 27,27; 27,40; 31,7; 31,47; 33,13; 36,24; 37,21; 37,36; 38,18;
38,29; 39,2; 41,16(+ 45); 43,23; 45,10; 45,18; 45,22; 49,3f; 49,5f; 49,8f; 49,27; Exod.
l,12(f); 1,20f; 3,14; 4,14; 4,25f; 5,21; 6,3; Judg. 6,15; 15,8; 1 Sam. 2,5; 4,21; 19,13;
20,20; 2 Sam. 1,21; 5,6; 8,18; 20,18; 20,19; 1 Kgs. 2,5; 2 Kgs. 2,14; Ps. 18(17),12;
18(17),46; 132(131),7; 141(140),7.
Ὁ Ἑβραῖος: Gen. 1,1; 2,8; 2,23f; 3,5; 3,22; 4,1; 4,4f; 4,7; 4,12; 4,15; 4,24; 4,26; 5,3;
6,5; 6,6; 6,13; 6,14, 6,19f; 8,4; 8,7; 8,21; 11,5(+ 7); 11,10; 17,14; 18,19; 19,22; 22,12;
22,13; 23,6; 23,15; 24,2; 24,63; 31,7; 31,47; 32,29(f); 36,24; 36,31ff; 37,36; 38,29;
49,3f; 49,5f; 49,27; Exod. 1,12(f); 4,25f; 23,19; Deut. 26,14; 1 Sam. 15,11; 20,41;
21,5; 1 Kgs. 1,38f; 2 Kgs. 3,4; Ps. 2,11; 19(18),5; Hos. 11,1.
I shall not go into a detailed discussion of this listing. Only a few observations
of rather a technical character should be made. Even from a superficial look at
the survey it immediately appears that in many cases both “the Syrian” and
“the Hebrew” are quoted for the same verse – either because of identical read-
ings or because of differences. In some cases where the Greek text of the com-
mentary is known, differences appear between the Greek and the Armenian
as to the quoting of the two versions. E.g. one tradition may refer to both, the
other to only one of them, and in one instance the Greek refers to “Aquila and
the Hebrew” against the Armenian: “the Syrian and the Hebrew”.33 In such
cases it is difficult to know whether one has added or the other has left out a
reference, or which one might have altered the original; at any rate it would
be unwise to assume a priori that either the Greek or the Armenian is right.
As Greek transmission has gone through catenists (Procopius and others) it
might be natural to assume that their editorial interference has been greater
than that of the Armenian translator. But when the Greek tradition transmits
references to “the Syrian and the Hebrew” (for Gen. 11,3), “the Hebrew” (for
Exod. 3,18), and “Aquila” (for Num. 7,3)34 which find no parallel in the Arme-
nian it would on the background of the otherwise ample amount of references
33 Ven. ed. 108,434-436, cf. Devreesse, Commentateurs grecs, 91 – about Exod. 4,25f.
34 Devreesse, Commentateurs grecs, 65; 88; 96.
to Bible versions be natural to assume that here the Armenian translator or

later copyists have omitted original references.
It would be premature here to discuss how much of the original Greek of
Eusebius’ commentary, including references to Bible versions, can be traced
in Procopius and other catenae, but two points should be underlined, firstly
that it will be important and profitable to utilize Eusebius’ commentary in the
further work on the editing of catenae, and secondly that if we are right in
our main point, that “the Syrian” was a Syriac translation, Eusebius will have
been one of the earliest authors to present in Greek the insights to be gained
from this version. In this way some of the problems which Greek renderings
of the Syriac have caused later investigators may be said to go back to him.
3. Gen. 39,2f in ὁ Σύρος. Montfaucon’s, Field’s, and Rahlfs’

Points of View Confronted with Eusebius’ Evidence
We shall now turn to an example where a Greek rendering of Syriac has led
investigators to wrong conclusions. The text is Gen. 39,2 which has been a key
text in the discussion from Montfaucon to Rahlfs – considered to be a “proof”
that ὁ Σύρος was a Greek version. The argument runs as follows: Hebrew
‫ מצליח‬is translated ἐπιτυγχάνων in the Septuagint, but according to certain
Greek witnesses, esp. Diodore and Procopius, “the Syrian” has κατευοδῶν
or κατευοδούμενος instead. The Greek words, however, are synonyms and
would therefore correspond to the same verb in Syriac, and only in two Greek
renderings would the difference appear.35 Rahlfs adds that the “proof” is not
“ganz unanfechtbar”; however, he only finds it “contestable”, because Sym-
machus, too, is said to have read (κατ)ευοδούμενος, and therefore it would
according to Rahlfs have to be considered whether this reading does in fact
belong to Symmachus having been – perhaps – attributed to “the Syrian”
through a false reading of the abbreviation Συ.
Now Eusebius’ commentary gives us the possibility of apprehending what
lies behind catenists’ and commentators’ reference to “the Syrian” in Gen.
39,2f. Briefly paraphrased Eusebius states that in verse 2 “the Syrian” uses a
35 Montfaucon, Hexapla, 31f; Field, Hexapla, LXXXII; Rahlfs, “Quis sit”, 422f.
verb corresponding to that used in verse 3 by the same version,36 thus in a way
confirming the point of Montfaucon, Field, and Rahlfs that two Greek synonym
words are likely to represent one word in Hebrew/Syriac. But the conclusion
should be that Eusebius tries to bring out the correspondence between verse
2 and verse 3 in the Syriac through the statement that “the Syrian” in verse 2
uses a verb corresponding to εὐοδόω,37 which is the verb of the Septuagint in
verse 3. Therefore it is only the brief rendering of this statement in later authors
– amounting to the bare “the Syrian reads (κατ)ευοδόω in verse 2” – that has
misled later researchers who have concentrated their attention on verse 2 in
isolation from verse 3. It should be noted that the Peshitta (in accordance with
the Hebrew Bible) uses the same verb in verse 2 and verse 3: ‫צלח‬. Therefore the
right understanding of the context in Eusebius would make it probable that
the version in question is closely related to the Peshitta – and would support
the probability of “the Syrian” being in Syriac, once again.
It should be added that the way in which Eusebius here draws exegetic
conclusions from a correspondence in the choice of words in neighbouring
passages or phrases is characteristic. A parallel example can be found e.g. in
his comments on Gen. 4,4f38 where the Septuagint in verse 4 uses δῶρα about
Abel’s offerings, in verse 5 θυσίαι about those of Cain. (Correspondingly the
Armenian lemma reads պատարագք and զոհք, resp., as does the Zohrab
Bible). But “the Hebrew” and “the Syrian” have identical words in the two
verses, Eusebius tells us. In fact, the Hebrew Bible reads ‫ מנחה‬in both verses,
the Peshitta ‫ ܩܘܪܒܢܐ‬in both verses.
The passage commenting on Gen. 39,2f in the above understanding thus
fits in very well with Eusebius’ exegetic method and his technique in using
the versions, esp. those in Semitic languages.
Parenthetically it might be added that Eusebius’ clear reference to “the
Syrian” makes Rahlfs’ deliberations about “Συ” being faultily interpreted as
ὁ Σύρος instead of Σύμμαχος superfluous.
36 Ven. ed. 88,130-133. The text runs: եւ էր տէր ընդ Յովսեփայ եւ էր այր կորովի ասորին ասէ
յաջողեալ, որպէս եւ ասէ իսկ թէ զամենայն ինչ եւ առնէր աստուած յաջողէր (= And the Lord was
with Joseph, and he was a proficient man. The Syrian says “prosperous”, as it is said (in the follow-
ing): God made all that he did to prosper).
37 Armenian յաջողեմ is a close equivalent to Greek εὐοδόω.
38 Ven. ed. 40,742-745.
4. Theodoret of Cyrrhus and Rahlfs’ Argument

from his Comments on Judg. 12,6
4.1. Theodoret’s quaestio on Judg. 12,6 as used by Rahlfs
As mentioned already, Rahlfs finds “einen volständig sicheren Beweis” of the
language of “the Syrian” being Greek in Theodoret of Cyrrhus’ quaestio 19
on the Book of Judges.39 Unfortunately we do not find any comment on this
verse in Eusebius. Rahlfs’ phrasing, however, would compel us to add some
remarks on Theodoret’s use of and comments upon “the Syrian”, even if it
must be noticed that a thorough treatment of this text would lie outside the
scope of the present paper.
In quaestio 19 Theodoret comments upon the passage in the story of Jephtha,
where the Ephraimites are revealed through their dialect, pronouncing ‫סבלת‬
for ‫( שבלת‬Judg. 12,6). The difference in the Hebrew Bible is marked through
‫ ס‬vs. ‫ש‬. This dialectal difference in pronunciation is paralleled by Theodoret
with differences between the Syriac dialects of his own time, and he further
tells us that “the Syrian” renders the difference in Judg 12,6 through the use
of σεμβλά, resp. σεμβελὼ, i.e. “durch einen syrischen Dialektunterschied”40
(as a vs. o corresponds to the well-known difference between the Eastern and
the Western dialect of Syriac). Further Rahlfs notes that this rendering has no
similarity with that of the Peshitta, where the use of ‫ ש‬and ‫ס‬, resp., is taken
over from the Hebrew Bible.
The reason why “the Syrian” should be in Greek, is that “der Unterschied
zwischen σεμβλά und σεμβελὼ wäre in der syrischen Schrift gar nicht zum
Ausdruck gekommen”.41
All the fundamental statements made by Rahlfs are true, of course. How
then could we venture to challenge his conclusion? My argument follows two
lines: One is the positive conclusions to be drawn from the examination of
Eusebius, the other is a questioning of the negative elements in Rahlfs’ arguing
about linguistic and versional matters.
First, of course, as is obvious, the evidence submitted above from Eusebius
has led us to the conclusion that judged from his commentary there can be
39 PG 80, 505-508.
41 Rahlfs, “Quis sit”, 426. (Incidentally, Rahlfs’ evaluation of Syrian script does not prevent him from
taking the difference as to the final vowel to reflect Syriac dialectal variations. Because of the refer-
ence to Melito (cf. note 8) Rahlfs therefore – revealing some doubt – considers whether such differ-
ences could go back to the second century A.D. (p. 427)).
no doubt that “the Syrian” is a Syriac version. Our arguments should not be
repeated here, but it should be added that if the main point in Rahlfs’ conclu-
sions should be accepted as being valid for Theodoret, we would maintain
that they cannot be transferred to Eusebius; in that case we would therefore
have to assume a Greek intermediary link between the Syriac Syrian known
to Eusebius, and Theodoret, a Greek Syrian.
The second line of argument concerns Rahlfs’ reasoning about linguis-
tic factors, esp. the use of Syriac. As has appeared, Theodoret uses dialectal
variations in the Syriac language of his time (as does therefore also Rahlfs)
to explain the readings of “the Syrian”. In our excursus in the next section
we will look a little closer into Theodoret’s use of Syriac and “the Syrian”. It
should be noted here already that the material presented in the next section is
for good and factual reasons not of a quantity to be very conclusive; however,
evaluated in a cautious and unbiased way it would rather lead to the conclu-
sion that the language of the version is Syriac than to the opposite. The most
important thing to be said about Rahlfs’ line of argument about Syriac being
excluded as vehicle of “the Syrian” is that he bases it on the presupposition
that dialectal differences could not be expressed in script in the fifth century,
if they concern vowels. Of course, it is true that Syriac script is a consonant
script, and that the two systems of vocalization now known as Eastern and
Western Syriac are usually considered to be of a later date of origin. As Theo-
doret, however, incorporated the dialectal phenomena, including questions
of vowels, in his reasoning (of course rendered in Greek in his commentary),
it seems to me that sound method would necessitate the question to be left
open how such a linguistic phenomenon could be expressed at such an early
time, rather than considering – a priori – the expression of such an element
to be impossible. There is a certain element of reasoning from non esse to non
posse in Rahlfs’ argument.
As has been stated already, we agree with Rahlfs that “the Syrian” is not
identical with the Peshitta, but this lack of identity should not in itself be taken
as proof that the version in question could not be in Syriac. Particularly, it
should be noted that Greek σ is used to render both ‫ ש‬and ‫ס‬, so the reference
to Peshitta’s consonants being identical with those of the Hebrew Bible is given
an exaggerated value, if it is taken to indicate that there are two – and only
two, reciprocally exclusive – ways of expressing the difference between the
dialect of the Ephraimites and that of the other tribes: either a different initial
consonant or a different final vowel.
4.2. Excursus: Theodoret’s Use of Bible Versions, in particular ὁ Σύρος, in

his Commentaries on Historical Writings of the Old Testament
As has been stated already, this section is not intended as an exhaustive study
of Theodoret as a source of information about biblical versions. On a later oc-
casion I hope to return to the question. Here, however, only a few observations
should be made, necessitated by Rahlfs’ remarks quoted above. It should be
noted that I have not examined all of Theodoret’s commentaries on the Old
Testament. According to Field he seems to be an important witness for “the
Syrian” in Jeremiah and Ezekiel,42 which indeed is not the case regarding the
historical writings of the Old Testament, to which we have confined our prob-
ings in order to cover material parallel to that treated of by Eusebius.
Besides the references to Bible versions, in particular ὁ Σύρος, we have con-
centrated on his statements about linguistic factors. Already from biographical
considerations and from the geographical location of Theodoret’s diocese it
is usually assumed that he has some – maybe even extensive – knowledge of
Syriac.
In his quaestio 60 on Genesis Theodoret advances some reflections on the
history of languages, esp. Syriac (and Hebrew). Syriac is the oldest language
in history, he maintains, – for Adam, Cain, Abel, and Noah are Syriac names.43
From the following quaestio44 it appears that he has expected an objection to
the effect that Hebrew has the precedence of age. This objection, however,
is met with the following reflection: Hebrew is not a “natural”, but a “holy”
language – created by Moses for didactic purposes: Ἡ οὖν ἑβραία πόθεν
ἤρξατο; – Οἶμαι αὐτὴν ἱερὰν εἶναι φωνήν. … διὰ τοῦ Μωσέως ὁ τῶν ὅλων
θεὸς ταὺτην ἔδωκε τὴν γλώτταν, διδακτὴν οὖσαν, οὐ φυσικήν.45
Such “historical” observations are not the only reflections on linguistic
subjects. As has appeared from his comments on Judg. 12,6 he also reveals
some knowledge of contemporary dialects etc. It will have appeared that
his reflections of a general and theoretical character are at variance with the
practical and precise considerations expressed by Eusebius when dealing with
questions of translation and textual evidence. Also as regards quantities there
is great difference between the two authors. The table below will show the
42 Field, Hexapla, LXXVII.

43 PG 80,165.
44 PG 80,165-168.
45 PG 80,165.
distribution of Theodoret’s references to Bible versions in his quaestiones on

historical writings of the Old Testament.
Gen. Exod. Lev. Num. Deut. Josh. Judg. 1 Sam. 2 Sam. l Kgs. 2 Kgs. Total
The Septuagint 46
1 1 1 1 4 8
“The other 1 2 1 3 1 2 10 20
interpreters”47
Aquila 6 1 2 1 1 1 19 9 40
Theodotion 1 1 2
Symmachus 2 4 1 3 2 2 2 3 3 1 23
The Hebrew 1 1 1 3
The Syrian 1 1 1 1 4
Total (excl. the 9 7 7 3 4 3 4 25 15 4 11 92

Septuagint)
It is evident that Theodoret cannot – on the basis of 4 references – be counted

among the important witnesses for readings of “the Syrian” in the historical
writings of the Old Testament. This, however, does not a priori exclude im-
portant information being incorporated in the few references.
In the following we shall briefly comment upon the 4 quotations of “the
Syrian” – with specific reference to the question whether they give any direc-
tion as to the question of its language. One of the four references, of course,
is that to Judg.12,6 that has been treated of already. It has appeared that the
reading of “the Syrian” is here commented upon, among other things, through
information about contemporary Syriac dialects.
Turning to the first of the references (to Gen. 36,24), already the way in
which it appears in Migne’s edition incurs doubt as to whether 4 is the right
figure for references, as there are serious reasons for doubts about the authen-
ticity of this passage, which – even outwardly – has the weak position of an
insertion between quaestiones 92 and 93 on Genesis.48
Already J.L. Schultze, and after him J.-P. Migne, suggested that the passage
might in fact be attributable to Diodore of Tarsus. In Deconinck’s edition of
46 For the same reasons as those stated above for Eusebius the number of explicit references to the
Septuagint is small. It is not included in the “total” figure.
47 Οἱ λοιποὶ or (often) οἱ ἄλλοι mean “the Three”.
48 PG 80,201.
catena fragments attributed to Diodore the fragment in question is included

under the fragments douteux.49 There are two versions of the fragment, one
corresponding to Eusebius’ interpretation, according to which the Ιαμιν of
the Septuagint means “water” in Syriac and Hebrew, whereas the other ver-
sion gives “source” as the reading of “the Syrian”. Deconinck notes that in
the Nicephoros-catena an unknown scholiast attributes the first version of
the fragment to Diodore, the second one to Theodoret, but he adds: “ce té-
moignage n’a pourtant rien de décisif”.50 Thus it will have appeared that the
authenticity of this fragment as part of Theodoret’s commentary is doubtful,
but should it belong – in its second version – to this text, two things might be
noted: 1) Theodoret does not take over Eusebius’ understanding, 2) there are
reflections about the Syriac word for “source” connected with the reference
to “the Syrian”.
The last mentioned fact is valid also for the reference to 1 Kgs. 1,9.51 Here
again the subject is the Syriac word for “source”; it is noted that “the Syrian”
calls the source of Rogel Ἀïνὰ.
Finally, quaestio 39 on 1 Kgs., concerned with 1 Kgs. 12,10, mainly consists
of a reference to the parallel in Chronicles (2 Chr. 10,10). It is said that the
reading here is clearer than that of 1 Kgs., and then it is added: “The Syrian
interprets in the same way, as does also Josephus”.52
To summarize: Evidence about “the Syrian” of the historical writings of
the Old Testament is very scarce in Theodoret. There are references to Syriac
usage in other passages containing no references to “the Syrian”, as when e.g.
we are informed – regarding 2 Kgs. 2,3 (quaestio 6)53 – that “son” is used in
Syriac and Hebrew about a single specimen of a certain species, in this case
“the sons of the prophets” = the prophets. Such cases are not too numerous,
and as far as references to “the Syrian” are concerned, it can be concluded, that
if questions of authenticity could be answered in the positive for Gen. 36,24,
such references are usually connected with considerations of Syriac usage; the
word “usually”, however, has no great strength when used about 2-3 cases
out of 3-4.
Even if we would therefore have to admit that our probings have not al-
49 Deconinck, Essai, 164f.

50 Deconinck, Essai, 165.
51 In Migne’s edition it is part of the last quaestio on 2 Sam. (No. 45), cf. PG 80,667s.
52 PG 80,704.
53 PG 80,748.
lowed us a very clear or full picture of Theodoret as a witness of “the Syrian”,

on the other hand, it seems permissible to maintain that the “certainty” of
Rahlfs’ “proof” has little background in an overall and unbiased evaluation of
Theodoret’s references to Syriac language and “the Syrian”; therefore, at any
rate, our excursus has given supplementary reasons for the point of view – to
put it cautiously – that Theodoret’s quaestiones furnish very weak grounds for
arguing against conclusions suggested by the Eusebian text as to what was
“the Syrian” of the mid-4th century.
5. Conclusions
In this paper an attempt has been made to show the importance of Eusebius of
Emesa’s commentary on historical writings of the Old Testament, particularly
in one respect: as evidence of “the Syrian” and of the fact that this version is
in Syriac. This has been argued, firstly on the basis of the general impact of
Eusebius’ references to linguistic questions and to Bible versions, esp. “the
Hebrew” and “the Syrian”; secondly on the basis of numerical facts when
counting his references to versions; and thirdly on distinctive interpretations
that have been analyzed for their textual basis. Particular weight has been
laid on the explanation of his reading of Gen. 39,2f, which makes it possible
to invalidate one of the main reasons given for the theory of “the Syrian” be-
ing in Greek. It has been shown that the arguments of Montfaucon, Field and
Rahlfs on this point are untenable, and that nothing prevents, indeed, facts
rather strengthen the probability of the very natural assumption that “the
Syrian” is in Syriac. In addition Rahlfs’ specific argument from Theodoret’s
comments on Judg. 12,6 has been considered, and it has been shown that the
“vollständige Sicherheit” of his proof is open to serious objections.
Therefore it is hoped that this paper will have argued convincingly for the
fact that it is important for patristic scholars as well as students of the Old
Testament and of early Syrian church history to take account of Eusebius’ com-
mentary, not least for its evidence about the Syriac Bible of the 4th century.
vidence of the Syriac Bible Translation in
Greek Fathers of the 4th and 5th Centuries
E
In a recent article1 I have argued that the Bible version referred to as “the Syr-
ian”, ὁ Σύρος, in Eusebius of Emesa’s Commentary on Historical Writings of
the Old Testament can very clearly be shown to be a version of the Bible in
Syriac. The main reason why it has been necessary to argue about what might
very well be considered to be obvious, is that scholars of previous genera-
tions, particularly the Old Testament scholar Alfred Rahlfs, and before him
the famous editor of the Hexapla, Frederick Field, and the renowned Patristic
scholar and editor Bernard de Montfaucon have argued rather strongly that
Patristic references to “the Syrian” concern a Greek version.2
One of their main arguments (concerning Gen. 39,2) can be disproved very
directly on the basis of the evidence of Eusebius,3 and the reason why this
was not seen by earlier scholars, is that Eusebius’ work has come down to us
in Armenian and was not published until a few years ago;4 however, one of
Rahlfs’ supplementary arguments (by himself described as “einen vollständig
sicheren Beweis”) is based on a quaestio on Judg.12,6 in Theodoret of Cyrrhus,
and I have felt obliged in the article referred to, to leave open the question
whether it was feasible that the 5th century author, Theodoret, should have
referred to the Syriac Bible version through an intermediary link in Greek,
even if I suggested that to my mind this is by no means a necessary – or even
probable – assumption.
1 Henning J. Lehmann, “The Syriac Translation of the Old Testament – as Evidenced around the
Middle of the Fourth Century (in Eusebius of Emesa)”, Scandinavian Journal of the Old Testament 1,
Aarhus 1987, pp. 66-86 [above, pp. 107ff].
2 For references cf. Lehmann, “The Syriac Translation”, p. 66 [above, p. 107], notes 1-3.
3 Cf. Lehmann, “The Syriac Translation”, pp. 78f [above, p. 117f].
4 Eusèbe d’Emèse, Commentaire de l’Octateuque, préparé par P. Vahan Hovhannessian, Mekhitariste,
Venise-St. Lazare 1980. The Greek fragments of this text as published, e.g. in Robert Devreesse, Les
anciens commentateurs grecs de l’Octateuque et des Rois, Studi e Testi 201, Città del Vaticano 1959, do
not contain Eusebius’ comments on Gen. 39,2f.
To these considerations of mine it can be added that Hans Norbert Sprenger

in the introduction to his edition of Theodore of Mopsuestia’s Commentary on
the Minor Prophets touches upon the question of Rahlfs’ arguments,5 and he, too,
voices serious doubts as to their reliability. Sprenger assumes Theodoret himself
to be the author of the Greek renderings that are meant to illustrate dialectal dif-
ferences in 5th century Syriac, and he adds, “dass rein philologisch orientierte
Untersuchungen dem Problem des Σύρος kaum gerecht werden können”.6
What is perhaps even more important for a general evaluation of the evi-
dence of Greek-writing fathers about “the Syrian”, is Sprenger’s statement,
that it is obvious that Theodore of Mopsuestia considers ὁ Σύρος to be an
unknown Syrian who translated the Hebrew Bible into Syriac.7 I see no reason
to doubt this statement by Sprenger, and thus Theodore of Mopsuestia adds
his testimony to that of Eusebius.
Still, it remains an interesting question that is not answered in what has
been said so far, what is the character and amount of evidence about “the
Syrian” to be gained from Theodoret of Cyrrhus’ commentaries on writings
of the Old Testament, and what I intend to do in the following pages, is to
make a few comments on that question.
First it is necessary to make a few remarks on Theodoret’s references to
Bible versions in general and on figures and distribution. The main fact to be
noted is that he very often refers to the three Greek versions: Aquila, Sym-
machus, and Theodotion, in order to reach an understanding – if possible,
a better understanding – of the text than that which can be gained from the
Septuagint. For that purpose he also uses Josephus, the Jewish historian, and
to a certain – modest – extent other extra-Biblical sources, such as presumably
onomastica etc., or he refers from the Book of Kings to the Book of Chronicles,
from a historical writing to one of the prophets etc. Often, particularly when
the Septuagint reproduces a Hebrew word or phrase, he gives the Greek
translation or solution of that element on the basis of one or more of the other
versions and sources.
About the Greek Bible versions Theodoret in a number of cases uses such
summary indications as “the three” or “the other translators” – and the like,
5 Theodori Mopsuesteni commentarius in XII prophetas. Einleitung und Ausgabe von Hans Norbert
Sprenger, Göttinger Orientforschungen, V. Reihe: Biblica et Patristica, Band 1, Wiesbaden 1977, pp.
79-83. I owe the reference to Sprenger’s treatment of the question to Dr. Lucas van Rompay, Leiden.
6 Theodori comm. in XII proph., ed. Sprenger, p. 82.
7 Ibid.
E v i d e n c e o f t h e S y r i a c B i bl e T r a n sl a t i o n i n G r e e k F a t h e r s 127
but more often he refers to one, two or all three of the important versions by
their names.
In total – in my counting (on the basis of Migne’s text) – including the
summary references – he refers to Aquila 285 times, Symmachus 489 times,
Theodotion 165 times, and the Quinta 3 times. Against such high figures for
the three stand 61 references to “the Hebrew” and a corresponding figure for
“the Syrian”, viz. 62.8
Now, out of the 62 references to “the Syrian” 34 can be found in Theodoret’s
commentary on Jeremiah (incl. Lamentations and Baruch) and 16 in his com-
mentary on Ezekiel; so for the whole Old Testament outside these prophetic
books there remain very few references to “the Syrian”.
What should be noted in particular here, is that in the commentary on
Jeremiah there is no reference at all to either Aquila, Symmachus, or Theodo-
tion. Apart from Daniel, where there are no references to versions at all, the
Jeremiah-commentary stands quite apart in this respect. It does so also, because
such references to a Greek version of the Bible, as can be found here, refer to
that version as one single entity, the Greek version.9
This difference between the Jeremiah-commentary and the rest of Theo-
doret’s commentaries on the Old Testament, of course, calls for considerations
and explanations, primarily considerations of literary genre or – even more
precisely – considerations of time and situation of the author at the time of
the composition of one work and the other.
Regarding the question of genre it can be mentioned – very briefly – that
among Theodoret’s commentaries on writings of the Old Testament there are
both quaestiones in loca difficilia and running commentaries commenting upon
the full body of a Biblical book. It cannot be argued that quaestiones-commen-
taries, as might be expected, perhaps, have, proportionally, a greater share
8 I shall not here refer to the full statistic information behind the above counting, but only add three
remarks: 1. A general reservation about the quality of Migne’s text might be apposite. 2. In some
instances it may be a matter of discussion whether, what is here counted as a reference to “the Syr-
ian” should be taken to apply to a Syrian author or to Syriac language and usage in general rather
than to the Bible version as such. 3. Usually it seems obvious that such designations as “the other
translators” (apart from such cases, where one of “the three” is explicitly excluded) concern “the
three”, and do not include other translations. – However, even if such reservations as those made
here mean that I do not insist on the absolute and final exactitude of the figures given above, I do
find that for a general evaluation of the weight with which one version or the other appears in
Theodoret, their evidence is clear enough.
9 Cf. esp. PG 81,580: in Jer. 12,5; PG 81,597: in Jer. 15,10; PG 81,660: in Jer. 31,2; PG 81,749: in Jer. 51,20
(Cf. also PG 81,566: in Jer. 8,6, for a closely related formula about the Greek version).
of references to the versions than do the others. It might be noted in passing

that what appears in Migne as Theodoret’s commentary on Isaiah is in fact a
collection of catena quotations, but this “commentary” does not stand out in
particular when compared with the rest, as far as references to Bible versions
are concerned.
As there is thus no distinct difference to be noted between genres, I would
rather turn to the question of the time and the particular situation of the author
when composing, e.g. the Jeremiah-commentary, if that could be determined
with any degree of probability. Permit me to suggest considering Theodoret’s
situation in 448-449, when he was prohibited by imperial decree from going
outside his diocese, as an occasion when he may have been deprived of the
usual tools of his work (including the Hexapla) and thus a setting for the
particular features of the commentary of Jeremiah. However, I must empha-
size that this dating and explanation must for the time being remain entirely
hypothetical.
I find no reason, it should be added, to take the peculiarities of the Jere-
miah-commentary to be reason enough to regard it as inauthentic, as there
are contentual and other links with other writings by Theodoret. Therefore,
it is the more interesting that the way in which Theodoret in the commentary
on Jeremiah compares “the Syrian” with testimonies of the Greek translation
(referred to in the singular) and with elements in Greek language, must in
my evaluation leave the reader with a clear impression of the Greek being
compared to a version in Syriac, i.e. presumably the vernacular of many of his
readers or listeners; and even if this cannot be stated with the same degree of
certainty for other Biblical commentaries by Theodoret (which may perhaps
rather be intended for a Greek-speaking audience) with their scarcity of refer-
ences to “the Syrian”, to my mind there is nothing against the view, that the
impression gained from the commentary on Jeremiah – covering more than
half of the references to “the Syrian” – holds true in general.
As far as the general evaluation of Theodoret’s knowledge of Syriac is con-
cerned, it is usually assumed that he is well versed in that language, whereas
his knowledge of Hebrew is limited.10 On the basis of his commentaries on
the Old Testament alone one could hardly find evidence for more than what
might be called a working knowledge of both languages.
As bishop of Cyrrhus and a brave fighter against the ecclesiastical use of
10 Cf. e.g. Otto Bardenhewer, Geschichte der altkirchlichen Literatur, vol. IV (repr. 1962), p. 221.
E v i d e n c e o f t h e S y r i a c B i bl e T r a n sl a t i o n i n G r e e k F a t h e r s 129
the Diatessaron he must, of course, have met with questions concerning the
Syriac language and people speaking it, and he does, indeed, in a number of
instances refer to Syriac usage – past and present. Incidentally, he is one of
the authors to tell us that Syriac is the oldest language (for Adam, Cain, Abel,
and Noah, are Syriac names, he states), whereas he considers Hebrew to be
the didactic-hieratic language of Moses.11
Leaving aside Theodoret’s assumptions in the field of history of language
and turning again to his use of “the Syrian”, this in my evaluation can be de-
scribed as basically different from that found in Eusebius of Emesa, even if a
number of external features might be alike. Particularly, I would call attention
to the fact that the great value and importance attached to “the Syrian” by
Eusebius is based on the fact that Syriac, to use his own wording, is a neigh-
bour language to Hebrew, the original language of the Bible.12 In one instance
Theodoret uses the same terminology of Syriac being neighbour to Hebrew,13
but whereas this statement is of fundamental theological and hermeneutical
significance in Eusebius and determinative for his exegetic practice, this is not
the case in Theodoret. As mentioned already, this in no way prevents him from
referring to Syriac usage – without necessarily referring to the Syriac Bible
translation. In fact, he does so quite often, e.g. with a view to understanding
names, figures, expressions of measure and weight, and extraordinary words
and phrases in the Septuagint.14 In some cases he discusses grammatical gender
and number etc. in Hebrew and Syriac; thus e.g. he discusses the phenomenon
of plural words in Hebrew and Syriac over against the singular in Greek; and
in one such instance he adds the following phrase: “such an interpretation I
found in the Syrian”.15 To my mind it is unnatural to take such a phrase in such
a context to refer to anything but a Bible version in Syriac; and therefore I find
that other instances where this is not as clearly demonstrable as here should
be read in the light of such an allegation. It may well be true that there is a
11 Cf. PG 80,165ff (quaestiones 60 and 61 on Genesis). For discussions of this topic in Syriac literature,
cf. Lucas van Rompay, Le commentaire sur Genèse-Exode 9,32 du manuscrit (olim) Diyarbakir 22, CSCO
483-484, Scriptores Syri 205-206, Louvain 1986, esp. vol. 484, p. 88, n. 9.
12 The Armenian word used is դրացի, cf. Eusèbe d’Emèse, Commentaire, p. 1. See Lehmann, “The
Syriac Translation”, p. 73 [above, p. 113], note 24.
13 PG 81,1448 (l.9: ἡ Σύρων φωνὴ γειτνιάζουσα τῇ Ἑβραῖᾳ). The question is about the word
φελμουνὶ in Dan. 8,13.
14 Comments upon expressions of measure and weight etc. are particularly frequent in the commen-
tary on Ezekiel, and account for rather a great portion of the references to “the Syrian”.
15 PG 81,1221. The question is about αιλαμ, αιλαμὼθ and θεε, θεείμ in Ezek. 40,7ff.
certain lack of clarity, and, indeed, a lack of numerical and theological weight
attached to Theodoret’s references to “the Syrian”, especially when compared
with Eusebius, and these circumstances may – in the clothing of Theodoret’s
Greek – have contributed to the false assumptions about the language of this
version.
However, considering Theodoret’s use of Bible versions in general and of
“the Syrian” in particular with a view to his use and knowledge of languages
as described quite briefly in this paper, I would, indeed, find it justifiable to
summarize and conclude in stressing four points:
1. Leaving aside the particular question of how to explain the differences

between the commentary on Jeremiah (and Lamentations) on one hand,
that on Ezekiel in a middle position, and the rest of his commentaries on
the other hand, it can be stated in general that the amount of evidence in
Theodoret about “the Syrian” is rather modest;
2. I leave it to experts to decide whether his knowledge of Syriac is deep – and
whether it is much deeper than that of Hebrew;
3. There are in his commentaries on writings of the Old Testament a number
of hints (even if not a great number) to the fact that the language of “the
Syrian” is Syriac. Therefore the often quoted view that it is Greek should
and could not be maintained through references to Theodoret.
4. There is hardly any reason to assume an intermediary link – Greek or
Syriac – between Theodoret and “the Syrian” as his source. Of course, he
must coin his renderings of the Syriac of “the Syrian” in his own Greek,
and nothing excludes the possibility that he may in certain instances have
taken over what had already been said by his teachers or other authorities
– possibly including comments upon readings of “the Syrian”.
In any case, an unbiased reading of Theodoret’s references to “the Syrian”

would to my mind lead to the view that considered as a witness to “the Syrian”
he belongs to the series of authors referred to in this paper; in other words:
Eusebius of Emesa, Theodore of Mopsuestia, and Theodoret of Cyrrhus must
be considered to testify to one unbroken chain of Greek evidence to the Syriac
Bible translation, a chain chronologically stretching from some time before the
middle of the 4th century to a time around the middle of the 5th century, a
chain, indeed, where different links have different appearances.
he Question of the Syrian Background of
the Early Armenian Church once again
T
Some Methodological Remarks
It lies in the very nature of the origins of the Armenian church that it is relevant
and appropriate to look for both Greek and Syrian sources of inspiration and
contacts on many levels and in many fields, those of exegesis, ecclesiastical
law, liturgical language and so on.
Also outside the ecclesiastical sphere related and parallel questions must
be asked about influences upon Armenian architecture, language, crafts and
trade etc.
The purpose of this paper is not to give a historical survey of the attempts
to assess the amounts of Greek and Syrian influx, nor to evaluate on a broad
scale the trends in recent research that stress the Armenian-Syrian relations,
nor those that point in an opposite direction. Rather, my purpose is modest
– and twofold: 1) to venture into the rash and dubious enterprise of asking
a couple of simple questions concerning fields where I can claim no specific
competence, viz. the history of art and the history of liturgy; 2) to illustrate the
complexity of the questions in fields where I hope to have some competence to
choose illustrating examples, viz. the field of patristics and that of the history
of the New Testament.
To me as a non-expert one of the most fascinating areas of progress in
Armenian studies is that of the history of art and architecture. The number
of publications in recent years has been overwhelming, and the amount of
work in measuring, depicting, describing, and interpreting remnants of early
Armenian churches certainly calls for admiration.
Some of the intriguing features of Armenian ecclesiastical architecture are
connected with the “bye-rooms” of the churches, “gavits”, “žamatuns” etc.
A kind of fore-court or “external nave” on the southern side of the church is
often supposed to be a feature of Syrian origin, and I am sure investigators
have good reasons for this assumption, reasons of an archaeological and geo-
graphical order, such as the findings of corresponding architectural elements
in the Syrian area.
Purely geographically speaking, questions of the precise borders of what

is called Syria could presumably easily be answered, and the amount of ar-
chaeological remnants so far identified could be registered; even a non-expert
might be able to find his way through the literature to relevant information.
More intriguing, however, is the question of interpreting the function of the
architectural element in question. Here the art historian must turn to the history
of liturgy, and it seems to me important to raise the question: Is there sufficient
textual basis for hypotheses set forth about the function of the southern “fore-
court” in a specifically Syrian introduction to the mass or in specifically Syrian
practises in relation to sacraments? Answers to these questions might then,
eventually, lead to further questions, e.g. about any Armenian preference for
elements from specifically Syrian churches and so forth, but the basic question
is: What is an art historical, architectural syriacism?
Having bordered upon the history of liturgy I should like to ask a couple of
questions concerning this field. They, too, amount to asking about the defini-
tion of “syriacisms”. If, for the field of architecture and art history, I formulated
my questions on a very broad basis, in the context of the history of liturgy my
basis for asking the questions is much more narrow, as I take my starting point
in one single paper by Gabriele Winkler about a “noticeable passage in the
Armenian creed”1. In the subtitle of this paper, the credal passage in question
is quoted: credimus et in Sanctum Spiritum qui descendit in Jordanem proclamavit
missum2; and Gabriele Winkler is mainly concerned with this mention of Jesus’
baptism in certain Armenian creeds and the function of the Holy Ghost in that
event, for which she tends to point to a Syrian background.
Gabriele Winkler certainly points to a number of interesting texts, e.g.
passages where Ephraem underlines the significance of Jesus’ baptism. But
the only texts outside the Armenian tradition, where the credal elements
as such are clearly identified, are Epiphanius’ Ancoratus, and the pseudo-
Athanasian Hermeneia, whose place of origin is unknown. As supplementary
evidence for a creed containing in the third article elements about Jesus as the
apostle – the one sent by the Father – or about the Spirit descending on the
river Jordan or proclaiming through the prophets, Gabriele Winkler refers to
Justin, Aphraates, the Apostolic Constitutions, and Cyril of Jerusalem, i.e. 1) a
1 Gabriele Winkler, “Eine bemerkenswerte Stelle im armenischen Glaubensbekenntnis”, Oriens Chris-

tianus 63, 1979, pp. 130-162.
2 Other readings instead of proclamavit missum are “proclaimed to (or: through) the apostles” (i.e. mis-
sis/apostolis).
T h e Q u e s t i o n o f t h e S y r i a n B a ck g r o u n d 133
second century apologist born in Palestine and with Greek as his vernacular,
2) the “Persian Sage” who of course wrote in Syriac – around the middle of
the fourth century, 3) a canonical-liturgical collection that is often believed
to stem from Antioch, i.e. a city where the population was mixed as in few
others, linguistically, sociologically and liturgically, in the churches presum-
ably mainly Greek-speaking, and 4) the famous bishop of Jerusalem about the
middle of the fourth century. It should be noted that for Cyril’s creed – and
for other texts – Gabriele Winkler uses the term “Syro-Palestinian” about
their locality, and it could be added that in cases where clear evidence is not
at hand, it is stimulating – and necessary – to try to find the hidden traces;
indeed, my remarks should not be understood as an unfair criticism of Gabri-
ele Winkler; she has indeed raised a number of stimulating questions, in this
paper as well as elsewhere. However, even allowing for her hypothesis that
some liturgical and dogmatic traditions – among them Syrian elements – were
suppressed as heretical in the fourth century, not least in Jerusalem, and that it
may be possible to find some of their traces in Syrian and Armenian literature,
and for the fact that a number of highly interesting Syrian texts – other than
creeds – are taken into account, where the motive of the sending of Christ by
the Father and the significance of Jesus’ baptism are given a central place, I
find that the same simple question as that addressed to the art historian can
be asked within the history of liturgy: What exactly is a liturgical syriacism?
And against the background of the paper by Gabriele Winkler here referred
to, this question could be specified in – among others – the following items:
Do the well-known liturgical connections between Jerusalem and Armenia
count as such? Does material originating from Antioch count as such? With
how many Syrian and Palestinian “schools” should we reckon? And to which
of them did the Armenians turn? Which Armenians and when?
Let me now turn to fields where I hope to be able to answer – or at any rate
specify – some of the questions about “syriacisms”. I first turn to the history
of the Bible text. Here it should be quite easy to define a Syrian element, as
that must be a reading in the Armenian Bible, the presence of which can only
be understood on the assumption of a Syrian Vorlage. The reason why I have
commenced in other areas is that it seems to me that in some cases linguistic
syriacisms are used as evidence of textual syriacisms; by including other areas
of research I did not only want to broaden the principal scope of my paper,
but first and foremost I wanted to find support for an exhortation about the
necessity of respecting the limits of each area.
I have tried elsewhere to substantiate my statement about the untenable
use of linguistic syriacisms in text-historical arguments, particularly the rea-

sons that led Louis Leloir to his answer “probably yes” to the question of a
Syrian Vorlage for the Armenian Epistle of James3. Leloir gave his opinion in
the Münster manual of the old translations of the New Testament4. I shall not
here repeat my discussion of Leloir’s arguments. I shall just resume – and
expand – one point, where the questions of method and principle to my mind
appear very clearly. I am thinking of the Armenian vocabulary of salvation.
In James 1,21 the Greek verb σῶσαι is rendered in the Armenian ver-
sion through կեցուցանել, which is an evident syriacism, and the theological
significance of this fact is commented upon by Leloir in some considerations
about the Semitic languages, partly taken over from Joseph Molitor5. Here it is
particularly asserted that the vocabulary of salvation in the Semitic languages
and also other elements of their religious or theological vocabulary is much
more “dynamic” than e.g. Greek; it is further maintained that Armenian and
Georgian translations in a number of cases have taken over the dynamics of
the Semitic languages, particularly from Syriac.
On this basis the text-historical conclusion is drawn that this linguistic fact
points towards a Syrian Vorlage of the Armenian Epistle of James.
Now, for one thing, it can be demonstrated very easily that if the linguistic
argument quoted – of a translational loan – carried any weight in text history,
the conclusion should have been the opposite, as Js. 1,21 is one out of five in-
stances in the epistle, where σώζω is used, but the only one where կեցուցանեմ
is used in the Armenian version6.
However, the Peshitta uses ‫ ܚܝܐ‬for σώζω four times in the Epistle of
James, so there is no specific correspondence between the choice of words
in the Syrian and the Armenian Bible, i.e. there is no specific textual corre-
spondence.
The perspective can be further widened, when taking the whole of the New
Testament into account. This certainly shows that in the Peshitta ‫ ܚܝܐ‬is the fa-
3 Henning Lehmann, “Some Questions concerning the Armenian Version of the Epistle of James”,
in Aarhus Armeniaca, Acta Jutlandica 57, Aarhus 1982, pp. 57-82 (hereafter Lehmann, “Armenian
James”) [above, pp. 37ff].
4 K. Aland, ed., Die alten Übersetzungen des Neuen Testaments, die Kirchenväterzitate und Lektionare,
Arbeiten zur neutestamentlichen Textforschung 5, Münster 1972. (Leloir’s contribution is found on pp.
300-313.)
5 Joseph Molitor: Grundbegriffe der Jesusüberlieferung im Lichte ihrer orientalischen Sprachgeschichte, Düs-
seldorf 1968, see esp. Leloir, p. 307.
6 Cf. Lehmann, “Armenian James”, p. 80, Table 7 [above, p. 64].
vourite word for σώζειν (86 out of 107 cases). In Armenian կեամ/ կեցուցանեմ
accounts for only 30 of the 107 cases, փրկեմ for 25, ապրիմ / ապրեցուցանեմ
for 48, and 4 further verbs for the 4 remaining cases7. Considering the situ
ation on Syrian ground it is no wonder that 29 out of the 30 cases with կեամ
/ կեցուցանեմ correspond to ‫ ܚܝܐ‬in Syriac – but that is of little interest in a
text-historical argument.
A thorough-going investigation of the Semitic vocabulary of salvation
should above all include Biblical Hebrew, it seems to me. By far the most
frequent verb for “salvation” in the Hebrew Bible is ‫ישע‬. In the Syriac Bible,
there is a clear preference for ‫ ;ܚܝܐ‬so in this case it seems unwise to treat
the Semitic languages as a whole. As regards the Syrian Bible, the second
most important word is ‫ܦܪܩ‬, at any rate, when you count the derivatives
of it, which are nearly always used for the nouns: saviour and salvation.
In Armenian ‫ ܦܪܩ‬appears as the loanword փրկեմ, and in Armenian as in
Syriac this is by far the most frequent stem in the nouns: փրկիչ for saviour,
փրկութիւն for salvation.
There is thus, certainly, very clear linguistic evidence in the Armenian
vocabulary of salvation of Syrian inspiration and influence; and it would be of
high value for a general and overall estimation of the cultural processes of the
Armenians’ absorption of Syrian and other elements, if historians of language
could tell us, firstly at what time and in what environment the loanword փրկեմ
and the translational loan կեցուցանեմ for “save” obtained their Armenian
naturalization, and secondly whether and how and why preferences changed
with the times.
I suppose that it would be difficult to answer the first question, that of
time and environment, as the borrowing must presumably have taken place
before the invention of the Armenian alphabet, i.e. before the fifth century.
Concerning the second set of questions – those of the reasons for choosing
one word or the other, I have made a few observations of very modest scope,
which will here be submitted for discussion.
If we first turn to the gospels for more specific information, we could as-
certain that out of the 18 (or 19) cases where the verb կեամ / կեցուցանեմ is
chosen to translate σώζω, 7 of them occur in healing narratives in the formula:
“Your faith has cured you”:
7 Cf. ibid., p. 77, Table 6 [above, p. 62].

Mt. 9,22 ≠ Mk. 5,34 ≠ Lk. 8,48
Mk. 10,52 ≠ Lk. 18,42
Lk. 7,50
Lk. 17,19
In 6 cases it occurs as the contrast of Greek ἀπόλλυμι / Armenian կորուսանեմ,

in such contexts as e.g.: “Whoever cares for his own safety is lost”:
Mt. 16,25 ≠ Lk. 9,24
Mt. 18,11 (≠) Lk. 19,10
Lk. 9,56
In 3 cases the verb occurs in the formula: “The man who holds out to the end
will be saved”:
Mt. 10,22; 24,13; Mk. 13,13
For a translator having three words at his disposal, which may be described
as close synonyms, the final choice may be determined by a number of factors;
first, I should think, the specific semantic connotation of each of the words. This
to my mind would account for the choice of կեամ to contrast կորուսանեմ /
ἀπόλλυμι. It should be noted, however, that at the time of the translation of
the gospels, the choice of word had not yet been standardized, as can be seen
from the fact, that the parallel in Mark to Mt. 16,25 and Lk. 9,24 (Mk. 8,35)
has ապրեցուցանեմ.
Secondly, it seems that a certain amount of tradition or setting of a common
standard had made itself felt already at that time, as appears from the standard-
ized choice in the phrase: “Your faith has cured you”. Thus it is interesting
to note that in the closest context of this phrase in Mt. 9,22, the verb փրկեմ is
used: “(v. 21) for she said to herself, “If I can only touch his cloak, I shall be
cured” (փրկիմ). (v. 22) But Jesus turned and saw her, and said, “Take heart,
my daughter; your faith has cured you” (կեցուցին). And from that moment
she recovered (փրկեցաւ)”.
That there is a process of standardizing going on, can be further demon-
strated, if we leave the New Testament and look at one of the few texts which
have twice been translated into Armenian. I am thinking of one of the homilies
by Severian of Gabala, occurring both in pure classical Armenian8 and in a

more “Hellenizing” translation9. As concerns the verb σώζω, which occurs only
twice in the Greek text10, the two translations agree in choosing the rendering
ապրիմ / ապրեցուցանեմ. But for the noun σωτηρία, the primary translation
has ապրանք, a derivative of ապրիմ, whereas the secondary translation has
փրկութիւն; and in the four cases where the noun σωτήρ occurs, the second-
ary translation has փրկիչ in all cases, the primary translation in only one. In
the other three cases the primary translation reads: Քրիստոս (= Christ), տէր
(= Lord), and տեառնորդին աստուած (= Son of the Lord God, incidentally, a
very interesting christological title).
These variations to my mind take us into a field which I would call the
semantics of religious language, or the semantic and emotional variations ac-
cording to variations of devotional and liturgical practise. To quote a modern
parallel: One could often guess whether a Protestant preacher belongs to an
evangelical, a fundamentalist, a high church or another movement from his
preference for certain nomina sacra or “christological titles”.
In another text, which I have had the opportunity to comment upon, viz.
the double translation of one of Eusebius of Emesa’s homilies, into Armenian
and Latin respectively, there is rather a large amount of variation in the use of
nomina sacra and christological titles. Here the inner ears of both the translators
seem to have adapted to the devotional or liturgical idiom of their environ-
ments11.
To illustrate the variety and the choice of words at the time, to which the
translation of Eusebius’ and Severian’s homilies belongs, the words chosen to
translate σώζω and derivatives will be listed here – in those cases, where the
Greek original is preserved, i.e. for homilies IX and XII in the Akinian collec-
tion12. In the ninth homily σωτήρ is found twice, once rendered through տէր
8 No. XI of the homilies edited by N. Akinian under the name of Eusebius of Emesa (“Die Reden des
Bischofs Eusebius von Emesa”, Handes Amsorya 70-73, 1956-59. No. XI is found in Handes Amso-
rya 73, 1959, cols. 1-30) [cf. below, pp. 141ff].
9 This version is published under the name of John Chrysostom in the Venice Mechitharists’ edition
of Chrysostom’s Homilies on the Epistles of S. Paul, vol. II, Venice 1862, pp. 694-715.
10 The Greek original of the homily is found in PG 59,653-664 (CPG 4202).
11 The homily in question is no. II in the series edited by N. Akinian (cf. note 8), Handes Amsorya 70,
1956, cols. 385-416. For my comments cf. Henning J. Lehmann, Per Piscatores. Studies in the Armenian
version of a collection of homilies by Eusebius of Emesa and Severian of Gabala, Aarhus 1975, see esp. pp.
45-102.
12 Cf. note 8. Homily IX: Handes Amsorya 72, 1958, cols. 161-182; Homily XII: Handes Amsorya 73, 1959,
cols. 161-182.
(= Lord), once through կենարար, σωτήριος is rendered կենդանացուծիչ.

In the twelfth homily there are a number of lacunae, so out of the five cases
where the verb σώζω is used, only two find their parallels in the Armenian
version, both with կեամ / կեցուցանեմ, and σωτήρ is rendered կենարար. In
other words, apart from the first mentioned instance, all cases have a deriva-
tive of կեամ.
Certainly, this material is modest in quantity, but even so, the examples
seem to me to illustrate that in this connection a lot of consideration needs to be
given to questions of the psychology of language, of devotional and liturgical
traditions, of the technique of translators, and, indeed, of the development of
the Armenian language in the early centuries of Armenian literature, before a
clear-cut description could be given of the factors that determined the choice
of words within e.g the vocabulary of salvation in a certain text, let alone that
of nomina sacra and christological titles. To use the modern parallel once again,
the pietist who spoke much more of “Jesus” and “the blood of the Lamb” than
of “Christ” and “the Lord”, and much more about the Son than about the
Father and the Holy Ghost, was little aware of the etymology of his favourite
words.
When the question of syriacisms is seen in this perspective, the Armenian
person or community – whether a Christian congregation or a pagan group
– who first took over ‫ ܦܪܩ‬and made it into փրկեմ and those who modelled
կեցուցանեմ on the afel of ‫ ܚܝܐ‬were certainly under Syrian influence; but the
translator of a liturgy or a homily or a gospel who used one or other of the
words – or preferred a third possibility – generations later, was influenced
by a number of factors such as those hinted at above, and his choice of one
of the words of Syrian origin could certainly not be taken as evidence of the
language in his Vorlage.
Incidentally and parenthetically our translator’s modern colleague should
– to my taste – avoid such renderings as vivificare or lebendig-machen for
կեցուցանեմ; the Armenian translators usually made a clear difference be-
tween σώζω, to save, and ζωοποιέω, to vivify13, and that difference should
not be blurred.
It may be appropriate to conclude this paper with an apology, because
I have called attention to a number of truisms. Certain observations of the
scholarly discussion of the question of Syrian influence on Armenian church
13 Cf. Lehmann, “Armenian James”, p. 80, Table 8 [above, p. 65].

history and culture and of the definition of a syriacism and the consequences
to be drawn from syriacisms in one field when asking questions in another,
however, seemed to me to make it appropriate, even necessary, to elicit such
truisms. A full picture of Armenian ecclesiastical and cultural history can of
course only be drawn on the basis of some kind of synthesis of the various
elements of that church and that culture; what I have tried to argue is, for one
thing, that when we speak of Syrian influence, we ought to be very precise
about what we mean by that, and for another, that a sound synthesis can be
reached only on the basis of analyses respecting within each field the laws
and demands of that field. A mixture of arguments from various fields and of
premature attempts to synthesize will only serve to make the attempted full
and final synthesis opaque, and opacity should be avoided.
What Translators Veil and Reveal
Observations on two Armenian
Translations of one Greek Homily
1. Introduction
The homily, referred to under the title: In Chananaeam et in Pharaonem; et quod
non volentis neque currentis, sed miserentis sit Dei (PG 59,653-664), which is – as
are so many homilies – in the Greek manuscript tradition handed down to
us under the name of John Chrysostom, is in Maurice Geerard’s Clavis Pa
trum Graecorum rightly placed as No. 4202, i.e. under the name of Severian
of Gabala.1
The attribution to Severian is due to B. Marx,2 whose views were accepted
by later scholars, in particular H.-D. Altendorf.3 In my book Per Piscatores I
added a few internal arguments to those brought forward by Marx;4 Ι further
argued5 that the very occurrence of the homily in the series of five Armenian
homilies by Severian edited by N. Akinian in 1958-596 (where his name – Sebe-
rianos – occurs in the title of two of the homilies), added external arguments
for the attribution to Severian.7
1 M. Geerard, Clavis Patrum Graecorum, II, Corpus Christianorum, Turnhout 1974.

2 B. Marx, “Severiana unter den Spuria Chrysostomi bei Montfaucon-Migne”, Orientalia Christiana
Periodica 5, 1939, 281-367.
3 H.-D. Altendorf, Untersuchungen zu Severian von Gabala (Diss.) Tübingen 1957.
4 Henning J. Lehmann, Per Piscatores. Studies in the Armenian version of a collection of homilies by Eu-
sebius of Emesa and Severian of Gabala, Aarhus 1975 (hereafter quoted: Per Piscatores), see esp. p.
319-327.
5 Cf. also Henning ]. Lehmann, “The Attribution of certain Pseudo-Chrysostomica to Severian of
Gabala confirmed by the Armenian Tradition”, Studia Patristica X, ed. F.L. Cross, Texte und Untersu-
chungen zur Geschichte der altchristlichen Literatur 107, Berlin 1970, 121-130 (hereafter quoted: Attribu-
tion).
6 N. Akinian, “Die Reden des Bischofs Eusebius von Emesa”, Handes Amsorya, 1956-1959. Homilies
I-VIII of this collection are by Eusebius of Emesa, homilies IX-XIII by Severian of Gabala, cf. Per Pis-
catores.
7 Cf. Per Piscatores, p. 319.
In Per Piscatores I added a few additional comments on the interesting fact

that this homily has been translated twice into Armenian. The translation
different from that published by Akinian was edited by the Mechitharists
of Venice in 1862 – here as in the Greek transmission as a homily by John
Chrysostom.8
Below, I shall register some of the characteristics of the two translations
in order to illustrate, on the basis of the early history of this text, some prob-
lems and questions of translation in Antiquity – some of them well-known,
indeed – and to emphasize the importance of taking the different text-historical
stages (including those of translations) into account when handling the exist-
ing manuscript material, Greek and Armenian. It should be underlined here
already that I have made no investigations of the manuscript material as such;
what is said below will not take us further than is possible on the basis of the
editions of the Armenian texts mentioned above and Migne’s edition of the
Greek text.
2. Differences between the two translations,

and their relation to the Greek
For reasons of space the following paragraph cannot give an exhaustive de-
scription of the linguistic features and other characteristics of each of the two
translations. What shall be attempted is to draw the attention to some of the
differences and correspondences in grammar and syntax, vocabulary and
style, the handling of Biblical material, and the disposition of the text. Through
this selective description it is hoped that the profiles of the two translators
will appear a little more clearly. Hopefully, some insights will also be gained
which may be of value for future investigators into the history of the text,
and help clarify which methods should be applied when using such material
as the two Armenian translations in the attempt to bridge the gap between
the original author of the late 4th and early 5th century and the theologian
and historian of today, who wants to take advantage of the text transmission
in its full breadth in order to understand, primarily Severian’s message and
context, and in the second place the historical and theological background of
those who wanted to make his homily available in the Armenian vernacular
for their congregations.
8 J.B. Aucher, ed., Յովհաննու Ոսկեբերանի Կոստանդնուպօլսի Եպիսկոպոսապետի Մեկնութիւն

Թղթոցն Պօղոսի, Venice 1862. Our homily is found in vol. 2, p. 694-715.
W hat T ranslators V eil and R eveal 143
2.1. Grammar and Syntax

In this section only one very prominent feature distinguishing the two trans-
lations shall be pointed out, viz. the way in which they handle Greek parti-
ciples. This can be done in a way sufficient to our purpose through a mere
counting.
In the Greek text as published by Migne there are in my counting 273 par-
ticipial constructions. The various ways of rendering these constructions by
the Armenian translators are shown in the following table, where A (as will be
the case in the following) is used for the Akinian text – or its translator, B for
the text of the Aucher-edition of 1862 – or the translator behind this version.
A B
Armenian participle with -եալ 47 74
Armenian participle with -ող 4 16
Armenian participle with -իչ 3 3
Armenian participle with -ոց 2 2
Armenian infinitive in the instrumental case 4 32
Armenian relative clause 37 69
Other possibilities 103 64
Rest 73 13
273 273
The most remarkable differences are those concerning the use of the Arme-
nian infinitive in the instrumental case and of the Armenian participle with
-եալ. In a brief description such as this I find it permissible to say that the
Armenian translator B considers the infinitivus instrumentalis as a participium
conjunctum, so the two figures taken together show the tendency in B of choos-
ing whenever possible a “participle” to render a participle. Thus this feature
in itself already reveals the B translator’s wish to make a direct tracing of the
Greek, and thereby it is already suggested that he is closer to the “Hellenistic
School” than translator A.
The renderings referred to under the headings “relative clause” and “other
possibilities” can be said to be the counterpart to the two groups of “parti-
ciples” first mentioned. Again translator B more often than A chooses the
closest “paraphrastic” rendering, that of the relative clause as compared with
other possibilities. And the figure 103 for “other possibilities” – showing that
translator A in more than a third of all cases chooses a “paraphrase” other

than a “participle” or relative clause – may justify my use in Per Piscatores of
the words “slavish” and “paraphrastic” for B and A, respectively, even if,
admittedly, as shown above, also translation B in some cases must be said to
use some modest kind of paraphrase.9
Two further observations might above all illustrate the limitations of this
survey. The relatively very great differences concerning participles with -ող
reveal that a more thorough investigation should – obviously – among other
things take into account whether the Greek has a past or a present participle.
Finally, it should be noted here that the reason why there is such a great num-
ber in A as 73 referred to under the heading “Rest” (against 13 in B) is mainly
due to the fact that A (as opposed to B) has nothing to correspond with the
introductory section of the Greek homily (amounting to about two and a half
Migne columns) (cf. below, section 2.4.).
2.2. Vocabulary and Style

The first of our selective probings into the vocabularies of the two translators
will concern the rendering of Greek components with εὐ-, as it is evident, even
at a first glance, that translator B, wherever possible, prefers a rendering with
Armenian բար- and also the rendering of other parts of the words in question
reveals this tendency to make a direct tracing or calque of the Greek – once
again hinting at his “Hellenistic” affinities.
In the following survey – after giving the Greek word – I therefore first list
the rendering in translation B. The figure following shows the number of occur-
rences in B. Then for comparison – the renderings and number of occurrences
in translation A are given. When words are given in parenthesis or quoted in
the grammatical case or conjugation of the text, it either indicates that doubts
can be raised whether we are dealing with a direct and proper rendering or a
more paraphrastic reproduction, or the form, construction and context of the
word is thought to give some hint about the translator’s technique or about
the need of further consideration of e.g. the textual Vorlage.
9 Cf. Per Piscatores, p. 323-325.

εὐγένεια B: բարետոհմութիւն: 7
A: ազնուականութիւն: 7
εὐγνωμονέω B: բարեմտեմ, բարեմտութեամբ: 2

A: 0
εὐγνωμοσύνη B: բարեմտութիւն: 2
A: բարք: 1
εὐεργεσία B: բարեգործութիւն: 5
A: բարերարութիւն / (զամենայն) երախտաւոր
բարերարութիւնսն / (երախտաւորեալ) զերախտիսն: 3
εὐεργετέω B: բարեգործեմ: 1
A: 0
εὐεργέτης B: ի բարեգործութենէն (2), բարեգործ (1): 3

A: կենարար: 1
εὐνοέω B: բարեմիտ: 1
A: 0
εὐνοία B: բարեմտութիւն: 2
A: 0
εὐνοϊκός B: բարեմտութիւն: 2
A: 0
εὐσέβεια B: բարեպաշտութիւն: 2
A: աստուածպաշտութիւն: 5
εὐφημία B: բարեբանութիւն: 4
A: 0
The fairly great number of instances where A has fewer occurrences than B or
none at all are mainly due to the difference in length mentioned already (cf.
below, section 2.4.). The wording “wherever possible” above mainly refers
to the fact that certain words with εὐ- (among them words representing the
central Biblical and liturgical vocabulary) seem to have found renderings in
the Armenian which even the B translator – as a matter of course – had to
respect. I have listed the following occurrences of that kind:
εὐαγγελικός B: աւետարանական: 2
A: 0
εὐγνώμων B: գեղեցիկ: 1
A: 0
εὐλογέω B: աւրհնեմ: 1
A: աւրհնեմ: 1
εὐπαράδεκτος B: դիւրընկալ: 1
A: (հասարակն պատուեալ): 1
εὐφραίνω B: զուարճացուցանեմ: 1
A: 0
εὐχαριστέω B: գոհանամ: 1 (Greek: 2)

A: 0
εὐχαριστἰα B: գոհութիւն: 2
A: 0
εὔχρηστος B: պիտանի: 1
A: (պիտանացու ի պէտս): 1
The lists speak for themselves. Particularly, it could be said, the translations of
such words as εὐγένεια and εὐσέβεια set apart the two translators as having
their base in classical Armenian vocabulary (A) and an “artificial” vocabulary
directly reproducing the Greek (B), respectively.
Also a listing of the translations of πρό and compounds with προ- sets the
two translators apart, one (A) having a preference for յառաջ with derivations,
the other (B) for կանխաւ and compounds with կանխ-, and for նախ- with
derivations. Figures are as follows:
A B
առաջագոյն 23 17
յառաջ etc. 15 9
առաջի 2 0
կանխաւ, կանխա- 1 23
նախ etc. 0 15
Rest 4 0
45 64
Here it would be less evident on semantic grounds to point out a classical or

“Hellenizing” element in one translator or the other, but again it is evident
that the linguistic background of the two translators differs widely. I am not
capable of judging whether one could here introduce the notion of “dialectal
differences”.
One of the instances of εὐ-words could lead us to a further observation on
the different trends in the two translators’ choice of words. I am thinking of
the one case where A translates εὐεργέτης: կենարար over against B’s Greek
calque: բարեգործ.
This to my mind takes us to the very important part of any ecclesiastical
translator’s vocabulary: that of God’s names and Christ’s “titles”.
In Per Piscatores10 I noted that the Greek homily uses σωτήρ 15 times. In
all these instances B has փրկիչ. Ten of the occurrences are in the part of the
homily not to be found in A. But the five remaining cases show an extreme
richness and variation in A, in so far as the following “translations” are used:
Քրիստոս, տէր, տեառնորդին աստուած, փրկիչ, տէրն եւ աստուած. The
reason for using quotation marks around the word “translations”, is the con-
sideration that should obviously be made here: whether at least part of this
variety was to be found in the Greek Vorlage for translation A.
If it is not all accounted for in the Greek Vorlage (which we may never
know), the vocabulary would presumably take us to the liturgical, ecclesiastical
and religious “milieu” of the translator; and what may be the most interesting
element in our present context is the occurrence of the very rare “Christologi-
cal title” տեառնորդի. As mentioned in Per Piscatores11 this title occurs 5 times
in the Eusebius homilies and 4 times in the Severian homilies of the Akinian
collection. Thus e.g. in the quotation of Mt. 21,16 in the Severian homily No.
13 of the collection, the mere “Jesus” of the Bible text is replaced by փրկիչն
տեառնորդի.
If the translation of this collection of homilies is due to one unknown Ar-
menian translator, the description of this element as belonging to his “eccle-
siastical background and milieu”, of course, only gives us little help – if any
– to identify the origin and history of this interesting “title” of Christ. But
even if we are not now able to determine more closely the historical genesis
and theological emphasis of such an Armenian word, it would indeed be
10 Cf. Per Piscatores, p. 322. For a broader discussion of elements in the Armenian vocabulary of
salvation, cf. also: Henning J. Lehmann, “Some Questions concerning the Armenian Version of
the Epistle of James”, Aarhus Armeniaca, Acta Jutlandica LVII, Aarhus 1982, p. 57-82, esp. p. 76-80
[above, pp. 61ff], and Henning J. Lehmann, “The Question of the Syrian Background of the Early
Armenian Church once again. Some Methodological Remarks”, Studia Patristica XVIII.4, ed. E.A.
Livingstone, Kalamazoo 1990, p. 255-262, esp. p. 257-261 [above, pp. 137ff].
11 Per Piscatores, p. 322, note 2.
quite unsatisfactory to maintain that the nomina sacra of the homily are there
by accident, and even if the anonymity of the translator as a person should
never be disclosed, some insights into his background and context might be
obtainable – and important.
The borderline between the analysis of vocabulary, the assessment of stylis-
tic features and the evaluation of a translator’s technique could and should not
be drawn too strictly. The following element might most often be referred to as
“stylistic ornament”. I am thinking of a point in S. Lyonnet’s characterization
of the vetus armena-translation of the Bible, by Lyonnet called arm 1: “Au lieu
d’écrire “l’Egypte, Jérusalem, Israël, la Galilée …”, arm 1 préfère les tournures
suivantes: “la terre des Égyptiens …”, “les régions des Galiléens … “.12 This
phenomenon is listed by Lyonnet as one of the “criteria” with which to de-
fine a vetus armena gospel reading. In our homily it can be observed that in a
quotation of 1 Sam. 4,8, translator A in fact translates τὴν Αἴγυπτον: զերկիրն
եգիպտացւոց. (B, as was to be expected, has the mere զԵգիպտոսն).
What is more interesting, maybe, for our purpose is that in a context where
no Bible quotation is concerned, our translator A renders ἐν τῇ Ρώμῃ through
ի հռոմայեցւոց քաղաքին. A little later on, ἐν τῇ Ρώμῃ is rendered յայսպիսի
քաղաքի. (B, in both cases has ի Հռովմէ).
The interest of these details is that they link our translator, not necessarily
(or only) to a certain Bible translation, but to the traditions or “techniques”
of translation among the translators responsible for that version. Thus, we
may through this observation have identified one element in his “profile” or
“milieu”.
2.3. Biblical Material Used

Considerations of space exclude any detailed examination of the Bible text used
in the two translations. However, having touched upon the stylistic element
that seems to connect translation A with the vetus armena, it might be appropri-
ate to add a few remarks about Biblical material used in the homily.
The first remark is of rather a formal character. In relation to my references
to Per Piscatores it seems appropriate to note that in my examination of the 13
homilies of the Akinian collection I gave particular attention to the gospel quo-
tations. The study of the Biblical material used in the homilies was restricted in
two ways. For one thing, when only gospel quotations were considered, this
12 S. Lyonnet, Les origines de la version arménienne et le Diatessaron, Biblica et Orientalia 13, Rome 1950, p.
53. Cf. Per Piscatores, p. 323.
was due to the fact that the gospels represent the area where – considering
the present state of research – elements of a vetus syra and/or a vetus armena
stratum could be identified with the greatest degree of probability. Secondly,
it should be noted that the quotations were only – or mainly – considered for
their value as criteria in the discussion of the question of authorship. For the
eight homilies belonging to Eusebius of Emesa (Nos. I-VIII) the representation
of vetus syra elements corresponding to related phenomena in Latin transla-
tions of homilies by Eusebius to my mind constitutes an important part of the
argument for Eusebius’ authorship.13 For Severian, however, the amount of
gospel quotations is small, and few clear characteristics other than those of
favourite quotations and exegetic topics could be used to compare the Arme-
nian homilies with existing Greek homilies.
In the homily under consideration there are very few gospel quotations.
Again translation B usually, rather “slavishly” follows the Greek. One interest-
ing exception shall be touched upon a little later on.
Translation A is again rather “free” compared with both the Greek homily,
and the Greek and Armenian Bible. Probings into the existing Greek manu-
script material14 do not so far seem very promising for a discovery of a precise
Vorlage for translation A, so I shall here – cautiously – confine myself to quot-
ing a few examples that might be taken to illustrate translator A’s affinity to
a certain Armenian tradition.
In Mt. 18,17 the ἔσται of the Greek (B and Zohrab: եղիցի) is rendered in
A: համարեալ լիցի. This reading is shared with the Armenian translation of
Chrysostom and Aphraates (as well as the Syriac text of Aphraates) and ac-
cording to Lyonnet has its origin in the vetus syra.15
In Mt. 22,29 the two verbs of the Biblical phrase, πλανᾶσθε, μὴ εἰδότες (B:
մոլորիք ոչ գիտելով; Zohrab: մոլորեալք ոչ գիտէք) in A are connected with a
եւ: մոլորեալ էք եւ ոչ գիտէք. This reading again is shared with the Armenian
Aphraates, but our translation does not share the further Syrian “taint” in
Aphraates16: յոյժ մոլորեալ էք դուք, եւ …
In Mt. 23,37 the ποσάκις of the Greek (B and Zohrab: քանիցս անգամ) is
13 Cf. esp. my discussion of homily II, which has a Latin parallel, Per Piscatores, p. 45-102 (esp. p.
78-102 for the gospel quotations of this homily).
14 Here I rely on information, mainly oral, from scholars acquainted with the Greek material such as
Profs. C. Datema and K.H. Uthemann, Drs. Holger Villadsen and Sever Voicu, whom I thank for
their assistance.
15 Cf. Per Piscatores, p. 326 and Lyonnet, op. cit. p. 232.
16 Cf. Per Piscatores, p. 327 and Lyonnet, op. cit. p. 51f.
rendered in A: բազում անգամ. This rendering is shared with the Armenian

translation of Eusebius of Emesa and John Chrysostom.17
Two Biblical quotations outside the gospels shall be commented upon very
briefly, Rom. 9,16 and Jer. 1,5.
The οὐ τοῦ θέλοντος, οὐδὲ τοῦ τρέχοντος of Rom. 9,16, which is very im-
portant for the central part of Severian’s theological and homiletic argument in
this text, in the Zohrab Bible is rendered as “heavily” as follows, in order, pre-
sumably, to bring out the full contents of the very close-knit Pauline formula:
ոչ ըստ այնմ ինչ է՝ որ կամիցին, եւ ոչ ըստ այնմ որ ընթանայցէ. Translator B,
however, chooses a more direct and short rendering (here, incidentally, using
relative clauses for the participles): ոչ որ կամին եւ ոչ որ ընթանայն.
Here translator A shares an element with Zohrab, viz. the preposition
ըստ, but otherwise chooses infinitives (governed by the ըստ) to translate the
participles of the Greek: ոչ ըստ կամել(ն) ուրուք (է) եւ ոչ ըստ ընթանալ.18
The quotation of Jer. 1,5 has indirectly been dealt with already, as we are
here in a verse using the preposition πρὸ twice. The πρὸ τοὺ με πλὰσαι σε
and πρὸ τοὺ ἐξελθεῖν … σε of the Greek homily19 are rendered in B very
directly: նախ քան զստեղծանելն զքեզ and յառաջ քան զելանելն զքեզ20. The
Zohrab Bible has two sentences introduced with մինչչեւ and the “periphrastic
pluperfect”: ստեղծեալ էիր and ելեալ էիր, respectively. Translation A shares
the մինչչեւ with Zohrab, but uses a participle with the subject in the gene-
tive: ստեղծեալ իմ քեզ and քո ելեալ, respectively.21 In this case we thus have
consistency between the three translators in the choice of verbs but a wide
scale of differences in choice of syntactic construction.
Only one specific feature about translation B shall be added. In the sec-
ond part of Mt. 23,37, where both the Greek NT and the Greek homily have
ἠθελήσατε, and Zohrab and translator A agree in using կամեցարուք, i.e. the
same verb as in the first part of the verse, translator B has a different verb:
ախորժեցէք22. Against the background of the usual word-by-word translation
and the “stereotypes” in choice of words in B, this is surprising. This to my

18 The ն and the է given in parentheses illustrate a difference between the two passages where Rom.
9,16 is quoted (l. 298f and 311f in the Akinian edition). For է cf. Zohrab’s rendering.
19 PG 59, 660. This reading corresponds to the LXX but for the position of σε (2°) which is in the LXX
placed before the infinitive.
20 Ed. Aucher, p. 709, l. 24-26.
21 Ed. Akinian, l. 343f.
mind focuses attention on the manuscript basis, as the difference would of

course find its easiest and most natural explanation if it could be demonstrated
to reflect a variant reading in the actual Greek manuscript used by translator
B. From the evidence available it seems difficult to point to a characteristic
feature of the translator’s “identity”, “school” or “tradition” as an explanation.
A look into the Armenian concordance showing us that ախորժեմ is never
used in the Armenian NT, 11 times in the Armenian OT does not help us
much further, and there is little of interest for our purpose to be gained from
the scriptural passages in question. Only, it could be noted that in Is. 1,19 the
two parallel clauses with θέλητε in the LXX are rendered ախորժեսջիք and
կամիցիք, respectively. It could therefore be asked whether a common “ideal
of variation” (not unknown in Armenian stylistics) should connect the transla-
tor of Isaiah and our translator B. For translator B this possibility should then
also be considered regarding his choice of նախ քան in one part of Jer. 1,5,
յառաջ քան in its parallel.
Without extending to the analysis of elements in these few Biblical read-
ings a weight of argument which it could not necessarily bear, it tends to
demonstrate the necessity of considering carefully the characteristics of the
translators, here – again – in particular their possible affinities to traditions
represented by certain translations of Biblical and Patristic texts.
As a provisional summary of our findings it might be permissible to state
that translator A shows some degree of relationship to a “vetus armena-group”
of translations, whereas the profile of translator B is so far mainly character-
ized through his close ties to the Greek text of the homily and some particular
features in stylistic and vocabulary belonging to a Hellenizing “school”, even
if not at the most excessive stage.
2.4. Arrangement of the text

The paraphrastic character of translation A – or its foundation in a Greek
Vorlage different from that of the Migne text – accounts for a number of varia-
tions where passages are of different length and style, so as to defy direct
comparison. I have touched upon a number of such occasions in Per Piscato
res.23 It makes little sense to go into any detail in the analysis of these instances
before the Armenian and Greek manuscript basis has been examined more
thoroughly. Here, of course, space also forbids any detailed argument.
23 Cf. esp. Per Piscatores, p. 321.

However, the two greater deviations between A and B pointed out already,24
should be given a brief comment, as they are both of consequence for the dis-
cussion of how to find the way back to the “original” text.
One is the difference in the arrangement of the final part of the text, viz.
the passage PG 59, col. 662, l. 66 – col. 664, l. 26 (or col. 662, l. 74 – col. 664,
l. 13) as compared with the Armenian translations. The interesting fact is here
that the two Armenian translations agree in their arrangement. I have argued
that the Armenian arrangement reflects the original order, because the line of
argument is more natural – or “logic” – here than in the Greek. If the passage
is divided into four sections, the order in the Greek is A-C-B-D.25 Whether or
not I am right about the answer to the question of what reflects the original
text, it seems obvious that the observation – including the agreement between
the two Armenian translations – should be taken into account in a description
of the text history of the homily, presumably in its Greek form, already.
I take the same stand, as far as the introductory part of the homily is con-
cerned. Here, as so often, translation A stands alone over against the Greek and
translation B, in leaving out the whole section about the woman of Canaan.26
There is nothing to combine this section with the rest of the homily, and so,
to say the least, the possibility of considering the “shape”, arrangement and
length of translation A as reflecting the original should at any rate be taken
seriously.
Therefore, if the forthcoming Greek edition of Severian’s homilies, as seems
to be the intention, will be arranged according to a “liturgical” disposition,
placing this homily under the Sunday where the pericope about the woman of
Canaan was read as text for the day’s sermon,27 this is understandable against
the background of the lack of evidence for placing Severian’s homilies in a
fixed and reliable absolute chronology of years.
However, if such a disposition is chosen it should be pointed out very
clearly that the bulk of the homily may not originally have had any connec-
tion at all with the Sunday in question, and that no conclusions for an abso-
lute chronology of Severian’s homilies could be drawn on the basis of links
24 Cf. Per Piscatores, p. 320f.

25 For details, cf. ibid.
26 Cf. Attribution, p. 125.
27 Cf. C. Datema, “Towards a Critical Edition of the Greek Homilies of Severian of Gabala”, Orientalia
Lovaniensia Periodica 19, 1988, 107-115, see esp. 111f.
between the introductory paragraph of the Greek version of this homily as

known today and other homilies by the author.
3. The author, the translators and other hands

The above observations have taught us little about Severian of Gabala and little
about Greek or Armenian manuscripts, at any rate as far as new information
is concerned.
In other words, they have not been very helpful in identifying or describing
such hands as those of copyists and editors of Greek and Armenian texts, who
have laid their hands on the text(s) in question through the close to 1600 years
that have elapsed since the homily was spoken viva voce to a congregation in
Constantinople or Gabala (or elsewhere?).
Hopefully, however, the observations have sharpened our attention on the
necessity and possibilities of describing the “hands” of the two very different
Armenian translators. In many ways the possibilities are restricted; the neces-
sity, however, in my judgment, does not only concern investigators interested
in that particular field of the history of culture made out by the remarkable
achievements of the early generations and centuries of Armenian translators,
but also – under the circumstances given – to investigators and editors of
Severian’s héritage littéraire in its original linguistic clothing, and Greek patristic
literature in general.
A closer identification and description of what belongs to the “profiles” of
the translators and their particular “milieu”, context and tradition may nar-
row down the field of direct reasoning from a reading in the Armenian to its
Greek Vorlage. On the other hand the identification of Armenian “particulars”
in the text tradition, and the isolation of such elements may add to the value
and weight of the “rest” of the Armenian translations in the procedures of
identifying Severian’s original address to his congregation.
he Noble Art of Abbreviating
in the Light of some Texts attributed
to Severian of Gabala
T
1. Preliminary remarks
Originally, this paper was intended to have as its modest subtitle the words:
“Some simple remarks”, for, what will be presented, will not be anything like
either a broad or a thorough analysis of rhetoric practice or principle in An-
tiquity in abbreviating procedures. What I shall try to do is – in a very simple,
straightforward and elementary way – to look at a few instances where texts
attributed to Severian of Gabala appear in more than one version, usually in
what may be considered a “full-length” version and some kind of abbrevi-
ated form.
On my way through the examples I considered more than once the pos-
sibility of transferring the adjective “simple” from the subtitle to the main
title, having ascertained that the procedure of abbreviating very often is quite
a simple one. So the art of abbreviating is, maybe, not always a noble one; on
the other hand, the abbreviated form of a text often has a particular strength
of expression, entirely its own, so I left the notion of “noble” in the title; but I
certainly ask and warn my readers to retain the notion of “simple” as a subtitle
for what follows.
2. Introductory remarks on the homily In natale

domini nostri Iesu Christi (CPG 4657)
First it would be natural to give a hint about the main reason(s) that led me
to go into the matter of abbreviated forms of texts by Severian. My point of
departure was the discussion about a certain homily, viz. the Pseudo-Chrys-
ostomic homily In natale domini nostri Iesu Christi (CPG 4657). The external
evidence for this homily, in my view, points rather strongly in the direction
of Severianic authorship, but internal arguments on the basis of the Greek text
(as published in Migne) have led modern scholars to deny that the homily in
question could be by Severian.
Let me elaborate a little further on the state of the matter.
For this homily, in fact, a number of potential authors have been proposed,
in ancient as well as modern times. A mere glance in M. Geerard’s Clavis1 tells
us that there is a Coptic version having been attributed by W. E. Crum (on
insufficient grounds, it seems) to St. Basil the Great,2 and there is a Georgian
version under the name of Epiphanius. On Syrian ground Severus of Antioch
considers the homily to be by John Chrysostom, as does most of the Greek
manuscript tradition.
One Oriental line of evidence was not known when Geerard published the
Clavis, i.e. the Armenian. In an article published in 19823 I have tried to elucidate
the Armenian transmission, as far as it is known to me. This includes a version
of the homily in two Armenian manuscripts of the library of the Jerusalem
Patriarchate (and two later manuscripts in the Mechitharist library in Venice) at-
tributing the homily to “Sewerianos, the Priest”,4 and two quotations in a series
of fragments attributed in MS Galata 54 to “Seberianos, Bishop of Gabala”.5 It
should be noted that in this series of quotations there are a number of fragments
belonging to texts which are generally held to be by Severian of Gabala.6
The Armenian transmission therefore fits very well with the first fact as-
certained in M. Geerard’s primary “nota”: Tamquam Severiani Gabalensis citatur
a Theodoreto.7
To me it is difficult not to consider such an early – and often reliable – Greek
source as Theodoret of Cyrrhus and the occurrence in the Galata series as
weighty external arguments for Severian’s authorship. Maybe less weighty, but
still of importance is the attribution to “Severian the Priest” in the Jerusalem
manuscripts.
However, modern authors such as B. Marx, J. Zellinger, H.-D. Altendorf
1 Clavis Patrum Graecorum II, cura et studio Mauritii Geerard, (CC), Turnhout, 1974.
2 W E. Crum, “Theological Texts from Coptic Papyri”, Anecdota Oxoniensia, Semitic Series 12, Oxford
1913, 18-20.
3 Henning J. Lehmann, “Severian of Gabala: New Identifications of Texts in Armenian Translation”,
Classical Armenian Culture, University of Pennsylvania Armenian Texts and Studies 4, ed. Th. J. Samu-
elian, University of Pennsylvania 1982, 113-124 (quoted: “New Identifications”) [above, pp. 67ff].
4 “New Identifications”, 114 [above, p. 68f].
5 Ibid., 117 [above, p. 71].
6 Cf. below and “New Identifications”, 118f [above, pp. 72f].
7 Op. cit., 587.
T h e N o bl e A r t o f Abb r e v i a t i n g 157
and Sever J. Voicu8 reject the attribution to Severian, and scholars responsible
for current plans of a new edition of the Greek Severian – such as C. Datema
and K.-H. Uthemann – seem to take the same stand.9
These sceptics mainly refer to one reason for their position, viz. that the
stylistic and linguistic characteristics of Severian are absent from this homily.
To quote Zellinger, after referring to the quotation in Theodoret, he comments
as follows: “W. Dürks10 ist es gelungen, die Rede aufzufinden, der das Stück
entnommen. Es ist das (sic!) die kurze, nur drei Mignespalten füllende, pseudo-
chrysostomische Homilie: Εἰς τήν γέννησιν … Dürks glaubt die Rede auf
Grund des Theodoretzitates Severian zurückerstatten zu sollen (De Severiano
Gabalitano 46-48; vgl. auch ebd. 19.66). Ich vermag aus sprachlichen Gründen
nicht zuzustimmen”.11 That is all Zellinger has to say about this homily, and
Marx and Altendorf only add little. Voicu regards this homily as attributable
to a certain anonymous Cappadocian author of the early 5th century. His
arguments are mainly stylistic, particularly linkages within a collection of 33
(or 30-odd) Pseudo-Chrysostomic homilies belonging, according to Voicu, to
the same author.12
At a first glance the short Greek text seems to contain few of the stylistic
elements usually referred to as characteristic of Severian, to which could be
added that a desideratum (which may be on its way to be fulfilled by those
responsible for the new Greek Severian edition) is a thorough stylistic analysis
of Severian’s homiletics. To some extent earlier research has left us with rather
superficial descriptions.
In my view, there is reason in general to question the validity of the verdict
“spurious” based on the absence in a certain text of locutions, otherwise often
used by an author. Methodologically, the simple fact that a homily is very short
makes a judgment of this kind all the more doubtful. Dealing with Severian’s
natale-homily, in my article quoted above I therefore wrote: “Great caution is
required before conclusions about authorship are drawn from general obser-
8 For references cf. below and “New Identifications”, 121, note 6 [above, p. 68].
9 I here rely on oral information and communications at the Oxford Patristic Conference in 1987.
10 G. Dürks, De Severiano Gabalitano, Kiel 1917.
11 Johannes Zellinger, Studien zu Severian von Gabala, Münsterische Beiträge zur Theologie 8, Münster i.W.
1926, 36, n. 1.
12 Sever J. Voicu, “Trentatre omelie pseudocrisostomiche e il loro autore”, Lexicon Philosophicum 2,
1986, 73-141, see esp. 99-101; S. J. Voicu, “Note sull’omelia pseudocrisostomica In natale Domini
nostri Iesu Christi (CPG 4567 (sic!))”, Memorial Dom Jean Gribomont (1920-1986), Studia Ephemerides
“Augustinianum” 1988, 621-626.
vations on style and language, especially where short homilies are concerned
which may very likely have been given their form by excerptors”.13
I refer, of course, to the fact that in the process of abbreviation – as also in
the process of translation – stylistic particulars will be liable to be smoothed
out or even disappear.
In the following, as mentioned already, it is my modest purpose to make
some simple and straightforward remarks about observations made on a few
texts by Severian or attributed to him which have undergone abbreviating or
excerpting processes.
3. Examples of abbreviated texts

3.1.
My first example is the homily on Jesus’ entry into Jerusalem (CPG 4246), so
far only known in Armenian. The long version of this homily is the last of the
Akinian collection – No. XIII, published in 1959,14 whereas the short version
was published more than 130 years earlier, viz. as homily No. XI in the volume
of 15 homilies edited by J. B. Aucher.15
It might be appropriate once again to quote Zellinger. His verdict on the
last five homilies of the Aucher collection runs as follows: “Für die Echtheit
der letzten fünf Reden liegt weder äussere noch innere Bezeugung vor, und
man muss restlos W. Dürks beistimmen, der aus sachlichen und stilistischen
Erwägungen den Namen Severians aus deren Titel streichen möchte. Zum
wenigsten ist über ein Non liquet nicht hinauszukommen”.16
In my book Per Piscatores17 it is shown that with the long version in our
hands we are in a situation fundamentally different from that of Zellinger’s
time. We can point to the external argument of the occurrence of the homily
among the group of five in the Akinian collection (homilies IX-XIII), and inter-
nally to a great number of correspondences to well-known Severian homilies,
as arguments for Severian’s paternity.
13 “New Identifications”, 115 and 121, note 6 [above, pp. 68 and 70].
14 N. Akinian, ed., “Die Reden des Bischofs Eusebius von Emesa”, Handes Amsorya 1956-1959. For
homily No. XIII see HA 1959, 321-360.
15 Severiani sive Seberiani Gabalorum episcopi Emesensis homiliae, ed. J. B. Aucher, Venice 1817. For hom-
ily No. XI see pp. 402-409.
16 Zellinger, op. cit., 71, with reference to Dürks, op. cit., 60-64.
17 Henning J. Lehmann, Per Piscatores. Studies in the Armenian version of a collection of homilies by Euse-
bius of Emesa and Severian of Gabala, Aarhus 1975 (quoted: Per Piscatores), 335-367.
The relation between the two versions is shown in Table 1.
Table 1
De adventu domini super pullum (CPG 4246) (Akinian XIII/Aucher XI)
Akinian XIII (line) Aucher XI (page, line) Number of lines (Akinian)

(a) 1-14 – 14
(b) 15-16 402,1-2 2
(c) 16-29 – 13
(d) 30-73 402,2 – 404,24 44
(e) 74-95 – 22
(f) 96-115 404,25 – 406,8 20
(g) 116-145 – 30
(h) 146-165 406,8-32 20
(i) 166-214 – 49
(j) 215-227 406,33 – 408,8 13
(k) 228-715 – 488
It appears that the short version covers less than one seventh of the long ver-
sion (99 out of 715 lines). It further appears that apart from the first two lines
the short version consists of four “blocks” representing about one half of the
first 200 lines of the homily in its long version.
A closer scrutiny of the relationship between the two texts, of course,
reveals a number of variants, most of them manifestly inner-Armenian, and
for the great majority of rather a trivial character: omission of “article”, եւ,
զ- etc. Choice of different words in some cases could be a copyist’s mistake,
in other cases there seems to be a choice between synonyms, which, of course,
would point to either different translators or the copyist’s change to a word
more familiar to him. As the texts for the greater part follow each other very
closely, I would plead for the latter possibility.
In only one or two cases one could ask the question whether the reason for
a variant could be that a Severianic “Zwischensatz”18 or characteristic idiom
had been omitted in the short version.19
18 Zelllinger’s expression, op. cit., e.g. 33.

19 I am thinking of the omission of the question in Akinian XIII, 47 (cf. Aucher, 402,20). Correspond-
ingly, the “superfluous” ոչ եթէ … այլ in Akinian XIII, 221 is left out in the short version (cf.
The comparison therefore shows that only a modest amount of stylistic

“smoothing out” or editorial manipulation for linguistic or other reasons seems
to have taken place in the passages used by the abbreviator.
If it is true to say that there is little in the Aucher-version to substantiate
Severian’s paternity to the text, so that Zellinger’s and later scholars’ judgment
is understandable (and I would certainly admit this, if only stylistic criteria,
traditionally applied to the text, are taken into consideration), it must there-
fore be ascertained that this is due mainly to the very choice of passages and
arrangement of the text, not to a very active effort to change stylistic details.
Furthermore it should be noted that a great part of what is left out is
concerned with a discussion of biblical material, first and foremost the whole
context of Psalm 8,3 quoted in Mt. 21,16 as Jesus’ answer to the indignation
of the chief priests and doctors of the law (Mt. 21,15) because of the shouting
of Hosanna to Jesus (and other elements of the narrative of Jesus’ entry into
Jerusalem).20 Also a great number of other biblical texts are commented upon
in the long version, among them a number of Severian’s favourite sources of
“testimonials” such as e.g. Joel 2,19 and 2 Sam. 24,10-17.21
As mentioned already, the amount of deliberate “smoothing” and alteration
is obviously very modest; it therefore seems to me justifiable to call the abbrevi-
ating method used a “subtraction technique”, i.e. a technique where a number
of “blocks” of the text are left out whereas other passages are preserved; and
as such a great amount of the biblical testimonial material so characteristic of
Severian is left out, not only the modern scholar using stylistic criteria but also
the investigator who would use the biblical material to solve the question of
authorship is left in a very difficult position by the abbreviator.
3.2.
Technically speaking my second example will have strong resemblances to
the first. In other respects it will differ.
I am referring to homily No. III in the Akinian collection for which again
a shorter version is found in Aucher’s Severian volume as homily No. I.
Aucher, 408,1), and it could be noted that the word անքնին (Akinian XIII, 154) which might be
called a favourite word of Severian’s is not to be found in the short version (cf. Aucher, 406,19).
20 Cf. Per Piscatores, 348ff. On the discussion of the very word “hosanna” in Severian (and other au-
thors), see my: “Hosanna – A Philological Discussion in the Old Church”, Armeniaca – Mélanges
d’études arméniennes, Venice 1969, 165-174 [above, pp. 13ff].
21 Cf. Per Piscatores, 357ff.
The relation between the two versions is shown in Table 2.
Table 2
De passione (CPG 3531,5) (Akinian III/Aucher I)22
Akinian III (line) Aucher I (page, line) Number of lines (Akinian)

(a) 1-116 2,1 – 10,12 116
(b) 117-157 – 41
(c) 158-177 10,12-34 20
(d) 178-189 – 12
(e) 190-197 10,34 – 12,7 8
(f) 198-207 – 10
(g) 208-246 12,7 – 14,12 39
(h) 247-278 – 32
(i) 279-290 14,12-25 12
(j) 291-332 – 42
(k) 333-339 14,26 – 16,1 7
(l) 340-519 – 180
Here, the abbreviated version is made up of six “blocks” of varying length

taken from the first two thirds of the homily.
Also, as far as this homily is concerned, the amount of variations in detail
is fairly small (see note 22), and again the passages left out by the abbreviator,
to a great extent consist of biblical material and exegetic discussion.
Technically speaking, therefore, the method of abbreviation largely seems
to be the same as that used in our first example, a “subtraction technique”.
On this background, therefore, it might be characterized as a surprising fact
that a number of circumstances, and not least, the history of research present
quite a different picture of this text.
First of all it should be noted that we are here concerned with a text which
22 The following differences of detail should be noted. In (a) the indication of the contents of the hom-
ily is shorter in Aucher I (l. 6-8) as compared with Akinian III (l. 6-11). Correspondingly, l. 17ff in
the long version are richer than the “parallel”, Aucher, 4,1ff. In (c) about 1½ lines are left out (Akin-
ian, 165f), maybe through homoioarkon. Aucher, 16,2-5 contains a final doxology which is evidently
secondary.
is not by Severian, and our best introduction to the history of research might
be a consideration of the reasons that nevertheless led Zellinger to accept the
short version as being by Severian.
Zellinger voiced a certain, modest amount of reluctance or uneasiness
about the attribution to Severian, saying: “Dabei wird sich … ein letztes klären-
des Wort kaum sprechen lassen”.23 On the other hand, he gives his approval
of Martin Jugie’s statement about the homily: “Son authenticité est hors de
doute … “,24 and even if it might be strange for Severian, that “Die Rede,
bescheiden an Umfang, wurde zu Jerusalem gehalten”,25 Zellinger finds the
explanation of what is unusual in the fact that “Was nun folgt, ist freilich
keine Rede im eigentlichen Sinne des Wortes, sondern eine des rhetorischen
Charakters stark entkleidete symbolartige Expositio fidei, die mit dem trini-
tarischen Teile des Athanasianums überraschende Ähnlichkeit aufweist”.26
To this statement Zellinger adds a quotation from Jugie, giving this de-
scription of the text: “C’est, peut-on dire, un resumé en phrases lapidaires de
la théologie de Dieu un et trine”.27 Zellinger even suggests that the expositio-
character of the text might be sufficient explanation of the fact that the words
ἀγέννητος and γεννητός are used, “die der Bischof sonst nach Inhalt und
Form in schroffster Weise ablehnt und deren wissenschaftliche Diskreditierung
er sich zum Ziel gesetzt”.28
Today we know that the expressions mentioned are there because they
belong to the central theological vocabulary of Eusebius of Emesa who is the
author of the text. The reason why I have chosen to consider some wrong as-
sumptions during the history of research at some length is that it seems to me
highly interesting that the abbreviating technique, even if technically speaking,
it is to a large degree of the same subtraction character as that of our previous
example, has left us with a text about which serious scholars have been led
astray by elements in the history of transmission as far as the identification of
the author is concerned and then argued for their wrong assumptions on the
23 Zellinger, op. cit., 72.

24 Zellinger, op. cit., 74, with reference (in note 6) to Martin Jugie, ˝Sévérien de Gabala et le symbole
Athanasien”, Echos d’Orient XIV, Paris 1911, 193-204.
26 Zellinger, op. cit., 75f.
28 Zellinger. op. cit., 78.
basis of considerations of “genre”, so to speak: expositio fidei, resumé en phrases

lapidaires, a “Traktat … auf Prägnanz und knappe Fassung … abgestimmt”29
Today, it is presumably easy to see that it is partly a matter of abbreviating
technique, and partly a matter of style and theology of an author concerned
with problems of an age about two generations earlier than Severian that
determine the character of the short version of the homily.
But when we consider in how many cases we are still bound to the ar-
guments of Zellinger and his contemporaries about style and language, the
importance of our attempt to gain an insight in the art and procedures of ab-
breviation in the patristic period seems to gain in strength from our second
example.
3.3.
My third example may be said to take us outside the field of abbreviating if
the notion of abbreviation can only be used in cases where a single homily
(or a text of another genre) is found in a short and a long version. In any case,
the phenomenon which I would call “the construction of a florilege homily”
seems to me to be of interest for our evaluation of what has happened in the
course of the history of transmission of Severian’s homilies.
I am thinking of the two series of fragments of Severianic texts in MS Galata
54. I have dealt with these series elsewhere,30 so I shall here only recapitulate
that one of the series consists of 11 quotations taken from six different homilies,
the titles of which are referred to very carefully in the “florilege”.31 The second
series contains 20 quotations taken from 8 of the homilies of the Aucher col-
lection (homilies I-VII and IX).32
What is of particular interest in this connection is that each of the series in
itself constitutes a unity of its own, an expositio as it were of a central theologi-
cal topic, a “florilege homily” to use the expression coined above, and thus to
my mind, they represent a particular, very selective, method of abbreviating
a number of texts in order to present certain subjects and themes of general
interest, in the MS Galata 54 series from one author, Severian, in a number
29 Ibid.
30 “New Identifications” and Henning J. Lehmann, “Severian of Gabala: Fragments of the Aucher Col-
lection in Galata MS 54”, Armenian Studies/Etudes arméniennes in memoriam Haïg Berbérian, ed. Dick-
ran Kouymjian, Lisbon 1986, 477-487 (quoted: “Aucher Fragments”) [above, pp. 95ff].
31 “New Identifications”, 117f [above, p. 73f].
32 “Aucher Fragments”, 481-483 [above, p. 98ff].
of cases – in real florilegia, as is well-known – core quotations about the same

subject from a number of different authors.
I very much doubt that such theological resumés or expositiones would eas-
ily have been identified by modern researchers as belonging to Severian, had
it not been for two facts: 1) the careful citation of the author’s name and the
title of the homily in question, and 2) the fact that some of the homilies quoted
belong to the least disputed core of genuine Severianic homilies.
It might be added that such chains of quotations might reflect the fact
that a number of homilies had been connected in a particular collection, as is
evidently the case as far as the Aucher homilies are concerned.33
I shall not go into a further description of MS Galata 54 or discussion of
the shaping of florilegia, which in many respects is a different matter from the
abbreviation technique met with in our two first examples, even if there is a
certain amount of correspondence in the search for theological “formulae”
and core passages.
Before returning to one of the homilies quoted in the first-mentioned Galata-
series, viz. the natale-homily, it might be of interest, however, to give a brief
survey of the contents of the two “florilege homilies”. This is done in Tables 3
and 4.
33 See the discussion of the relation between the Aucher collection and the quotations in the “Seal of
Faith” and about the history of the collection with or without the homily: Aucher No. I, “Aucher
Fragments”, 484f [above, p. 102ff].
Table 3
MS Galata 54: Severian Cento I (Aucher): Սեբերիանոսի Եմեսու
Homily No. CPG Aucher MS.Galata 54 page/col., line Aucher, page, line
(a) 3531,3 I 363a,21-28 4,1-5
(b) 363a,28-b,22 14,26 – 16,1
(c) 4240 II 363b,25 – 364,15 18,24 – 20,12
(d) 364,15-32 24,7-27
(e) 364,32 – 366,2 34,7 – 36,10
(f) 366,3-29 36,37 – 38,31
(g) 4241 III 366,32 – 367,6 78,5-14
(h) 4242 IV 367,8-20 160,16-33
(i) 367,21 – 368,8 162,32 – 164,21
(j) 4243 V 368,10 – 369,2 202,15 – 204,8
(k) 4244 VI 369,3-18 216,32 – 218,16
(1) 369,18-25 228,22-30
(m) 369,25 – 370,16 236,21 – 238,12
(n) 4198 VII 370,17-24 258,13-21
(o) 4215 IX 370,26 – 372,9 336,14 – 338,33
(p) 372,9-32 340,6-32
(q) 372,33 – 373,28 342,26 – 344,20
(r) 373,29 – 374,11 350,7-24
(s) 374,11 – 375,5 356,3 – 358,3
(t) 375,6-13 362,22-32

Table 4
MS Galata 54: Severian Cento II: Երանելւոյ Սեբերիանոսի Գաբա …34
Homily No. CPG MS Galata 54, page/col., line

(a) 4249 375,15 – 376,18
(b) 4657 376,19-25
(c) – 376,25 – 377,7
(d) 4699 377,9-19
(e) 4201 377,21-24
(f) – 377,24-33
(g) – 377,33 – 379,8
(h) – 379,9-16
(i) 4196 379,18-23
(j) 4295,17a 379,25 – 381,21
(k) – 381,21 – 384,6
4. The natale homily reconsidered

The correspondences and differences between the Greek text as published in
Migne and the Armenian text to be found in MS Jerusalem arm. 1 is shown
in Table 5.35
34 For manuscripts and editions see “New Identifications”, 117f [above, pp. 73f]. The fragment quoted
by Severus of Antioch referred to under CPG 4295,17a (cf. (j)) only covers part of the first quotation
of this homily in the Galata MS, which therefore is the only source – so far – for a hitherto unknown
part of this homily.
35 Cf. “New Identifications”, 114 (where, regrettably, there are a couple of errors in the figures given)
[corrected above, p. 69]
Table 5
In natale domini nostri Iesu Christi (CPG 4657)
PG 61 Number of lines MS Jer. arm. 1 Number of lines

col., line (PG) col., line (MS).
(a) 763,1 – 765,3 69 41a,48 – 42a,35 84
(b) – – 42a,36 – 42b,19 22
(c) 765,4-26 23 42b,20 – 43a,2 31
(d) – – 43a,3-45 43
(e) 765,27-44 18 43a,46 – 43b,16 19
(f) – – 43b,17 – 44b,3 83
(g) 765,45 – 768,12 141 – -
Expressed in the terminology chosen in this paper, three “blocks” of the text
as known in Greek (corresponding to less than half of the text – 110 out of
251 lines in Migne’s edition) find their parallels in the Armenian (84 + 31 + 19
= 134 lines in the manuscript) which has three further “blocks” (22 + 43 + 83
= 148 lines in the manuscript).
Technically speaking the “abbreviation method” now looks familiar. Only,
we have here two texts which might both be abbreviations; and it is of course
difficult to give any opinion of how long was the original unabbreviated hom-
ily.
Above I expressed my uneasiness about the rejection of the external evi-
dence for Severian’s original author’s rights to this homily. I further hope to
have shown what problems and consequences arise from the process of ab-
breviating, even in its least radical form, as far as the editorial accomplishment
is concerned, and hence find new reasons for an exhortation to be cautious in
verdicts of inauthenticity about short homilies.
As the Armenian version of this homily has never been published, I shall
finally give a few specimens of the contents of the Armenian “blocks” which
are not covered by the Greek.
In section b the comments about the virgin birth from the preceding para-
graph are continued, first with a parallel between Jesus’ birth and the way
in which Eve was “born” out of Adam (Gen. 2,21f). In order to obviate the
lack of understanding of the Jews there is then a reference to the narrative in
Exodus 17 of how Moses smites the dry rock with his dry rod making water
come out of the rock. Finally, reference is made to the narrative of the prophet
Habakkuk being carried by an angel to Babylon to feed Daniel in the lion’s

den (Dan. 14,34ff (apocryphal addition)).
All three references to Old Testament “testimonia” can be found in genuine
texts by Severian. The miraculous birth of Eve (Gen. 2,21f) as a foreboding of
the virgin birth is commented upon with a number of parallels in Genesis-
homily No. 5 (CPG 4194,5)36 and in Quomodo animam acceperit Adamus (CPG
4195).37
The reference to Moses’ dry rod miraculously making water come out of
the dry rock as an image of the virgin birth is known from e.g. In pretiosam et
vivificam crucem (CPG 4213).38
Finally, the wonder of Habakkuk being brought to Babylon is used by
Severian in more than one context.39
After section c about Jesus’ birth in Bethlehem the first part of section d
presents Hab. 3,3 as a testimony that the birth should take place in Bethlehem
in Judea. A little further on also Hab. 3,2 LXX is quoted and commented upon,
and after quotations of Ps. 71,6 and elements from Ps. 49 (vss. 2.3.5) attention
is again – in section e – turned to the historical reports of Jesus’ birth.
The elements in section d taken from the prayer of Habakkuk have very
close parallels in a genuine Severianic homily, viz. again: In pretiosam et vivi-
ficam crucem (CPG 4213).
Thus the comments that Teman of Hab. 3,3a (Θαιμαν) (which can be trans-
lated “South”, it is maintained) is Bethlehem, and that the shadowy mountain
(Mount Taran) of 3,3b is Zion are found in practically the same wording in
Combefis, 255f, including the explanation that Mount Zion is called shadowy
because it is overshadowed by the strength of God. The verb ἐπισκιάζω is
that used about the virgin in the Annunciation narrative.40
Also the explanation that the two creatures of Hab. 3,2 should be under-
stood about life in this aeon and life in the aeon to come or about the Old and
36 PG 56,482f.
37 Savile, 5,650ff. Cf. Johannes Zellinger, Die Genesishomilien des Bischofs Severian von Gabala, Alttesta-
mentliche Abhandlungen VII,I, Münster i.W. 1916, 40-46.96ff.
38 Combefis, 225.
39 Cf. Zellinger, Studien, 95. References are to the following homilies: CPG 4194,5, 4196, 4213, 4244.
40 This topic is dealt with twice in CPG 4213, apart from Combefis, 255f, also 274f. In the latter pas-
sage, the shadowy mountain is considered to refer to the Virgin Mary, and it is added that τινες
(“some”) take it to mean Zion. Such an inconsistency is not unseen in a text by Severian. In any case
the problem of inconsistency is an internal one for CPG 4213, and of course does not detract from
the value of the parallel between the first quotation here and that in CPG 4657, arm.
the New Testament is shared with the homily In pretiosam et vivificam crucem
(CPG 4213).41 Here even minor details of the two texts are identical.
Now, whereas the last half of the Greek (section g) is a fulfilment of the
promise in the title of the Armenian, viz. that the homily shall be concerned
also with the wise men’s adoration (Mt. 2,1ff), the last part of the Armenian
(section f) is mainly concerned with the trial, apology and death of Stephen
(Acts 6-7).
These chapters have not been commented upon by Severian elsewhere
to such an extent as to give material for comparison, so on the basis of the
criterion for internal argument, which I have mainly used, viz. the use of the
Bible, there is little to be gained. I shall here abstain from going into stylistic
or other details.
By way of conclusion to this paper I find it permissible to say, that the
Armenian version of the natale-homily has furnished us with valuable com-
parative material for reaching a positive conclusion from internal evidence
corresponding to that suggested by strong external arguments, viz. that the
homily is by Severian. It is my hope that our way through some elementary ob-
servations on the process, which some texts attributed to Severian underwent
in order to find an abbreviated form, has been of some value for deepening
our attentiveness to what would happen in such a process, thereby assisting
us on our difficult way “back to the original text”.
41 Combefis, 272f.
he So-Called “Absurd” Punctuation
in John 1,3-4
T
Neglected Witnesses of the Old Church
Eberhard Nestle, the famous New Testament editor and scholar, begins his
article “Zur Interpunktion von Joh 1,3.4”, published in 1909, with the fol-
lowing words: “Eine monographische Untersuchung dieser Frage und im
Zusammenhang damit die nach dem richtigen Text dieser Verse wäre eine
sehr lehrreiche Aufgabe”, and his concluding remarks run as follows: “Ein
ganz interessantes Stück Geschichte der Exegese hängt da an einem einzigen
Komma oder Punktum”.1
Insofar as Ed. L. Miller’s book Salvation-History in the Prologue of John,
published in 19892, can be considered the first full monograph concerned
with the exegetic and theological consequenses of the various possibilities of
punctuation in vss. 3-4 of the Johannine prologue, it can be maintained that
80 years passed before Nestle’s wish was fulfilled.
However, the questions of punctuation have, naturally, been dealt with
quite often in the abundant literature of the eight decades between Nestle and
Miller, dealing with the Gospel of John and, in particular, its prologue. We
shall make no attempt here to summarize the exegetic discussion;3 it should be
emphasized, however, that as a background and counterpart of our remarks
below on some particular questions related to the history of exegesis – and
the history of the text – a number of studies which appeared in the 1950s and
1960s are of prime importance. I am thinking of contributions by such scholars
1 Eberhard Nestle, “Zur Interpunktion von Joh 1,3.4”, Zeitschrift fur die neutestamentliche Wissenschaft
und die Kunde der älteren Kirche 10, 1909, 262-264.
2 Ed. L. Miller, Salvation-History in the Prologue of John. The Significance of John 1:3/4, Supplements to
Novum Testamentum LX, 1989.
3 For a general bibliographic overview reference can be made to Miller, op. cit., 2f, note 3, and 110ff
(Bibliography).
as K. Aland, J. Gennaro, P. Lamarche, H. Langkammer, I. de la Potterie, and

J. Mehlmann.4
For the field of text history and text criticism Aland’s discussion may well
be called “mandatory”, as does Miller.5
From a purely technical point of view scholars have not – taken as a whole
– been too helpful to readers, as the numbering of the readings in question has
been overthrown a couple of times. To our purpose it is useful first to list the
readings in I. de la Potterie’s numbering (which, in fact, goes back to Theodor
Zahn6):
I. ἐγένετο οὐδὲ ἓν (οὐδὲν), ὃ γέγονεν. Ἐν αὐτῷ ζωὴ ἦν.

II. ἐγένετο οὐδὲ ἓν, ὃ γέγονεν ἐν αὐτῷ. Ζωὴ ἦν.
III. ἐγένετο οὐδὲ ἓν. Ὃ γέγονεν, ἐν αὐτῷ ζωὴ ἦν.
IV. ἐγένετο οὐδὲ ἓν. Ὃ γέγονεν ἐν αὐτῷ, ζωὴ ἦν.7
The main topic of discussion has been the choice in Bible manuscripts, edi-
tions and commentaries between what is often called “the old reading”, cor-
responding to nos. III and/or IV in the above survey, and what is called “the
new reading”, that of no. I. Corresponding to this historical sequence Aland
chooses the designations A1 and A2 for III and IV, resp., and B for I, while
Miller simplifies correspondingly, choosing however the designations “Read-
ing I” and “Reading II”, corresponding to no. I and no. III/IV, respectively.
4 K. Aland, “Eine Untersuchung zu Joh. 1:3-4: Uber die Bedeutung eines Punktes”, ZNW 59, 1968,
174ff. The article is reprinted with minor updatings in Kurt Aland, Neutestamentliche Entwürfe,
München 1979, 351-391. References below will be to the 1979-version. J. Gennaro, Exegetica in pro-
logum sec. maximos ecclesiae doctores antiquitatis christianae, Rome 1952. P. Lamarche, “Le Prologue
de Jean”, Recherches de science religieuse 52, Paris 1964, 497-537. H. Langkammer, “Die Zugehörig-
keit des Satzteiles ὃ γέγονεν in Joh 1,3.4 bei Hieronymus”, Biblische Zeitschrift, Neue Folge 8, 1964,
295-298. J. Mehlmann, “De mente S. Hieronymi circa divisionem versuum Jo 1,3s”, Verbum Domini
33, 1955, 86-94. J. Mehlmann, “A Note on John 1:3 “, Expository Times 61, 1955-56, 340f. I. de la Pot-
terie, “De punctuatie en de exegese van Joh 1,3.4 in de traditie”, Bijdragen 16, 1955, 117-135. I. de
la Potterie, “De interpunctione et interpretatione versuum Joh. 1,3-4 “, Verbum Domini 33, 1955,
193-208.
5 Miller, op. cit., 12.
6 Das Evangelium des Johannes ausgelegt von Theodor Zahn, Kommentar zum Neuen Testament IV, quoted
here from the 3/4th edition 1912, Excurs I: Die Satzabteilung 1,3.4, 706-709.
7 In this survey I have combined Zahn and de la Potterie, insofar as the latter, in nos. I and II, leaves
out the comma before ὃ, and Zahn does not list the variant reading οὐδὲν, as de la Potterie does, in
I. In the following I mainly refer to the four possibilities under the numbers used by Zahn and de la
Potterie.
T he S o - C alled “ A bsurd ” P unctuation in J ohn 1 , 3 - 4 173
Sometimes Miller finds it necessary to distinguish between de la Potterie’s

readings ΙΙΙ and IV, then referring to a “Reading IIA” and a “Reading IIB”.
Lamarche chooses designations corresponding to those of Aland, adding a
“C” for de la Potterie’s reading no. II.
It may be helpful to the reader to summarize these numberings in a table
(Table 1).
Table 1
de la Potterie Lamarche Aland Miller

I B B I
II C
III A1 A1 IIB
IV A2 A2 IIA
As appears from this survey, reading no. II in de la Potterie’s list has attracted
less attention than the other three readings. This, no doubt, is due to the fact
that this reading was described by Zahn already as “offenbar absurd”, which
statement is quoted, consentingly it seems, by Aland.8
As opposed to Zahn and Aland, Miller seems to prefer a timbre of polite-
ness toward such authors in the history of the church as have advocated this
reading, quoting it “for the sake of completeness”, and adding the remark:
“But this punctuation barely occurs in the tradition and commands no claim
to authority”.9 In reality, thus, his verdict is identical with that of Zahn and
Aland. Lamarche does not call reading no. II absurd, but “improbable”: “Cette
coupe n’est pas impossible, mais improbable elle est rarement adoptée”. And
he has also noticed that “Cette hypothèse comporte encore une variante si en
plus de la ponctuation après ἐν αὐτῷ on ajoute un point après οὐδὲ ἔν”.10 This
observation to my mind is important, as it rightly suggests a pointed form of
Reading II.
In the light of the verdicts quoted and the discussions implied, what follows
below, will have a very modest scope. No attempt will be made to compete
with authors giving broad theological and text-historical outlines. What will
8 Op. cit., 357.

9 Op. cit., 14, note 36.
10 Lamarche, op. cit., 514.
be presented, will be a few observations on the distribution of the so-called

“absurd” or “improbable” reading and its possible background, contents, and
context, particularly in its pointed form.
Before turning to these observations four prefatory remarks might be ap-
propriate, one of principle, one fairly brief reference to the actual “state of
the arts” in contemporary exegesis and theology, and two notes of a more
technical order.
First, to my mind, the only really “absurd” thing for scholars concerned
with questions of history would be not to try to register what attempts have
been made throughout history of placing the right and proper commas and
full-stops in a given text and to understand the background, contents, and
context of such attempts.
Second, as was hinted at above, it may be true to say that it is the prevailing
point of view among scholars that one reading (III and/or IV) is the old – and
therefore the (most) original, even the “authoritative” – one, while the “new
reading” is the outcome of theological precisions against heretic exegetes, be
they Arian or maybe Gnostic. The symbolic contemporary sign of the victory
of this point of view is the inclusion of the “old” punctuation in the text of
Nestle’s Novum Testamentum from the 26th edition, for which Kurt Aland
was responsible. From Erasmus until Nestle’s 25th edition the text chosen in
printed New Testaments would usually be that of the “new” reading.11
The principal theological point of view in Miller’s book could very briefly
be summarized in the statement that ὃ γέγονεν of verse 4a is about the Lo-
gos, not about creatures. In his own words – and in the light of the notion of
“salvation-history” – Miller’s point of view is presented in brief as follows:
ὃ γέγονεν ἐν αὐτῷ ζωὴ ἦν καὶ ἡ ζωὴ ἦν τὸ φῶς τῶν ἀνθρώπων is a reference to the
life-giving and light-imparting incarnation of the Logos. In order to make this case we
will have to argue against the several attempts which, to make sense of Reading II,12 take
ὃ γέγονεν in vs. 4a as being in various ways a continuation or extension of the πάντα
11 The difference between the “old” and the “new” punctuation and some of its theological implica-
tions could be illustrated through quoting vss. 2-4 of the translations in the Old and New English
Bible, respectively: “The same was in the beginning with God. All things were made by him; and
without him was not anything made that was made. In him was life; and the life was the light of
men” (1876 version); “The Word, then, was with God at the beginning, and through him all things
came to be; no single thing was created without him. All that came to be was alive with his life, and
that life was the light of men” (NEB, 1961).
12 III/IV in the above survey, cf. Table 1.
in vs. 3 (something like, “that (creation) which has come into being through him”). But
then, and more constructively, we will have to provide evidence for our own interpreta-
tion according to which the perfect tense ὃ γέγονεν signals a movement, logically and
temporally, to the historical incarnation of the Logos (something like, “what has (now)
come about through him”).13
Miller has in no way met with total consent. For contrasting views (both on
punctuation and theology) I shall here confine myself to a brief reference to
Aland’s pages about modern commentators (not least Rudolf Schnackenburg)
(partly added in the 1979 edition of Aland’s article).14
In a number of works referred to above (note 4) direct and indirect refer-
ences to the theological and exegetic discussions of the Old Church are given.
One reference should be added, viz. to M. F. Wiles’ book The Spiritual Gospel,15
in which the author treats the general lines of the history of Johannine exegesis
in the age of the Church Fathers.
My third remark will be a reminder of the fact that authors of the Old
Church who quote the first verses of the Fourth Gospel, very often break off
their quotation at a point which makes it difficult to judge what was their
punctuation “between verse 3 and verse 4”. Aland (and Miller) may be right
in taking Lamarche to task for having drawn too far-reaching “positive” con-
clusions from this “negative” fact,16 but even then it should be borne in mind
that although this very fact – as well as the fact strange to the modern exegete
that one author may “represent” more than one reading – makes it difficult in
some cases to give a final coherent judgment about an author examined for his
contributions as an exegete and a user of the Bible, one should be attentive to
the fact that even in situations where “verses 1 and 2” are used in a particular
theological discussion one should not jump to conclusions about the reading
and punctuation presupposed in “verses 3 and 4”, to which could be added,
as emphasized by Aland, that Bible manuscripts are – as a rule – not very
helpful in deciding questions about the original punctuation.17
Our fourth and final prefatory remark should point out that the possibilities
13 Miller, op. cit., 14f.

14 Aland, op. cit., 360ff, see esp. 361f with note 12.
15 Maurice F. Wiles, The Spiritual Gospel: The Interpretation of the Fourth Gospel in the Early Church, Cam-
bridge 1960.
16 Lamarche, op. cit., 517, cf. Aland, op. cit., 367, and Miller, op. cit., 30.
17 Aland, op. cit., see esp. 365f. 368ff.
of punctuation may not be exhausted through de la Potterie’s survey. As hinted

at by Lamarche in the passage quoted above18 it is necessary to consider the
possibility of Reading II as an intermediate form between I and III/IV in the
sense that two full stops might be conceivable – before and after the clause:
ὃ γέγονεν (,) ἐν αὐτῷ. Syntactically, of course, this clause is not a full one,
but as we shall see, it may be taken to express a theological message and to
be a stylistic element natural to the prologue of John.19 Above I took this to
be a “pointed form” of Reading II. It might be useful to choose a particular
designation for this reading. Below I shall therefore indicate this reading
through a “IIp”:
IIp. ἐγένετο οὐδὲ ἕν. Ὃ γέγονεν ἐν αὐτῷ. Ζωὴ ἦν.
Turning now to the evidence of two authors of the 4th and the 5th centuries
it might be in accordance with the modesty of our contribution to introduce
it through correcting a minor misprint which has survived in Aland’s article,
even in the 1979 version. In Aland’s note concerned with patristic evidence
for the “absurd reading”, among the very few witnesses mentioned there is a
text to be found “unter den Dubia des Chrysostomus”, viz. “MG 63, 514f”.20
The reference should be to col. 544f.
This correction, however small it may be, immediately takes us beyond
the sphere of minor inexactitudes and trivial corrections, as it helps us to
identify the author of the homily in question, the reference being to the hom-
ily: In illud: In principio erat verbum (CPG 4210). This homily is by Severian of
Gabala, as was demonstrated originally by Johannes Zellinger (with references
to Dupin and Tillemont).21 The Armenian transmission has made it possible
to strengthen further the attribution to Severian,22 which, incidentally, seems
18 See above, p. 173 with note 10.

19 Lamarche is aware of this, cf. his comments on punctuation “C” (= II), op. cit., 514, and his refer-
ences to Eusebius of Emesa and others, op. cit., 519, note 81, cf. below.
20 Aland, op. cit., 357, note 7.
21 Johannes Zellinger, Studien zu Severian von Gabala, Münsterische Beiträge zur Theologie 8, Münster i.W.
1926, 37ff, see esp. 38.
22 Cf. Henning J. Lehmann, “The Attribution of certain Pseudo-Chrysostomica to Severian of Gabala
confirmed by the Armenian Tradition”, Studia Patristica 10, Texte und Untersuchungen zur Geschichte
der altchristlichen Literatur 107, 1970, 121-130 (see esp. 122-124), and Henning J. Lehmann, Per Pis-
catores, Studies in the Armenian version of a collection of homilies by Eusebius of Emesa and Severian of
Gabala, Aarhus 1975 (see esp. 273-286). The Armenian version was published by Nerses Akinian as
to have been generally accepted on the grounds given by Zellinger. There is

therefore no need here to go into any detail about the question of authorship,
but one element of a stylistic-exegetic-homiletic order pointed at by Zellinger
should be mentioned. I am thinking of his reference to the homily De sigillis
(CPG 4209) which shows a number of lines of connection with the homily on
John 1,1ff, important to Zellinger’s argument. Among other things, in De sigillis
the frequency – πολλάκις – of the use of the verbal form ἦν in the first verses
of John’s Gospel is emphasized as a means by which the evangelist brings out
what is characteristic of the divine Logos. This then becomes part of an argu-
ment that heretics should abstain from insisting on an οὐκ ἦν (understood:
about the divinity of the Son). This πολλάκις will turn out to be an element
stressed in some of the texts to be referred to below.
In the Armenian translation of the homily on John 1,1ff (CPG 4210) there
is a reminiscence of the πολλάκις-motive, as in the third line of the Akinian
edition there is a comment on the ἦν of John 1,1f running as follows:
Զէրն, զէրն, առ չէրն. եւ զԱստուածն առ աստուած, առ չաստուածսն
over against the brevity of the Greek: ἦν, πρὸ τοῦ, οὐκ ἦν.23 Another differ-
ence between the Greek and the Armenian concerns the passage referred to
by Aland. Where the Greek has the punctuation No. II, the Armenian has a
punctuation corresponding to No. I.24 However, the Armenian evidence about
Severian’s readings is not exhausted through these references. It is particularly
noteworthy that in a section of one of the homilies edited by J. B. Aucher com-
menting upon the first verses of John there is a quotation of John 1,4, which
clearly takes ζωὴ ἦν to belong to the following:
վասն այնորիկ ի նոյն յարէ, թէ կեանք էր, եւ կեանքն էր լոյս մարդկան25.
It should be noted that in one text Severian seems to represent the “new”
No. IX of the collection which he gave the title “Reden des Bischofs Eusebius von Emesa”, Handes
Amsorya 72, 1958, 161-182.
23 Akinian, HA 1958, 161f, 3f – cf. PG, 63, 543, 17f.
24 Akinian, HA 1958, 163f, 32f – cf. PG 63, 544, 22ff. If Akinian’s punctuation is in accordance with the
manuscript, it presumably rather illustrates the “normalizing” trend in the history of transmission
than two “original” versions of the homily.
25 Severiani sive Seberiani Gabalorum episcopi Emesensis homiliae nunc primum editae per P. Jo. Baptistam
Aucher, Venice 1827. The homily referred to is No. V. For the quotation see 196.
punctuation, viz. in the first Genesis homily,26 thus illustrating an “ambiguity”

corresponding to that found in Epiphanius.27
Without entering into the debate Aland/Miller versus Lamarche hinted
at above I should, however, also like to note that a number of quotations in
Severian of the beginning of the prologue of John are too short to determine
with certainty what punctuation they represent.
Before leaving Severian it should finally be noted that his theological op-
ponents in the discussion about the first verses of John are very clearly the
pneumatomachoi. Severian’s main argument is that it is evident that the πάντα
of verse 3 does not include the Holy Ghost – just as Moses and the prophets
are not included in the πάντες of Jesus’ words in John 10,8: “All who came
before me are thieves and robbers”.28
Severian’s way of reasoning might be called an argument from theological
common sense in interpreting apparently all-inclusive statements of the Bibli-
cal text. And, as has appeared, in his implying of that method to John 1,3.4 he
can make use of Reading No. II, so it may be concluded that although writing
at this late stage of (post-)Arian theological discussion he has not (in general)
felt obliged or been taught to use the “new reading” taking ὃ γέγονεν with
the preceding clause.
The most important gain for patristic scholarship from Akinian’s edition
of the above mentioned collection of homilies might be the bringing to light
of eight homilies by Eusebius of Emesa (Nos. I-VIII). This verdict may be true
also when applied to the discussion of the punctuation in John 1,3.4, as we
find the “absurd reading” (No. II) in one of the Eusebian homilies, as well.
The relevant part of section 24 of Homily No. III runs as follows:
Վասն առաջին ծննդեանն ասէ. “Ի սկզբանէ էր Բանն. եւ Բանն էր առ Աստուած. եւ

Աստուած էր Բանն. սա էր ի սկզբանէ առ Աստուած”: Չորեքին ասէ զ”էրն”, զի յայտ
արասցէ զէականն: Իսկ վասն այլոցն արարածոց ասէ. “ամենայն ինչ նովաւ եղեւ. եւ
առանց նորա եղեւ եւ ոչինչ, որ ինչ եղեւ նովաւ”: Զլինելոցն երեքկին եւ զարարչէն
չորեքկին:29
26 PG 56: 431f.
27 On Epiphanius, cf. e.g. Lamarche, op. cit., 515; Miller, op. cit., 49f.
28 This way of reasoning and the combined reference to John 1,3 and John 10,8 is found both in the
homily on John 1.1ff: PG 63, 545 (cf. Akinian, HA 1958, 163f, 39ff), and in Aucher, Homily V (De Pas-
cha…), 195f.
29 Akinian, HA 1957, 115f, 46-52.
About the first birth it is said: “In the beginning was the Word, and the Word was with
God, and the Word was God. The same was in the beginning with God”. Four times it
says “was” in order to make manifest what exists (of itself).30 About the other (beings),
the creatures, it says: “All things were made by Him, and without Him was not any
thing made, that was made by Him” – three times about what was made, and four times
about the Creator.
Whether or not Akinian has manuscript support for his punctuation, two very
clear observations can be made from the above quotation: 1) that Eusebius here
represents the “absurd” Reading II (whether II or IIp cannot be decided here),
2) that in his comments he uses what was referred to above as the πολλάκις
element. It is apparently an important (traditional?) element of the exegesis to
count the ἦν’s which are all about the Creator and the “first birth/offspring”
and the ἐγένετο’s and γέγονεν’s which are all about the created world.
Both elements can be further substantiated through the Latin transmission
of two Eusebian texts. Lamarche seems to be the only writer in recent years
to have noticed this.31 The texts in question are two homilies in the so-called
Troyes collection.32
What shall be attempted here is to elucidate both theological topics and
elements of commentatorial technique characteristic of Eusebius, as they ap-
pear in the homiletic paraphrases of John 1,1ff, which constitute the contents
of the passages in question. In order to do that I shall quote them rather ex-
tensively.
An important principle inherent in Eusebius’ interpretation of the prologue
of the Gospel of John is that of separating those elements that belong to the
description of the divine order from those that describe the economy of the
incarnation and those which introduce John the Baptist.
The relationship between the Father and the Son, and the human imagery
that can be used to describe it is a central topic in Eusebius’ theology. Thus,
characteristically, his quotation of John 1,1f in the homily De Fide, section 7,
is preceded by considerations (in section 6) about the image of the Lord as
30 “What exists (of itself)” is chosen as rendering of the Armenian (զ)էական(ն).

31 Lamarche, op. cit., 519 with note 81.
32 Eusèbe d’Emèse, Discours conservés en latin, ed. E. M. Buytaert, vol. 1: La Collection de Troyes, Spicile-
gium Sacrum Lovaniense 26, Louvain 1953.
iustitiae sol and the reality of the visible sun, qui non in principio fuit, sed quarto
die per Filium factus est.33
Each time an image belonging to the created world is used about the Son,
it is important to stress that this does not include Him among the creatures or
suggest any need in Him. Therefore the in principio and the apud Deum of John
1,1 is explained, emphatically, to mean: non tamen in novissimus temporibus …
non tamen post saecula … ante omnes … ante caelos: non enim necessarios habebat
caelos ad habitandum. The heavens are created for the sake of themselves and
of those who live beneath them (section 7).34
Then the author turns to the verbs of the first verses of John’s Gospel, first
in section 8 the instances of ἦν/erat.
Erat, erat, erat. In principio erat Verbum; ecce semel erat; et Verbum erat apud Deum, et iterum
erat; et Deus erat Verbum, et ecce tertio erat. Hoc erat in principio apud Deum. Post quatuor
has voces adhuc dubitas quod erat? Et iterum: Vita erat et vita erat lumen hominum. Si dicit
evangelista (qui ex pectore dominico haurivit) Erat, et hoc frequenter, ex his, quae saepe dicta
sunt, nequidem unum tenes?
As appears, the frequency of the occurrences of erat is emphasized (frequenter,

cf. πολλάκις). Before turning to the verbs about created beings, the author em-
phasizes that the “eternal birth” did not mean a dividing of God, the Father:
Bonus enim est Pater et non est divisus generans Filium: potens est et non erat ei impossibile
implere suam bonitatem. Tempora enim et saecula et spatia deorsum sunt. Noli comparare uni-
genitae illi naturae nihil ex his, quae facta per ipsum sunt.35
The core notions in Eusebius’ theology: nonnatus about the Father, and unige-
nitus about the Son are now commented upon (sections 9-10), and the incom-
parability of the Only-Begotten with any of those created per ipsum is drawn
to the fore, and this section ends:
33 Buytaert, Discours 1, 83.

Non est unus ex his, quae per ipsum sunt: non est ex hoc mundo, sed et mundus per ipsum; non
caelum cum Filio, sed et caelum per Filium. Quid opus est partibus exhibere naturam? Omnia
per ipsum sunt; et nihil sine ipso; et tertio: Quod factum est, per ipsum.36
The following section (11) then draws the contrast between what is said in
verses 1-5 about the divine Logos and in verses 6ff about the human being, John
the Baptist. Factus est homo … is the translation given here of John 1,6: Ἐγένετο
ἄνθρωπος (where the Vulgate has Fuit homo). This translation corresponds
very well with Eusebius’ emphasis on the fact that the use of this verb takes
John the Baptist to belong to those characterized in verses 3-4 and 10 through
the use of ἐγένετο and γέγονεν, i.e. belonging to the created world.37
However, the most important observation in our context is, of course, that
Eusebius, unmistakably, uses Reading II in the form where Quod factum est,
per ipsum obtains the character of an independent clause, IIp.
A number of the elements described here reappear in Homily XVI of the
same collection (De hominis assumptione I).38 From this context I shall only quote
the passage showing, once again, the choice of Reading IIp, and its theological
context:
Omnia per ipsum facta sunt; dixit semel et non tacuit, sed repetit et ait: Et sine ipso factum est
nihil. Et iterum non tacuit sed dicit: Quod factum est, in ipso ut nihil suspicaret[ur] evadere
posse virtutem Filii. Sed nutu Patris sufficiebat Unigenitus opificans; faciebat vero non mutans
sed respiciens ad Patrem; noscens voluntatem Patris usus est virtute sua: habet enim acceptam
a Patre.39
Here the context makes it all the more clear that the words: Quod factum est,
in ipso are read as an independent clause dealing with the participation of the
Only-Begotten Son in the creation through his virtus, another key notion in
Eusebius’ theological vocabulary.
Speaking in terms of the history of theology, Eusebius’ comments on the
prologue of John take us to the discussions of the 330s or 340s in the wake

37 Ibid.
38 E.g., the contrasting of the divine Logos and John the Baptist is here elaborated, Buytaert, Discours
1, 365. Here verse 6 is quoted in the form of the Vulgate: Fuit homo missus …
of the Council of Nicaea. His theological specifications are directed against

Sabellian theology on one side and Arianism on the other.40
The next question to be asked here would then be that of Eusebius’ place
in the history of the Bible text and in the history of exegesis. In general terms I
have tried to demonstrate that he is rather an important witness to the Syrian
Bible of his time, i.e. a pre-Peshitta version. This, of course, is true in particular
about his commentary to the Octateuch where he must be considered one of
the most important witnesses to ὁ Σύρος, which is a translation into Syriac.41
Furthermore, it has been demonstrated that his gospel quotations, as they
appear in the Armenian translation of his homilies, have characteristics in
common with vetus syra witnesses.42
Concerning John 1,3-4 Aland is very definitive in his statement that the
Old Syriac represents the “old punctuation”. As is natural, Aland first refers
to the manuscript material. For John 1,3-4 only Syrus Curetonianus is extant,
and this manuscript “liest eindeutig A”.43 As mentioned above it is Aland’s
view that manuscript evidence of punctuation should not be overestimated.
He therefore immediately procedes to supplementary evidence. Here, by far
the most important text is Ephraem’s Commentary on the Diatessaron.44 In my
view, Aland is obviously right in stressing the fact that this text gives indisput-
able testimony to the full stop after οὐδὲ ἔν. This is most clearly demonstrated
through Ephraem’s use of an “and” before quoting the following words.45
Surprisingly, however, Aland has no comment on the second “and” in
Ephraem between the translation of αὐτῷ and ζωή. This, in Louis Leloir’s
preface to his edition of the Syriac original of Ephraem’s Commentary on the
40 Cf. P. Smulders, “Eusèbe d’Émèse comme source du De Trinitate d’Hilaire de Poitiers”, Hilaire et
son temps. Actes du Colloque de Poitiers 29 septembre – 3 octobre 1968 à l’occasion du XVIe Centenaire de la
mort de saint Hilaire, Paris, 1969. 175-212, see esp. 180f. 188ff.
41 On this question, see Henning J. Lehmann, “An Important Text Preserved in MS Ven. Mekh. No.
873, dated A.D. 1299 (Eusebius of Emesa’s Commentary on Historical Writings of the Old Testa-
ment)”, Medieval Armenian Culture, eds. T. Samuelian and M. Stone, University of Pennsylvania
Armenian Texts and Studies 6, 1983, 142-160 [above, pp. 77ff], and Henning J. Lehmann, “The Syriac
Translation of the Old Testament – as Evidenced around the Middle of the Fourth Century (in Euse-
bius of Emesa)”, Scandinavian Journal of the Old Testament 1, Aarhus 1987, 66-86 [above, pp. 107ff].
42 See Per Piscatores, passim.
43 Aland, op. cit., 366.
44 Incidentally, in 1968 Aland considered the Diatessaron to be “wahrscheinlich” older than the vetus
syra, in 1979 it was so “mit Sicherheit”.
Diatessaron is commented upon rather thoroughly. I shall quote his statement

on accentuation and punctuation in full:
Le texte du manuscrit syriaque est reproduit avec la plus grande fidélité possible; en
matière d’accentuation notamment, et de ponctuation, devant la grande difficulté, ordi
nairement, à reconnaître ce qui est de première et ce qui est de seconde main, j’ai transcrit
ce que j’ai trouvé dans le manuscrit. L’effet de cette servilité est parfois bizarre; ainsi pour
Jean i. 3-4, où le syriaque (cfr p. 6,1. 2) a un point, et après dhw’, quae facta est, et après bh,
per eum. La présence, ensuite, d’un w, et, semble indiquer qu’Éphrem finissait le membre
de phrase après per eum; j’ai donc traduit: Et illa res quae facta est per eum, et illa vita est;
mais cette option ne me donnait évidemment aucun droit de modifier le texte dans un
sens qui répondît à mon opinion.46
To my mind, Ortiz de Urbina, in his edition of the Vetus Evangelium Syrorum

brings out the text presupposed by Ephraem more precisely, translating: Y todo
cuanto era, era en El. Y El era la vida, y esta vida era luz de los hijos del hombre.47
The second w/et/y points to a reading with a full stop after αὐτῷ. Whether
or not the Syriac Ephraem in the following phrase reads ἐστίν instead of ἦν,
cannot be decided on the basis of the Syriac. The Armenian has the imperfect
էին. At least, Aland’s statement: “v. 4a beginnt mit ὃ γέγονεν und wahrschein-
lich vorgesetztem καί und hat am Schluss ἐστιν”48 is too condensed to make
the reader understand that Ephraem – like Eusebius – is a witness of Reading
II in the form with a full stop both after οὐδὲ ἕν and αὐτῷ – IIp.49
There are elements in Ephraem’s commentary (sections 4-5) which bear
some resemblance to what was called above the πολλάκις motive as well as
to other elements in Eusebius’ use of the prologue of John. I shall not go into
that in any detail. But one further element of Ephraem’s reading of John 1,4
should be noted, viz. the use of the demonstrative pronoun hnwn before the
second ζωή.50 This could be described as all the more natural after the preced-
46 S. Ephrem, Commentaire de l’évangile concordant, ed. L. Leloir, Chester Beatty Monographs 8, Dublin
1963, VII.
47 Biblia Polyglotta Matritensia, Series VI. Vetus Evangelium Syrorum et exinde excerptum Diatessaron Ta-
tiani. Editionem curavit Ignatius Ortiz de Urbina S.I., Madrid 1967, 4.
49 It should be noted that the Armenian translation of Ephraem does not have the second et. Cf. Le
loir’s edition, CSCO 137, 1953, 5. This may be due to either a technical reason in the transmission
and copying of the text or to a “normalization” by the Armenian translator.
50 Cf. esta in Ortiz de Urbina’s translation (see above, note 47).
ing pointed short phrase about the Logos: Ζωὴ ἦν, and it is interesting that the
reading with the demonstrative pronoun is also used in the quotation of John
1,4 in Ephraem’s Fourth Discourse to Hypatius in a refutation of a Manichaean
interpretation of John 1,4.51
In Ephraem’s discussion with the Manichaeans here, the issue at stake
is their idea of a “Primal Man” and the Syriac reading of singular instead
of the Greek plural of τῶν ἀνθρώπων in John 1,4. Western authors such as
Augustine and Ambrose also point to the Manichaeans as representatives of
a wrong understanding of John 1,4. Thus Augustine chooses Reading III as
a bulwark against the misleading Manichaean interpretation of the notion of
vita in their use of Reading IV.52 Ambrose in his discussion of the possible
readings in John 1,3f maintains that the Arians, when they take Logos to be
a creature, represent a Manichaean doctrine. Here, however, the issue is the
dualism which the Manichaeans read out of Reading IV, which is taken by
them to mean that if “what came into being through Him, was life”, it must
be concluded that there is something else – which did not come into being
through Him, and was death.53
It is interesting to note that after this discussion Ambrose refers to vari-
ous readings used by plerique … docti et fideles and alii, among which is found
Reading IIp. It is particularly worth noting that this reading is not refuted,
but paralleled with S. Paul’s words in Acts 17,28 and Col. 1,16:
Plerique enim docti et fideles sic pronuntiant: Omnia per ipsum facta sunt, et sine ipso factum
est nihil, quod factum est, alii sic: Omnia per ipsum factum sunt, et sine ipso factum est nihil,
deinde pronuntiant quod factum est et subiungunt in ipso, hoc est: Quidquid autem ‘factum’ est
in ipso. Quid est ‘in ipso’, apostolus docet dicens: In ipso enim et sumus et vivimus et movemur
et alibi: Omnia in ipso creata sunt.54
A corresponding – neutral – reference to Reading II can be found in a much

later (11th century) author, Theophylactus,55 whereas Theodore of Mopsuestia
51 C. W. Mitchell, S. Ephraim’s Prose Refutations of Mani, Marcion, and Bardaisan I-II, 1912-21, see esp.
I,121f (text), xc (translation), II, clvii (commentary).
52 Augustinus, Tractatus in Johannis Evangelium I, ed. R. Willems, CCSL, 36, 9f.
53 Ambrosius, De fide, III, 6, 41ff, ed. O. Faller, CSEL 78, 1962, 122f.
54 Ambrosius, op. cit., 123.
55 Theophylactus, Enarratio in Evangelium Joannis, PG 123, 1145.
is the first author to ridicule it.56 Neither of these authors gives more precise
hints about their sources for Reading II.
The evidence presented above, however, seems to me to allow the state-
ment that the occurrence of Reading IIp in Eusebius and Ephraem shows that
this reading was used by authors who wrote their main works in the second
and third quarters of the fourth century, and should now be well known as
primary witnesses to the Syrian Bible text of that period. And about the theo-
logical context of the reading we can observe that these authors considered it
a suitable bulwark against the absurdities promoted by Arian, Sabellian, and
Manichaean heretics.
Furthermore, through the evidence of Ephraem, we seem to be taken to
Tatian’s Diatessaron as the most probable source of the reading. We have also
seen that later authors (such as e.g. Severian) seem to have learnt some of the
exegetic topoi now best known from Eusebius. Others (as e.g. Ambrose and
Theophylactus) took notice of the reading without feeling the need to ridicule
it, whereas Theodore of Mopsuestia, a strong opponent of the reading, can be
seen as a forerunner of modern writers such as Zahn, Westcott, and Aland, who
have condemned the reading to a resting-place in “absurdity”. The Armenian
transmission in spite of its lack of precise rendering in the case of Ephraem
turned out to be very helpful in finding our way back to those authors who
used “Reading IIp” in a fruitful and interesting way, theologically, exegeti-
cally – and therefore historically – speaking.
56 Theodori Mopsuesteni commentarius in evangelium Johannis Apostoli, ed. J.-M. Vosté, CSCO 115, 1940,
25f (Syriac text), CSCO 116, 1940, 17 (translation)
What was Theodoret’s Mother Tongue?
– Is the Question Open or Closed?
If one seeks the answer to the question, what was Theodoret of Cyrrhus’
mother tongue, in the prominent reference books of the 20th century, there
seems to be little doubt as to the answer. I shall only quote two of them, one
from the first half of the century, the other from the second half, viz. Barden-
hewer’s Geschichte der altkirchlichen Literatur, and Johannes Quasten’s Patrol-
ogy, respectively. Bardenhewer states that Theodoret “von Hause aus syrisch
sprach”,1 and Quasten tells us that he was “acquainted with several languages
besides his own, which was Syriac”.2
It seems that Bardenhewer’s statement is built on the estimate forwarded
by Léon Parmentier, the famous editor of Theodoret’s Ecclesiastical History
in the prominent series of Die Griechischen Christlichen Schriftsteller.3 The idea
that Theodoret’s primary language was Syriac, was further elaborated by Paul
Peeters,4 and his view was taken over by Pierre Canivet, the most productive
editor of works by Theodoret in the later decades of the 20th century.5 Jean-
Noël Guinot, too, in his large work on Theodoret as exegete of the Bible (1995)
(where questions of language are crucial, of course), even if he maintains that
Theodoret has used Greek instruments in his exegesis, states with no reserva-
1 Here quoted from the edition of the Wissenschaftliche Buchgesellschaft, 1962, based on the 1st-2nd edi-
tion, 1924, vol. 4, p. 221.
2 Johannes Quasten, Patrology, III, 2nd imprint, 1963, p. 538.
3 Theodoret, Kirchengeschichte, ed. L. Parmentier, GCS 19, Leipzig 1911 (cf. below).
4 P. Peeters, Orient et Byzance. Le tréfonds oriental de l’hagiographie byzantine, Subsidia hagiographica 26,
Bruxelles 1950 (quoted Tréfonds).
5 Pierre Canivet, Histoire d‘une entreprise apologétique au Ve siècle, Paris 1958 (quoted Entreprise), esp.
p. 25ff, and Le Monachisme syrien selon Théodoret de Cyr, Théologie historique 42, Paris 1977, esp. p. 38f
and 251f.
tion that “il est certain que Théodoret connaissait le syriaque et que c’était sa
langue maternelle”.6
Only a couple of voices seem to have raised doubts as to the validity of
the assumption quoted. Most explicit was Nicola Festa, the Italian editor and
translator of Theodoret’s Cure of Greek Maladies7, but also Canivet’s co-edi-
tor of the History of the Monks of Syria, Alice Leroy-Molinghen, characterizes
Theodoret’s linguistic background as an environment where Greek was used
“couramment”.8 We shall return to the answers given by the authors here
quoted – and their arguments.
The use of ὁ Σύρος in the 4th and 5th centuries:

The question of Theodoret’s language in the context
of the history of the versions of the Bible
In spite of the remarkable consensus about Syriac being Theodoret’s primary
language, to my mind there is good reason to re-open the question. Before
turning to the history of research leading to the prevailing view I shall therefore
point to one of the presuppositions for this pleading for a re-opening of the
discussion. The fundamental issue at stake was the discussion of the language
of the Bible version called “the Syrian”, ὁ Σύρος. I shall return to Theodoret’s
use of that version; here I shall only mention that I have discussed Eusebius
of Emesa’s use of this version in two articles.9 To-day, however, it is the more
natural to refer to R. B. ter Haar Romeny’s book from 1997, where, character-
istically, he introduces Eusebius of Emesa as “a Syrian in Greek dress”.
From the title of his book, it appears (indirectly, at least) that Romeny
has concentrated his attention on the Genesis part of Eusebius’ Commentary
on Historical Writings of the Old Testament, and inside that part on the biblical
quotations. Romeny’s book, indeed, represents an important step forward
in Eusebian research, not least in identifying Eusebius’ contributions to the
Greek catenae, and it is of course (from my point of view) satisfactory that
Romeny entirely agrees that ὁ Σύρος is a version in Syriac, a point which I
6 Jean-Noël Guinot, L’exégèse de Théodoret de Cyr, Théologie historique 100, Paris 1995 (quoted L’exégèse),
p. 195. Cf. below, p. 207.
7 Teodoreto, Terapia dei morbi pagani, a cura di Nicola Festa, vol. I, Edizioni Testi cristiani, Firenze 1931
(quoted Terapia).
8 Cf. below, p. 192.
9 L 1984, 1987, cf. also L 1989 [above, pp. 77ff, 107ff, 125ff].
W hat was T heodoret s M other T ongue ? 189
strongly emphasized in 1984.10 (Already in 1975 I argued that through Latin

and Armenian translations, in some instances, Eusebius’ Syriac Bible was
recognizable.11)
Romeny’s laudable concentration on Eusebius, however, means that for
other writers of the Old Church he mainly builds on other scholars, e.g. on
Guinot as far as Theodoret is concerned. As, however, the estimate of the OT
and LXX scholar, Alfred Rahlfs (as given in 1915) – that in Theodoret’s com-
ments on Judges 12,6 he had found “einen vollständig sicheren Beweis” that
ὁ Σύρος was a Greek version – still seems to be given some consideration,12
Romeny’s sound and evident evaluation of the language of this important
version as used by Eusebius, is here referred to as a new source of inspira-
tion for reconsidering the question of languages (and Bible texts) concerning
a Syrian bishop writing about 100 years later than Eusebius, i.e. Theodoret of
Cyrrhus.
Before turning to Theodoret himself I shall substantiate my introduction
to the state of research. For the sake of brevity this will not be a full history,
but concentrate, first, on Canivet’s portrait of Theodoret and its sources –
representing the consensus or majority described above, and second, on the
“minority”, i.e. the two representatives of doubt.
Pierre Canivet’s portrait of Theodoret – and its sources

Pierre Canivet’s book Histoire d’une entreprise apologétique (1958) is concerned
with Theodoret’s work Graecarum affectionum curatio (or Cure of Greek Mala-
dies – hereafter quoted: Curatio).13 In his description of Antioch, Theodoret’s
native city, Canivet first deals with the “politique religieuse et survivances
païennes de 385 à 430”. Then he turns to “le milieu social antiochien”, where
his keyword is cosmopolitisme, a determining factor in commerce, culture,
religion and demography.
For a linguistic description of Antioch, again the word “cosmopolitism”
10 R. B. Ter Haar Romeny, A Syrian in Greek Dress. The Use of Greek, Hebrew, and Syriac Biblical Texts in
Eusebius of Emesa’s Commentary on Genesis, Louvain 1997, see esp. p. 71ff.. Cf. Guinot, L’exégèse, p.
186ff.
11 Cf. Henning J. Lehmann, Per Piscatores. Studies in the Armenian version of a collection of homilies by Eu-
sebius of Emesa and Severian of Gabala, Aarhus 1975. Cf. also below, pp. 235ff.
12 Alfred Rahlfs, “Quis sit ὁ Σύρος?”, Kleine Mitteilungen II, Nachrichten von der Gesellschaft der Wissen-
schaften zu Göttingen, Phil.-hist. Klasse, 1915, p. 420‑428. See esp. p. 425ff.
13 Entreprise, cf. above p. 187, note 5.
is valid: Latin is the official language of legislation, Greek is the language of

culture. In a certain competition with Greek, Syriac has a twofold function: it
is the everyday language of communication, but it is also a literary language;
and in this respect it is of increasing importance, mainly because of Ephraem
the Syrian and because of the translation of central ecclesiastical literature
from Greek into Syriac.
Turning to Theodoret, Canivet first refers to the 17th century editor H. de
Valois (Valesius) for the view that Theodoret’s family was of Syrian origin.
Next he turns to Léon Parmentier, from whom he quotes the crucial passage
in the introduction to Theodoret’s Ecclesiastical History:
Seine Mutter war nach den Begriffen der Zeit eine Heilige (…). Das Kind, das durch ein
Wunder nach sechzehnjähriger Ehe geboren und sogleich Gott geweiht war (…), wuchs
im Umgang mit den berühmten Asketen auf, deren Leben es später schreiben sollte (…).
Die Sprache, in der diese Mönche mit ihm redeten und die auch seine Muttersprache
war, ist das Syrische. Dies geht aus den Worten Graec. aff. cur. V S. 145 ff (Raeder) hervor:
ταῦτα λέγω οὐ τὴν Ἑλλάδα σμικρύνων φωνήν, ῇς ἀμηγέπη μετέλαχον, und es scheint
sich aus derselben Stelle … klar zu ergeben, dass Theodoret niemals lateinisch verstand.
Daher kommt es, dass er das Griechische mit der Correctheit eines Fremden schreibt, der
sich bemüht hat, es nach den Regeln der Grammatik zu lernen”.14
In addition to Parmentier, Canivet calls the attention to reports about Theo-

doret’s mother’s contacts with monks, whose primary (or only) language was,
presumably, Syriac, and to the fact that the prevailing language in the diocese
of Cyrrhus was Syriac. He also refers to his own remarks about metaphors
used by Theodoret, but these remarks, as far as I can see, only amount to the
description of Theodoret’s rhetoric as having a saveur orientale.15
Canivet rounds off his presentation of Theodoret’s cultural and linguistic
profile with the following description:
D’origine araméenne, il était bilingue et il faut le ranger parmi ces Syriens grécisés qui
s’exprimaient sans doute mieux en grec que dans leur langue nationale. A ce titre, Théo-
doret demeure un des derniers grands témoins d’une culture qui est sur son déclin et
14 Parmentier, op. cit., p. XCIX.

15 Canivet, Entreprise, p. 25. Cf. Canivet’s introduction to the edition of Curatio: Théodoret de Cyr,
Thérapeutique des maladies helléniques, ed. Pierre Canivet, SC 57, I-II, Paris 1958 (quoted Thérapeu-
tique), p. 67.
d’un grec dont la pureté s’altère de plus en plus au contact de langues plus populaires
et plus pratiques.16
In this passage the formula syrien grécisé presumably contains a reference to

Paul Peeters, the famous Orientalist, whose series of lectures (given in 1943)
on “le tréfonds oriental de l’hagiographie byzantine” was published in 1950 –
after the author’s death. Peeters, in this book, in quite passionate language
attacks classical philologists for their “automatic” preference for Greek textual
evidence as compared with Oriental versions, through which procedure they
often overlook, even discredit, the significance of le tréfonds oriental.
Even if this criticism could sometimes with good reason be taken up 50
years later, in Peeters’ case it also becomes the background for the throwing
into relief of Syrian spirituality and culture in general and of Theodoret in
particular – seen as a “hellenized Syrian”, “un type … de ces Orientaux chez
qui la culture hellénique, portée à un rare degré d’excellence, a recouvert sans
l’effacer le caractère ethnique”.17
A little further on we shall discuss the two narratives in Theodoret’s History
of the Monks of Syria (or Φιλόθεος ἱστορία, Historia religiosa – hereafter quoted:
Hist. rel.), from which Peeters draws his principal arguments for Syriac having
been Theodoret’s primary (or “ordinary”) language.18 Here I shall only add
the question of what kind of connection there might be between Leloir’s and
Molitor’s prioritizing of Syriac as a “dynamic” language and Peeters’ search
for the Syriac or Oriental tréfonds in Theodoret’s work and elsewhere. But this
question concerns rather the history of ideas of 20th century Europe than 5th
century Syria.19
16 Canivet, Entreprise, p. 27 with notes 3 and 4.

17 Peeters, Tréfonds, p. 89.
18 Peeters, Tréfonds, pp. 91. 95. 105. 120f.
19 Cf. L 1982,1 [above, pp. 37ff]. Here I discuss the background of Louis Leloir’s and Joseph Molitor’s
comments on the Armenian and the Georgian versions of the New Testament, in particular the
question of a possible Syriac Vorlage, where to my mind considerations of Syriac being more “dy-
namic” than Greek, and questions about judging text historical issues on the basis of the history of
languages raise serious methodical problems. I see a certain parallel in the importance attached to
Theodoret’s “Syrian” background by Peeters.
The voices of doubt: Nicola Festa and Alice Leroy-Molinghen

Parmentier’s reading of the crucial phrase in Curatio V is explicitly contra-
dicted by Nicola Festa – or rather “by Theodoret himself”, as Festa puts it.20
The Italian scholar holds that the phrase ῇς ἀμηγέπη μετέλαχον expresses
“modest irony”, and argues that this is supported by the context; this then
leads to the conclusion that Greek was Theodoret’s language from his earliest
years and – probably – from his family background.21
Pierre Canivet and Alice Leroy-Molinghen shared the responsibility of
editing Theodoret’s Hist. rel. in the Sources Chrétiennes (1977‑79).22 From what
we know already about Canivet, it is no surprise to find the evaluation in his
chapter on L’auteur et l’oeuvre, that Theodoret’s mother tongue was Syriac. The
phrasing of the context, however, is interesting:
Il a certainement fait des études classiques; on ne sait pas dans quelles conditions, peut-
être à l’école d’un maître païen d’Antioche, en tout cas selon d’excellentes méthodes si
l’on en juge par la qualité du style et la culture de cet homme dont la langue maternelle
était le syriaque.23
Alice Leroy-Molinghen in her chapter on the Histoire du texte refers to Par-

mentier in the following way: “Parmentier a caractérisé le grec de Théodoret,
en disant qu’il l’écrivait avec la correction d’un étranger qui s’était efforcé de
l’apprendre suivant les règles de la grammaire”. Interestingly, however, she
adds: “Peut-être est-ce moins vrai qu’il ne paraît, et les classes cultivées de la
société syrienne usaient-elles couramment de la langue grecque”.24
Combining this statement with Peeters’ formula the hellenized Syrian, one
could be tempted to ask that if the process of hellenizing begins at the very
moment of the birth of a “Syrian” child, what then is his mother tongue? In
modern times perhaps one would ask: Are we concerned with a case of bilin-
guality? – maybe adding: What is real bilinguality? Or should we, rather than
searching (for ideological reasons?) for other explanations, consider Theodoret
20 Festa, Terapia, p. 18: “Ma ciò è contradetto esplicitamente da Teodoreto stesso”.

21 Festa, Terapia, p. 19.
22 Théodoret de Cyr, Histoire des moines de Syrie, I-II, ed. Pierre Canivet et Alice Leroy-Molinghen, SC
234. 257, Paris 1977‑1979 (quoted Histoire des moines).
23 Histoire des moines, SC 234, p. 14f.
24 Histoire des moines, SC 234, p. 112.
as a boy, born in a Greek-speaking environment in the cosmopolitan city of

Antioch and thus, naturally, having Greek as his mother tongue?
Theodoret’s language. A reconsideration

In the following we shall not concentrate too much on Theodoret’s birth and
infancy. Instead I shall try to extract seven points from current research and
from the existing oeuvre of Theodoret, in order to try to reach an evaluation
of his use and knowledge of languages, and their context.
The first three points will take up certain details in current research, first an
element in Nicola Festa’s description of the “modest irony” as used in Theo-
doret’s rhetoric, then an element in the characterization of Theodoret’s Greek,
as presented by Alice Leroy-Molinghen; and finally, N. Fernández Marcos’
studies of Theodoret’s knowledge of Hebrew will be used in an attempt to
understand his general linguistic equipment.
Then, in points 4‑7 we shall turn to the context of Theodoret’s writings,
first considering certain narrative elements in the stories most often quoted as
“proofs” of his (vast or primary) knowledge of Syriac. Then we turn to some
of his ideas about the history and hierarchy of languages, both Greek (5) and
Syriac (6); and finally (7) the use of Syriac in his Biblical exegesis will be com-
mented upon. (It should be noted that this article, of course, does not allow
me to present an exhausting analysis of Theodoret’s commentaries on the
Bible; as noted above, however, I have added material about the Commentary
on Jeremiah (7c), which was not included in the Danish original.)
1. The use of ἀμηγέπη in Theodoret

It might be true to say that Parmentier and Festa agree on one point, formally
speaking; i.e. in taking the author’s self-evidence about his language as a natu-
ral starting-point. However, as noted already, they entirely disagree in their
interpretation of the passage in question (Curatio V). I have already given the
verdict that I find Festa’s interpretation to be sound and correct. Here I shall
only add two observations, which might support Festa’s method of reading.
They both concern the word ἀμηγέπη. In addition to the Curatio passage,
this word is found in two places in Theodoret’s letters, viz. in letters Nos. 81
and 147. In both cases Theodoret describes, how, during his time as bishop of
Cyrrhus, he has – ἀμηγέπη – “to a certain extent” contributed to architectural
improvements of the – otherwise – inelegant city.25 To my mind the author’s

rhetoric technique used here is a characteristic example of “rhetoric mod-
esty” used in a particular way, viz. in the form of an ironic litotes, whereby
ἀμηγέπη is given the meaning “to a great extent”, “remarkably” (instead of
“to a certain extent”). Festa’s expression “modest irony” about this rhetorical
device is very well chosen.
2. Theodoret’s Greek
The whole of Theodoret’s oeuvre is written in Greek, and it is unanimously
stated in reference books and introductions that his language could be char-
acterized as pure and “classical”. To quote just one example, Alice Leroy-
Molinghen underlines the absence of neologisms in the Hist. rel.: “Toujours
est-il que, dans l’Histoire Philothée, Théodoret témoigne de la défiance à l’égard
des néologismes”.26 Some of the explanations given of this fact, when combined
with the assumption of Syriac being his primary language, have an almost
absurd ring.
3. Theodoret’s background in terms of language, culture, and theology

Two basic assumptions often form the setting for explanations of Theodoret’s
background: 1. He knew many languages – as maintained by Quasten, e.g.
(cf. above).27 2. He must have been able to communicate in a Syriac speaking
environment.
Two very general considerations might prepare our way into more specific
comments and investigations. For one thing I would agree that it is hard to
believe that a person could have a lifelong career in a certain environment
without being acquainted, to some extent, with the prevailing language of that
environment. For another thing, however, it is an open question, what amount
of knowledge is needed in order to characterize a person as one who “knows”
a language. A modern parallel to Theodoret’s situation as a preacher in Syria
could be taken from such modern churches, where it is emphasized that their
pastors should “know” Hebrew and Greek in order to base their sermons
on the languages of Holy Scripture. An acceptable “working knowledge” to
be demanded in such cases would rarely mean that the persons in question
25 I refer to Y. Azéma’s edition of Theodoret’s letters: Théodoret de Cyr, Correspondance, 1‑4, SC 40; 98;
111; 429, 1955‑1998. See esp. SC 98, p. 196f and SC 111, p. 142 (with note 2).
27 See p. 187 with note 2.
would master modern Greek or Hebrew or be able to read, on a broader scale,

extra-Biblical literature in the languages in question.
Theodoret’s knowledge of one of the Biblical languages, Greek, is indis-
putable. His knowledge of Hebrew has been studied by a Spanish scholar, N.
Fernández Marcos. As introduction to his own studies he points to two evalu-
ations, both about 100 years old, viz. those of E. Venables, who considered
Theodoret to be a “master of Greek, Syriac, and Hebrew”, and of C. J. Elliott,
who described him as having “a superficial acquaintaince [of Hebrew] as one
living in Syria would have acquired by intercourse with those who spoke a
cognate language”.28
On such a background Marcos makes a thoroughgoing, methodical inves-
tigation into Theodoret’s references to the Hebrew Bible in his two quaestiones-
commentaries, to the Octateuch and Samuel-Kings, respectively. He reaches
the conclusion, that he has met with no instances that necessarily presuppose
any knowledge of Hebrew.29 Where Theodoret refers to Hebrew words and
texts, his knowledge, according to Marcos, is rather “indirect”, i.e. he builds
on such sources as Greek onomastica, for example.
The evidence referred to by Marcos is very convincing. One could add that
Theodoret – apart from his knowledge of singular Hebrew words – reveals
a certain basic understanding of syntactical and grammatical phenomena in
Hebrew. On our way to an evaluation of his knowledge of Syriac – and with
a view to the modern parallel in Theodoret’s colleagues of to-day, I might be
tempted to consider the possibility of a positive answer such as, for example:
Theodoret had some knowledge of Hebrew, rather modest, indeed, but on the
other hand sufficient to be designated a “working” knowledge of the Biblical
language.
4. Narration, genre, and context in Theodoret’s works

As promised we now turn to the context of Theodoret’s oeuvre. First we shall
attempt to bring forward some of the narrative elements of the most important
“proof” passages about his knowledge of Syriac. In particular, two of the nar-
ratives in Theodoret’s Hist. rel., have played an essential role in Peeters’ and
28 N. Fernández Marcos, “Teodoreto de Ciro y la lengua hebrea”, Henoch 9,1, 1987, pp. 39‑54 (quoted
Lengua hebrea), see esp. pp. 39f with notes 3 and 5.
29 Marcos, Lengua hebrea, p. 51: no he encontrado ningún solo caso cuya exégesis o comentario obligue
a postular algún conocimiento del hebreo por parte de Teodoreto.
Canivet’s arguments, viz. the narrative about the Ismaelite nomads,30 and that
of the demon speaking Syriac. We therefore first turn to these narratives and
their context. Elements from the stories about Aphraates, Macedonius, and
Maësymas add interesting material; therefore some supplementary remarks
about those chapters will be presented.
a. The narrative about the Ismaelite nomads

This narrative is part of the chapter (26) about Simeon Stylites in Theodoret’s
Hist. rel. The context is as follows: After the description of Simeon’s life and
ascetic achievements, before he took up life on top of the pillar (Sections 1‑11),
the author concentrates on the circumstances and effect of the mounting onto
the pillar (Sect. 12). Such an act is stronger than words, not least in the eyes of
the Ismaelites, who, due to their awe towards the holy man on the pillar, give
up their former superstition (Sect. 13). (In the last part of Sect. 13 (presumably
a secondary interpolation) the pillar is compared with a lampstand allowing
the light to be spread far around.)
The narrative of Sect. 14 is about an episode where Simeon the Stylite
sends a group of Ismaelites to Theodoret in order that he may give them
the blessing of the Church (ἡ ἱερατικὴ εὐλογία). Barbarians as they are, the
Ismaelites believe, however, the bodily contact with the bishop’s beard and
limbs and clothes to be more effective than the words of the blessing. Simeon
must therefore shout loudly at the Barbarians in order to make them leave
Theodoret at ease.
In Sect. 15 it is Simeon’s turn to be disturbed by the Ismaelites, because
two of their tribes compete about Simeon’s blessing. Theodoret – using many
words (πολλοῖς χρησάμενος λόγοις) – now tries to make the competing
Barbarians understand that the blessing is intended for both the one group
and the other. Simeon, from his pillar, threatens them “as if they were dogs”
in order to put a stop to the strife. The important thing, however, Theodoret
concludes, is the firm belief inherent in the wildness (“mania” – Gr. ἐμάνησαν)
of the Ismaelites that the holy man’s blessing contains an immense power (…
μεγίστην ἔχειν δύναμιν τοῦ θεσπεσίου ἀνθρώπου τὴν εὐλογίαν).
In my view it is important to note that there is nothing in the narrative about
the outward forms and means of communication, such as e.g. the language
used. The narrative dynamics is not taken from the opposition between Greek
30 Peeters used the word bédouins about the Ismaelite nomads.

and Syriac, but from three different binary oppositions or dialectic word-
pairs, that of word and deed, that of the formal representative of the Church,
i.e. the bishop, versus the holy man of the desert, and that of Barbarian versus
Christian.
The second bipolarity might be quite important, even if in the story as
such it seems to be a secondary element, whereas it is obvious that the third
word-pair is important where, in Theodoret, Christianity has to a great extent
taken over the role of “the Greek” as opposed to “the Barbarian”. It should be
emphasized here that the purpose of the whole work (as stressed by Theodoret
in the prologue) is edifying. The main point of the story is thus the miraculous
power of the preaching in evoking conversion, and it could not be expected,
that all outward circumstances should be told in such a story. Therefore ad-
ditional speculations as to these circumstances must be considered as a kind
of interpolation. It could be added, that in some instances, as we shall see,
Theodoret is very attentive on linguistic facts; but here we are told nothing
about the language or dialect of the Ismaelites.
b. The narrative of the Syriac speaking demon

This narrative belongs to the chapter about Jacob Cyrrhesticus (21). He, too, like
Simeon, is one of Theodoret’s contemporaries. Thus, in this chapter, there are
again a number of tales about meetings of the holy man with the bishop, e.g.
about how Theodoret took care of Jacob when he was ill, trying to persuade
him to drive his ascetic endeavours to no further excesses (Sects. 6‑8.10).
On the other hand, as Sect. 15 tells us, Theodoret has been supported by
Jacob. Before illustrating this the author gives a description of how tough his
fights have been with the Marcionites, who have stuck at nothing, be it sorcery
or demoniac intervention.
Thus one night a wicked demon (ἀλιτήριος δαίμων) appeared, shouted
at him – in Syriac – accusing and threatening him because of his fights with
Marcion (Τί πολεμεῖς, ὁ δεῖνα, Μαρκίωνι;…), and Theodoret only escaped
being stabbed, because the demon saw that he was protected by “the choir of
martyrs and Jacob”.
In the morning Theodoret asks one of his companions whether he, too,
heard the voice of the demon (Sect. 16). It turns out that everybody has heard
the voice, but nobody has tried to awaken Theodoret, who was thought to be
fast asleep. It also becomes evident to Theodoret that the demon’s final refer-
ence (to the martyrs and Jacob) was to an “amulet” in the shape of a bottle
filled with “martyr oil” hanging at the bedhead, and a cloth belonging to
Jacob, on which he (Theodoret) had rested his head. The tale of this episode
is then followed by further accounts (in Sects. 17‑18) about Theodoret’s fight
with the Marcionites and Jacob’s assistance in this fight.
As will have appeared, the main narrative line of argument is concerned
with Theodoret’s clash with the Marcionites, and it would be natural to assume
that the narrative element of the demon speaking Syriac (which is indeed em-
phasized here) points at the fact that this is the language of Marcion’s follow-
ers. It seems to me also to be worth a moment of consideration, whether the
morning dialogue contains a hint of a narrative element stating that Theodoret
might not have understood the demon.
Be that as it may, it is not correct, at any rate, to render the narrative as
containing an element of Theodoret’s “conversing in Syriac with the demon”
(as Peeters does). Otherwise Peeters’ remarks are of a “socio-linguistic” or
social character, pointing at Syriac as being “ce patois de petites gens”. This
remark is found in a passage where Peeters considers the question, of how
far Theodoret shared the prejudice against and contempt for Syriac held by
the Antiochian upper-class:
peut-être eut-il la faiblesse de céder lui aussi au dédain fastueux que certains beaux
esprits d’Antioche et d’ailleurs affectaient pour ce patois de petites gens. Il laisse en effet
paraître qu’il tenait quelque chose de ce préjugé en nous confiant qu’il lui arriva en rêve
de parler syriaque avec le diable.31
Canivet’s comment on this narrative only contains the basic geographic / lin-
guistic information that “le syriaque est la langue dans laquelle on s’exprimait
couramment en Cyrrhestique”.32
It seems to me that Peeters’ and Canivet’s observations (however rational
and correct they might be seen from various 20th century angles) hardly
meet the drama at narrative eye-height. Seen at this level the Syriac element
to my mind strenghtens important perspectives of the narrative. Looking for
a “rationale” of these perspectives, I find the best suggestion to be that the
Syriac language was an important vehicle of 5th century Marcionitism; and in
that respect Theodoret’s experiences might be parallel to those from his war
against the use of the Diatessaron, of which Syriac was often the linguistic
vehicle, too.
31 Tréfonds, p. 91.
32 Histoire des moines, SC 257, p. 95, n. 3.
c. References to Syriac in other parts of the History of the Syrian Monks

The above cases are not the only ones, where references to the Syriac language
are of some importance in Theodoret’s Hist. rel. I shall refer to two groups of
cases, one group, where names of persons or places are described as being
Syrian, another group, where the use of Syriac is seen as a positive element
of the narrative or event in question.
The phenomenon first mentioned can be parallelled with a number of
related cases in Theodoret’s exegetical writings. I shall only call attention to
four cases in the Hist. rel., viz.:
2,1: Ἰουλιανὸς, ὃν Σάβαν ἐπίκλην οἱ ἐπιχώριοι τιμῶντες ὠνόμαζον …33
4,2: … τῇ δὲ ἐγχωρίῳ φωνῇ Τελεδὰν αὐτὴν ὀνομάζουσιν.34
10,9: … κώμη Μαρατὼ συριστὶ καλουμένη.35
13,2: … ὅθεν καὶ Γουββᾶν αὐτόν τινες ἐπωνόμαζον – ἀπὸ δὲ τῆς σύρας εἰς τῄν ἑλλάδα
φωνὴν τοῦτο μεταφερόμενον λάκκον σημαίνει τὸ ὄνομα.36
An interesting detail, to my mind characteristic of Theodoret, is the use of the

words ἐπιχώριος and ἐγχώριος in the first two examples. The words used are
thereby described as belonging to the language “of the country(side)”.
In other cases the attention on facts concerned with languages and their
use could be described as giving positive information about the history and
life of churches and monasteries, as e.g. the use of hymns in both Greek and
Syriac in the liturgy (cf. e.g. 4,13 about the monastery of Teleda, and 5,5f about
Publius, whose activities in this connection are seen as a fulfilment of the mis-
sionary command of Matth. 28,19).
Finally, three examples should be quoted, where the Syriac language has
a particular role to play in the narrative in question. The narratives are about
Aphraates, Macedonius, and Maësymas, three figures prior to Theodoret’s
own time.
The chapter about Aphraates (8) is introduced through remarks about human
nature being equal in all people, Greek and Barbarian alike; they have therefore
equal opportunities of access to philosophy. This is evident in Aphraates who

was of Persian origin, thus belonging to an extremely ill-bred (lit. “unlawful”)

nation (ἐν Πέρσαις τοῖς παρανομωτάτοις). Even then, he reached the highest
level of virtue, higher than anybody reared in a pious home. Some details are
told about Aphraates’ life, in Edessa, and in Antioch as well, and it is under-
lined that when he came to Antioch, under the guidance of the Holy Ghost he
was able to attract people through his preaching against the heretics (we are
taken to heresy-stricken Antioch of 360), even if his sermons were given in a
half-Barbarian language, as he had learned only a few Greek words: … καὶ τῆς
ἑλλάδος φωνῆς μαθὼν ὀλίγα ἄττα ῥημάτια εἷλκε μὲν ὅτι πλείστους εἰς τὴν
τῶν θείων λογίων ἀκρόασιν· τῇ δὲ μιξοβαρβάρῳ κεχρημένος γλώττῃ τὰς τῆς
διανοίας ὠδῖνας προέφερεν … (8,2). Theodoret draws a parallel to St. Paul’s
words in 2. Cor. 11,6: “I may be no speaker, but knowledge I have”.
In the chapter about Macedonius (13) we are again taken to Antioch. Mace-
donius in his speech reproaches the emperor in the case about the statues,
well-known from John Chrysostom. The editors (Canivet / Molinghen) even
consider the possibility that Chrysostom could be Theodoret’s source. Be that
as it may, Macedonius according to Theodoret gives his sermon in Syriac (Sect.
7). There is a translator present so as to make his words understandable to
the general of the imperial army. But how could such a rustic, ill-bred person
preach with such strength and such courage? The answer is (Sect. 8) that this
is due to the Holy Ghost.
The presentation referred to of Macedonius: ἀνὴρ παιδείας μὲν ἀπάσης
ἀμύητος, ἐν ἀγροικίᾳ δὲ τεθραμμένος37 is entirely parallel to the presentation
of Maësymas: Ἐγένετό τις Μαησυμάς ἐν τοῖς πρὸ ἡμῶν χρόνοις, Σύρος μὲν
τὴν φωνήν, ἐν ἀγροικίᾳ δὲ τεθραμμένος.38 It is interesting here to note the
description of the two figures as belonging to rural environments (ἀγροικία),
parallel to ἐπιχώριος and ἐγχώριος above.
In my evaluation three things are characteristic and important in such
contexts: 1) Theodoret quite obviously takes what may summarily be called
“Christian Greek culture” as his starting point. 2) He reveals an interest in
Syriac dialects – as the languages of the countryside, through which, indeed,
the gospel can be preached. 3) He only refers to such elements, where they
are significant – in terms of history and theology, and not least in terms of
narrative effect and content. But the passages in question never reveal a deep
and thoroughgoing knowledge of Syriac on the part of Theodoret himself.

5. Theodoret’s Christian Greek culture

The best introduction to what is here given the term “Theodoret’s Christian
Greek culture” is given by Theodoret himself in the 5th book of Curatio. We
are here in the context of the phrase, on which Parmentier and Festa deeply
disagree, as we saw above. The book in question generally deals with human
nature. It contains a rich arsenal of quotations from Greek authors of the clas-
sical period. This in itself is interesting enough; more important, however,
is the following: In Sect. 55 one element of the creation narrative of the Old
Testament is emphasized, viz. that only one man is created, Adam. Therefore
all human beings, to whichever people they belong, have one and the same
nature (or essence: οὐσία).
This is not contradicted by the existense of two genders (Sects. 56‑57); nor
could the variety of languages be used as an argument against the one and
same nature (now: φύσις) of man; and it is added that both among Greeks and
Barbarians do we find individuals striving for virtue and others who perform
evil deeds.39
This remark introduces a fairly long chapter about languages and cultures
(Sects. 58‑75). References to Homer and Xenophon testify to Greek acknow
ledgement of the striving towards virtue to be found in Barbarians. Language
is not the important thing, as can be learned from the fact that the majority of
prophets and apostles did not know Greek. Even then, they have proclaimed
the divine teaching (ἡ θεία διδασκαλία) all over the whole world. Two well-
known themes from apologetics and homiletics of the Early Church are added
here: Writings by fishermen, publicans and a tentmaker have superseded the
idle talk of the philosophers; so now we take Abraham and his predecessors,
Abel, Enoch, and Noah as our models rather than the schools of philosophers,
who belong, incidentally, to a much later period.
The texts of the Old Testament, originally in Hebrew, have now been trans-
lated into the various languages of the whole world, and the proclamation
of the Christian message has reached all layers in society, in the countryside
as in the cities. Referring again to St. Paul (here 1 Cor. 14,11: If I do not know
the meaning of the sound the speaker makes, his words will be gibberish to
me, and mine to him) Theodoret draws a parallel between the strangeness to
the Greek ear (or even the Barbarian character) of such languages as those of
the Illyrians, the Paeonians, the Taulantianians, and the Atintanians,40 and
39 Thérapeutique, SC 57, p. 245‑251.

40 Probably, these languages are mentioned as belonging to the “neighbourhood” of classical Greece.
the inability of non-Greeks to understand the Greek dialects: the Doric, the
Aeolian, and the Ionian.
From the old historians, Theodoret adds, we know about the great wisdom
of other peoples than the Greek, the Persians, the Indians, even the Ismaelites,
and, certainly, the Egyptians and the Romans. This is the context (illuminat-
ing, it seems to me) where is inserted the remark, discussed by Parmentier
and Festa: “I do not mention this in order to detract from Greek, a language,
which I might presumably and adequately call mine”.41 (Here I permit myself
to render the “modest irony” of ἀμηγέπη through the adverbs “presumably”
and “adequately”. Cf. the discussion above.)
6. Syriac in history according to Theodoret

We have just been acquainted with some aspects of Theodoret’s general view
of the languages of the world. In some of his texts we also find ideas about the
role, the Syriac language has played in history. I shall comment on two ele-
ments (the most important, I think), viz. 1) that Syriac is the oldest language of
all, 2) that Syriac was the language of the Aramaeans of the Old Testament. The
first assumption is presented in his quaestio No. 60 on Genesis.42 Theodoret’s
main argument is that names such as Adam, Cain, Abel, and Noah are Syrian
names. He knows about other authors, who believe that Hebrew is the oldest
language. This discussion is pursued further in the following quaestio, No.
61,43 where the author maintains that Hebrew is a holy language or a didactic
language – to be distinguished from other languages, the natural ones. (Οἶμαι
αὐτὴν ἱερὰν εἶναι φωνήν … τὴν γλῶτταν, διδακτὴν οὖσαν, οὐ φυσικήν.)
This idea is substantiated through the observation that Jewish children of
his day are taught the languages of their environments, not Hebrew, as their
primary languages.
This may add little to the “portrait” of Theodoret’s knowledge of Hebrew
as described by Marcos (cf. above), but the second observation might be of
some relevance to the understanding of Theodoret’s approach to history and
41 Thérapeutique, SC 57, p. 250.

42 Theodoreti Cyrensis Quaestiones in Octateuchum, ed. N. Fernández Marcos & A. Sáenz-Badillos, Ma-
drid 1979, p. 56. Cf. the data given by Lucas van Rompay in Le Commentaire sur Genèse – Exode 9,32
du Manuscrit (olim) Diyarbakir 22, CSCO 484, Scriptores Syri 206, Louvain 1986, p. 88, n. 9. Cf. Hen-
ning Lehmann, “Sprog, nation og religion. Nogle oldkirkelige observationer”, Sprogets mesterskab.
Festskrift til Johannes Sløks 70‑årsdag, red. Kjeld Holm and Jan Lindhardt, Aarhus 1986, p. 83‑108, esp.
p. 100ff.
43 Cf. the edition quoted (in note 42), p. 57.
to his overall interpretation of the Old Testament. A number of OT quotations

make it natural for him to comment on linguistic facts such as the language
of the Aramaeans and that of the Hebrew people.
For example, the story about the Hebrew-speaking, Assyrian chief officer
Rab-shakeh, of 2 Kings 18 (cf. Is 36) invites the juxtaposition of Syriac or Ara-
maic, the language in which Eliakim wants Rab-shakeh to speak (συριστὶ,
verse 26) over against Hebrew (here ἰουδαϊστὶ, verse 28).44
From Theodoret’s comments on Is. 19,18 it appears that Hebrew was “the
language of Canaan” (Χανανίτιδα δὲ γλῶσσαν τὴν Ἑβραΐδα λέγει, ἐπειδὴ
τὴν Χαναναίαν εἰσῴκησαν οἱ Ἑβραῖοι), and in the same context (conc. v.
23) he gives the information, that “Assyrian” can in the OT mean “Syrian”
(Ἀσσύριοι πάλαι καὶ οἱ Σύροι προση(γορεύντο)).45
By way of conclusion it seems appropriate to note that there is no romantic
approach to Syriac. Its status as the oldest language of the world does not
place it in any “golden age”. It is the language of the Aramaeans of old as of
the Syrians of his own time. All languages can transmit valuable proclama-
tions and important insights. Theodoret, however, considers himself as the
heir to classical Greek, and – indirectly – to Hebrew as the holy and didactic
language of the Old Testament and Old Israel. But as a representative of the
church, the New Israel, he does not need to have any great knowledge of the
language of the Old.
It is a truism, of course, that Theodoret should not be measured by any
modern standard of presentation of the history of languages. But his historical
horizon, as far as the use of languages is concerned, is an important element
in the understanding of his writings.
7. Theodoret’s use of Syriac (and “the Syrian”) in his biblical exegesis

a. Biblical versions and other sources
We now turn to such writings from Theodoret’s hand, as are concerned with
the exegesis of the OT. Focus will be on the question of which Bible text he
uses, and, in particular, what traces there might be of any use of “the Syrian”
(and, eventually, any other Syriac elements). Some of the more important
examples of his allusions to Syrian culture, language, and history will be
referred to as well.
44 PG 80, 785ff.
45 Théodoret de Cyr, Commentaire sur Isaïe, I-III, ed. Jean-Noël Guinot, SC 276. 295. 315, 1980‑1984, esp.
SC 295, p. 140f og 146f.
It should be noted that the text corpus relevant for this investigation is vast;
it is not, however, in all cases available in good, critical editions. So, no doubt,
my survey will call for corrections and supplements, maybe immediately, at
any rate whenever the manuscript material is sorted through and better edi-
tions might be published.
It should also be noted at the outset, that the basic Bible text used by Theo-
doret is the Septuagint (in an Antiochene-Lucianic recension). But very often,
in order to understand this or that biblical passage, he refers to the other old
Greek versions, and to “the Hebrew” and “the Syrian” as well. When doing
so, his favourite word is σαφέστερον – signifying that this or that recension
would render the meaning “more clearly”, “more precisely” than the Sep-
tuagint.
A counting of references gives the following picture):46
The Hebrew 65
The Syrian 61
Aquila 365
Symmachus 597
Theodotion 247
Quinta 3
The most conspicuous fact of the table is, of course, the distribution of quota-
tions and the prominent place of Symmachus whenever Theodoret needs a
control instrument for his Septuagint text – and the inconspicuous place of
“the Syrian”. Further, about the quotations from the Syrian it should be men-
tioned that 32 belong to the Commentary on Jeremiah, 14 to the Commentary on
Ezekiel – leaving us with very few references for the rest of the Old Testament.
We shall return to the Commentary on Jeremiah.47
46 When the figures in the survey given here differ a little from the figures given in my paper of 1987
(L 1989), the main reason is that in 1987 I had only done a provisional count from Migne’s edition,
whereas here I have used the SC-editions, where such exist. Another element should be mentioned
too, viz. that sometimes it is difficult to judge whether Theodoret has a particular verse of the Bible
in mind or refers to Syriac usage in general.
47 In this paper I include very little comparative material; it is natural, however, here to mention that
for the use of “the Syrian” (compared with other versions) Eusebius of Emesa leaves us with a very
different picture as compared with Theodoret (cf. L 1987 and Romeny, A Syrian in Greek Dress).
Commenting on the Bible, apart from versions of the OT, Theodoret also
uses other sources such as e.g. onomastica and Josephus’ works, and inside the
Septuagint he sometimes reveals knowledge of textual variations in different
manuscripts.48 His critical attitude to the text can be quite radical. In Jonah
3,4, e.g., the Septuagint has the reading “three days”, whereas “the Three” in
accordance with the Hebrew and the Syrian read “forty days”. Theodoret’s
comment is that “40” is more probable (εἰκὸς) and more true (ἀληθέστερον);
therefore it is probably the original reading of the Septuagint; however, the
very first copyists made a mistake, and this mistake has survived!49
Furthermore, a couple of characteristics of Theodoret, the exegete, should be
mentioned: 1) When the versions use different words to translate one Hebrew
word, Theodoret in his comments often combines the various possibilities
in a harmonizing rendering. 2) Of course he reads the Old Testament in the
light of the New Testament and the traditions of the Church. In this light the
Septuagint acquires a particular authority, because it is the version of the OT
used in the New. His arguments for the authority of the Septuagint are not
only theological, but historical as well. Thus he argues that the Septuagint was
completed before Christ’s birth, whereas “the three” versions are later, and
so, e.g., the πάρθενος of Is. 7,14 is there with both theological and historical
right – over against νεᾶνις.50
b. References to Syrian (and Hebrew) facts and words

Quite often the reason for implicating the “other” biblical versions in the ex-
planation of a Septuagint rendering is the fact that this translation has left an
element “untranslated”. This is often the case, where names are concerned,
both personal and geographic.
Commenting on Jonah 1,3 the exposition about the name Θαρσις, e.g.,
expands into a geographical excursus,51 explaining, among other things, how
certain exegetes (τινες) have taken the place to be identical with Ταρσις,
meaning India. However, Theodoret concludes that the Greek versions and
the Hebrew of Is. 23,14 prove that Θαρσις is Carthago, which is supported by
the readings of the Hebrew and the Syrian in Ezek. 27,12.
Frequent and characteristic elements in Theodoret’s commentaries are
48 Cf. Guinot, L’exégèse, p. 842f.

49 PG 81, 1733.
50 Cf. Guinot, SC 276, p. 286ff (NB p. 288f, note 3) and L’exégèse, e.g. p. 223f and 510.
51 Cf. Guinot, L’exégèse, p. 672f.
the explanations of figures and indications of measure and weight etc. It is

exceptional, however, that one Hebrew quantitative term, viz. the word hin
is explained in three different ways: concerning Exod. 29,40 with a reference
to Josephus, concerning Ezek. 4,11 with a general reference to Syrian usage,
and, finally, concerning Ezek. 45,24, where hin is identified with the Syrian
ἡμίναν (as rendered in Greek, of course).
Apart from such practical and historical explanations the analysis of a
particular word can dig one step deeper – as in the comments on Ps 81(80),16
and Lam 1,15, where it is noted that according to Syrian and Hebrew custom
the word καιρὸς (= time) can mean συμφορά (= fortune): Οὕτω γὰρ καὶ οἱ
Σύροι, καὶ οἱ Ἑβραῖοι [ταῦτας] ὀνομάζειν εἰώθασι.52
I have mentioned already that in a number of cases Theodoret notes differ-
ences as to grammatical gender or the use of plural versus singular in Hebrew /
Syriac and Greek, respectively (ἐναλλαγὴ ἀριθμοῦ). It is an interesting detail,
that in his comparison between the usual Hebrew plural of “heavens” with
the plural in a number of names of Greek cities,53 he uses “we” – ἡμεις – as
subject about Greek custom.
c. Theodoret’s Commentary on Jeremiah, Baruch, and Lamentations

As has been pointed out already, Theodoret’s Commentary on Jeremiah differs –
in more than one way – from what is characteristic for most of his commentar-
ies on writings of the Old Testament.
We shall, of course, be particularly concerned with differences as to the
use of Bible versions (in particular “the Syrian”). Before turning to that topic,
however, a couple of general peculiarities should be noted.
Theodoret in his preface (PG 81,496) emphasizes that he will comment upon
the Book of Jeremiah in great brevity, and, in fact, in a number of passages
his commentary could be described as a paraphrase – not much longer than
the original text itself. Examples of this concentrated paraphrastic technique
can be found, e.g., in his comments on chapters 24, 26(33), 35(42), 37(44), and
41(48). In other cases, such as e.g. chapters 31(38) and 46(26), his comments
are broader so as to look more like what is the case in other commentaries.
Related to this feature of brevity is, no doubt, the fact that anti-heretic
and anti-Jewish passages are much less frequent than in other commentaries.
52 PG 80, 1528, cf. PG 81, 785 (cf. below, p. 213).

53 Comm. in Psalm. 148,4 (cf. Marcos, op. cit., p. 42f with notes 16‑19 for parallels in the Commentaries on
the Octateuch and the Historical Writings).
(Again on that point, the comments on chapter 31(38) are, to a certain extent,
at variance with the rest of the text.)
It should also be mentioned that Theodoret follows the disposition of the
Hebrew Bible, not the differing disposition of the Septuagint.
Even if there are thus a number of characteristics to set this work apart,
there are also a number of lines of connection with Theodoret’s commentaries
on other writings of the Old Testament, so the radical solution: to consider the
Commentary on Jeremiah as a text that does not genuinely belong to Theodoret,
would distort the evidence. Below, some of the characteristics of Theodoret’s
“exegetical technique” are illustrated in combination with considerations about
lines that connect the Commentary on Jeremiah with other commentaries from
his hand.
We may then be left with considerations about particular circumstances
in the situation of the author when he wrote the text or – as far as versions
quoted are concerned – with particular conditions related to the textual facts.
In my 1987‑communication (L 1989) I put it as an open question whether there
were conditions of the first mentioned order to be taken into account. Here I
shall rather refer to circumstances of the second order.
Romeny refers to my 1989 presentation of the problem with no substan-
tial further comments.54 Guinot seems to prefer a literary approach to such
questions. For example, (for Ezekiel) he considers Theodoret’s basis to be
“un exemplaire biblique, partiellement revu sur l’hébreu, porteur de leçons
hexaplaires, et sans doute de références à la version du ‘Syrien’”,55 or he refers
to possibilities such as “une Bible glosée” with marginal textual variants,56 and
he is doubtful as to any direct use of the Peshitta. This is the background for
his reflections as to whether the Commentary on Jeremiah should be spurious.57
However, he hesitates to give a final verdict of that character.
In the following I shall try to illustrate more fully Theodoret’s use of “the
54 Romeny, A Syrian in Greek Dress, p. 73. 83 (Romeny, of course, could not know that over the years
since 1989 I had become more reluctant as to the amount of knowledge of Syriac to be assumed in
Theodoret, cf. the present article and L 2005,2, below.)
55 Guinot, L’exégèse, p. 222.
56 Guinot, L’exégèse, p. 251f.
57 Guinot, L’exégèse, p. 216: “Pour comble de malchance, le Commentaire sur Jérémie donné pour celui
de Théodoret à partir des chaînes en PG 81, n’offre pas des garanties suffisantes d’authenticité”, cf.
also p. 186, where in relation to the use of “the Syrian”, he says: “Si l’authenticité de ce commen-
taire était assuré, il serait logique d’admettre que Théodoret commente son texte en s’aidant de la
version syriaque, quitte à vérifier en certains cas la teneur exacte du texte hébreu”.
Syrian” in his Commentary on Jeremiah. First it might be worthwhile to refer to

elements of the text-historical circumstances and particulars of his references
to the Septuagint and the other versions.
The basic starting-point for these considerations stems from the fact that
Theodoret seems entirely to have given up the most fundamental element of
his “technique” as a commentator, viz. that of taking – for reasons of principle
as well as practical circumstances – the Septuagint as his textual base, using
the other versions, in particular Symmachus, as auxiliary instruments.58
In order to understand the total absence of references to “the three” the
most important fact to be taken into consideration might be the shape of the
Lucianic recension of the Jeremiah text.
I shall here confine myself to quoting the most important elements in
Ziegler’s description of this recension. First he notes that “sehr viele Stel-
len der lukianischen Rezension stimmen mit den Wiedergaben der jüngeren
griech. Übersetzer überein”.59 After a number of examples of “Austausch
verschiedener Wörter (Synonyma)” he further remarks that “die aufgeführten
Beispiele zeigen deutlich, dass Lukian an vielen Stellen die jüngeren griech.
Übersetzungen benützt hat”, often “nach den Gesetzen der Stilistik umgeän-
dert”, even if “Lukian hat nicht konsequent die Revision des alten Bibeltextes
durchgeführt”.60
Thus, there may be a possibility that Theodoret, whose textual basis, when
using the Septuagint – as mentioned above – is usually the Lucianic recension,
may have been conscious that in the case of Jeremiah this textual base differed
quite widely from what was the case in other books of the Old Testament, and
that his usual technique of comparison could only be applied in a collocation
of “the Greek version”, the Hebrew original (or column), and “the Syrian”.
In this perspective it is important to notice that in four instances the author
explicitly uses the expression “the Greek translation” for the text to be com-
pared with the Hebrew and/or the Syrian. These instances concern Jer. 12,5;
15,10; 31(38),2; 51(28),20. The Greek is referred to as ἡ ἑλληνικὴ ἑρμηνεία
(PG 81,580; 597; 749) or ἡ ἑρμηνεία τῆς Ἑλλάδoς φωνῆς (PG 81,660). Conc.
Jer. 8,6 a related formula is used, when after quoting “the Syrian” the Greek
58 Cf. above, p. 204.

59 Septuaginta. Vetus Testamentum Graecum. Auctoritate Societatis Litterarum Gottingensis editum, vol.
XV, Ieremias, Baruch, Threni, Epistula Ieremiae, ed. Joseph Ziegler, Göttingen 1957 (here quoted
from p. 85).
60 Op. cit., 86 and 92.
is referred to in the following manner: κατὰ μέντoι τὴν Ἑλλάδα φωνὴν (PG
81,556). Such formulae seem to be specific for the Commentary on Jeremiah.
In the Table below those instances are collected where “the Syrian” (and
“the Hebrew”) is/are referred to. They are quoted in the form to be found in
Migne. Important discrepancies between the Migne text and the information
to be gained from Ziegler’s edition may be referred to in the notes, but nei-
ther Table nor notes, nor remarks given after the Table intend to exhaust the
information about the Syrian version to be obtained from this text, let alone
the textual history of Theodoret’s commentary.
Technically speaking, the first column gives the references to chapter and
verse (where relevant in Jeremiah, the reference to the order of the Septuagint
is added in parentheses). The second and third columns give the references
to the Migne column in question and the (main parts of the) text given here;
and in the fourth column the references to the Syrian (s) and the Hebrew (h)
are quoted.
In the following I shall first give a few explanatory notes and comments
to the textual material contained in the Table, and after that, some relevant
supplementary material will be commented upon.
As will appear, a number of the differences listed concern proper names,
found among those names of geographical localities. The difference may
concern orthography and vocalization (25,25(32,11); 41(48),5); in other cases
we meet the well-known fact that Theodoret is eager to explain names left
“untranslated” in the Greek – cf. Σινὼχ (29(36),26), σιoνείμ (31(38),21) and
Σαὼν ἐεβεὶρ ἐμωήδ (46(26),17), or maybe translated in an unsatisfactory
manner (49,27(30,16)) or being sufficiently at variance in Heb. and Syr.,
resp., to attract attention – such as the “translations” of Iωσεδεχ (23,5f),
where furthermore, it is emphasized that none of the “translations” points
to Zerubbabel, and it is added that δίκαιoς as epitheton for Christ is found
in the Prophets in three formulae: ἀνατoλὴ δικαία, βασιλεύς δίκαιoς and
κύριoς δικαιoσύνης. It is also emphasized (about 23,4) that this prophecy
was not exhausted in Zerubbabel’s time. After him came Macedonians and
Romans (Greeks).
The longer variants, in principle, do not differ from those mentioned,
and the variations between the Greek and the Syrian (and Hebrew) are often
given an additional explanation, cf. 21,13, where Σόρ is said to mean συvoχή
or πέτρα.
References to “the Syrian” in Theodoret’s Commentary on Jeremiah, Baruch and

Lamentations
Jer., ch., v. PG 81 Greek Syr., Heb.

1,11 500 καρυΐνην ἀμυγδαλίνην (s,h)
3,2 517 ἐξεφύρθης ἐμιάνθης (s)
8,6 556 διέλιπεν ὁ τρέχων ἀπὸ τoῦ δρόμoυ πάντες τῇ γvώμῃ αὐτῶν
αὐτoῦ ὡς ἵππoς κάθιδρoς ἐv πoρεύovται, ὡς ἵππoς ὁρμῶν
χρεμετισμῶ αὐτoῦ εἰς πόλεμoν (s)
12,3 577 ἄγνισoν εὐτρέπισoν (s)
12,5 580 σoῦ oἱ πόδες τρέχoυσιν εἰς κακίαν, καὶ μετὰ πεζῶν τρέχoυσα
ἐκλύσoυσί σε. Πῶς παρασχευάσῃ ἐφ̒ ἡττήθης, καὶ πῶς ἀντιστῆναι
ἵππoις ἱππεῦσι δυνήσῃ; (s)
12,12 581 διεκβoλὴν ἀτραπὸν (s)
14,8 592 ὡς αὐτόχθων ὡς ὁδίτης (s)
15,10 597 oύκ ὠφέλησα, oὐδὲ ὠφέλησε oὔτε ὤφληκα, oὔτε δεδάνεικα
(s)
17,6 604 ἐv ἀλίμoις ἐν φωλεoῖς (s)
21,13 618 τὸv κατoικoῦντα τὴν κoιλάδα Σὸρ τὴv μεταξὺ ὁρῶν κειμένην ἐν
πεδινὴν κoίλῳ τόπῳ (s)
23,5f 628 Κύριoς δικαιoσύνη ἡμῶν Κύριε δικαίωσoν ἡμᾶς (s)
25,25 (32,11) 637 Ζαμβρή Ζεμβρά (s,h)
29,26 (36,26) 652 Σινὼχ εἰρκτήν (s)
30,20 (37,20) 657 μαρτύρια συναγωγὰς (s)
31,2 (38,2) 660 θερμὸν oἰκτιρμὸν (s,h)
31,21 (38,21) 664 σιoνείμ (σιωνιμ) σημεῖα (s)
31,38 (38,38) 669 ἀπὸ Ἀνανιὴλ ἕως τῆς πύλης τῆς γωνίας ἀπὸ πὺργoυ τῆς γωνίας
Ἀνανιήλ (s)
38,14 (45,14) 688 εἰς τὴν oἰκίαν Ἀσαλισαὴλ ἔσω τριόδων (s)
41,5 (48,5) 696 Σαλὴμ Σιλὼμ (s,h)
46,14 (26,14) 712 τὴv σμίλακα τὰ ἐν κύκλῳ σoυ (s)
46,17 (26,17) 712 τὸ ὄνoμα … Σαὼν ἐεβεὶρ ἐμωήδ τὸ ὄνoμα αὐτoῦ, ταράττων,
καὶ καιρoὺς παραφέρων (s)
48,31 (31,31) 724 κειράδας αὐχμoῦ κεφαλαὶ αὐτῶν κεκαρμέναι,

καὶ oἱ πώγωνες αὐτῶν
ἐξυρημένoι (s)
48,33 (31,33) 724 Οἶνoς ἦν ἐν ληνoῖς σoυ· πρωΐ oὐκ Οὐκέτι δὲ ληνoβατoῶντες
ἐπάτησαν, oὐδὲ δείλης oὐκ ἐπoίησαν λέγoντες, Ἰὰ, ἰά (s)
49,27 (30,16) 733 Υἱὸς Ἄδερ Βαραδὰδ (s)

Jer., ch., v. PG 81 Greek Syr., Heb.

50,8 (27,8) 740 ἐρίφoυς τράγoυς (s)
50,39 (27,39) 745 θυγατέρας Σειρήνων στρoυθoκαμήλoυς (s)
51,20 (28,20) 749 Διασκoρπίζεις Ἑτoίμασoν (s)
52,18 757 πoδιστῆρας νιπτῆρας πoδῶν (s)
Lam., ch., v.
1,12 784 ἐπεφύλλισε ἧψε (s)
1,22 785f ἐπιφύλλισoν αὐτoὺς, ὃν τρόπoν θλῖψoν αὐτoὺς, ὡς ἔθλιψας

ἐπεφύλλισας ἐμὲ ἐμέ (s)
3,16f 796 ἐξέβαλεν ἐν ψήφῳ τoὺς ὁδόντας μoυ, ἐν πέτρᾳ ἐστέρησέ με καὶ τῆς
καὶ ἐψώμισέ με σπoδόν πρoτέρας εἰρήνης, καὶ τῶν
παντoδαπῶν ἀγαθῶν (s)
3,29f 797 ἐν χώματι ἐν γῇ (s)
4,18 804 ἐθηρεύσαμεν ἐθήρευσαν (s)
In some cases a full and final evaluation of a parallel or a variation between

Bible versions in Theodoret would demand a much deeper examination, than
can be given here, not only into the history of the biblical text and its exegesis,
but also into the history of languages, semantics and metaphorics. Let me il-
lustrate this from the comments on Jer. 50,39 (27,39). Theodoret first comments
upon the word ἰνδάλματα of the first member of the verse (here translating He-
brew ‫צײם‬61) through the phrase: ἰνδάλματα δὲ τὰς τὼν δαιμόνων φαντασίας
καλεῖ. Then he turns to the second phenomenon of the verse, the θυγατέρες
σειρήνων. The immediate impression might be that he refers to a variant
reading, “the Syrian” reading: στρoυθoκαμήλoυς instead of “the daughters
of Sirens” of the Greek. However, for one thing, dictionaries consider, e.g.
conc. Is. 13,21, the meaning ostrich for σείρην.62 Secondly, it should be noted
that when commenting on Is. 13,21, Theodoret in the first place seems to take
the word σειρῆνες on “face Greek value”, meaning τoὺς ταῖς παντoδαπαῖς
καταθέλγoντας ἐξαπάταις. Then he adds that “the other interpreters” read
στρoυθoκάμηλoι instead of σειρῆνες. Finally, he remarks that the ἐχῖνoι of
Is. 13,22 are also animals of the desert (φιλέρημoν … τῇ ἐρημίᾳ).63
The lines of connection to the comments on Jer. 50(27),39 are obvious.
61 This word is also explained in the Commentary on Isaiah.

62 See e.g. Liddell & Scott, s.v. III.
63 Comm. Is., ed. Guinot, SC 295, 80.
Commenting on the word σείρην, Theodoret here states that it is “usual

in the divine Scripture” to use this word in the sense of “demon”, διὰ τὸ
καταθέλγειν καὶ ἐξαπατᾷν τoὺς ἀνθρώπoυς, i.e. the same choice of words
as in the Commentary on Isaiah. Then, too, the reference to the desert is given:
καὶ oὗτoι μέντoι κἀκεῖνoι τῆς ἐρημίας σημαντικoί – leaving it open to the
reader, to what extent the Syrian reading is used in order to place “both” in
the desert.
Guinot comments upon Theodoret’s remarks about the ὀνοκένταυρoι of Is.
13,22 that – in opposition to Cyril who takes these creatures – “en un sens plus
concret” – to be identical with wild asses, “le terme … n’est pour Théodoret
qu’une manière humaine de désigner les démons”.64
In addition to this observation it might be true to say that the important
thing for Theodoret is to bring out what is the “meaning”65 of the elements of
a given Bible text; and that to him means its basic “themes” – such as e.g. the
activity of demonic creatures and the desert as their place. This can be done
in a more complete form or in a condensed “abbreviation”, but even in the
brief form key notions and words may be present and the relevant information
from the existing versions may be included.
As has been mentioned already, Theodoret’s particular interest in explain-
ing names of places both on a linguistic, a historical and a geographical level
appears in the Commentary on Jeremiah. A few important cases (outside those
listed in the Table) should be mentioned.
Jer. 4,15 furnishes Theodoret with the occasion to comment upon the geo-
graphical extension of Israel in the past. Here, corresponding to his comments
on Jer. 39,5, he inserts the identification of Ribla (in Hamat) (Ῥεβλαθὰ τῆς
Ἐμὰθ) with (the great) Emesa, adding that there is another Emath, to be identi-
fied with Epiphaneia (i.e. Hama).
In a number of cases Theodoret’s comments on usage as to geographic
names are very brief, e.g. about the identity of On and Heliopolis (43(50),13),
the use of Ararat for Armenia (51(28),27), Bosor (of 49,13.18(29,14.19)) not be-
ing (the well-known) Bostra, sometimes restricting his notes to a reference to
other works of his (e.g. about 25,23(32,9): Thaiman and Dedan).
Corresponding brief annotations are made about non-Greek words other
than names of localities, e.g. μεχωνωθ (27,19(34,16)), φαγαρίμ and σπoδία
(31(38),40), μαναα (Bar. 1,10).
64 Comm. Is., ed. Guinot, SC 295, 80f, n. 2.

65 Note the choice of the word: σημαντικός.
In certain cases the historical content of an annotation or a reference is

somewhat richer, as when e.g. he comments upon Cypriots and Cappado-
cians as elements of the population of Palestine (47(29),4, very briefly; with
reference, however, to his Commentary on Isaiah).66
One case has already been quoted, where the formula used by Theodoret
might for an immediate reading lead to considering the reference to be to the
Hebrew and the Syrian Bibles (thus belonging to the Table), viz. his reflection
on Lam. 1,15, which is given in the form: Τὴν συμφoρὰν καὶ ὁ Σύρoς, καὶ ὁ
Ἑβραῖoς, καιρὸν καλεῖ.67 Here, however, the idea of course is to explain the
καιρὸς of the Greek as a “Syriacism/Hebraism”, and the reference is therefore
not included in the Table.
While in most of the cases quoted Theodoret concentrates on variant read-
ings for single words, one passage in particular can illustrate how the entire
context may be included or illuminated through one or two such observations.
In Jer. 48(31),31 Theodoret seems to take the Κίρ Αδας (Hebr. Kir-Heres) to be a
form of Greek κειράς (κειρὰδας). He first quotes the parallel of the Syrian, and
then after having commented briefly on verses 32 and 33 – again quoting the
Syrian for its rendering of the main elements of verse 33, he finally concludes:
καὶ διαφόρως τὰ αὐτὰ εἰρηκὼς, ἡρμήνευσε τὸ “κειράδας αὐχμoῦ”.68
It may be true to say that this passage – besides containing the most ample
quotations of the Syrian – is also in its conclusion the most explicit, as far as
its argument for the harmony among the varying versions is concerned.
This harmonizing feature has been mentioned above; however, I shall only
quote one further example here, viz. Theodoret’s comments on Jer. 46(26),14.
The reading of the Syrian quoted above, which is in accordance with the He-
brew Bible, is also shared by “the others” according to Ziegler’s apparatus.69
Theodoret’s interesting – and harmonizing – explanation of the Septuagint
reading runs as follows: The σμίλαξ of the Septuagint is a plant, which is well
chosen as a metaphor (τρoπικῶς) for the auxiliaries (ἐπικoύρoυς) of Egypt,
for the σμίλαξ is twined (συμπλεκoμένη) with its neighbour plants,70 and
66 Is. 9,11. See Guinot’s remarks, ed. Comm. Is., II, SC 295, 16f, n. 1.
67 Cf. above, p. 206.
68 PG 81,724.
69 It must be considered a minor deviation that the ἐν of the Syrian is omitted by “the others”.
70 Theodoret obviously takes σμίλαξ to mean “bindweed”. According to Liddell & Scott, apart from
bindweed (smilax aspera, convolvulus sepium) it may as well mean holm-oak (quercus ilex), yew (taxus
baccata) and kidney-bean (phaseolus vulgaris).
so, in reality, it points towards the same sense (αἰνίττεται) as the reading of
the Syrian.
Even if, quantitatively and technically speaking, the biblical references in
Theodoret’s Commentary on Jeremiah differ widely from other commentaries,
the elements analysed or pointed to here fit in very well, to my mind, in the
picture of Theodoret as an exegete and homiletic user of the Bible. It should
also be borne in mind, that the quantitative argument mostly concerns the
negative balance for “the three”, whereas – even if the number of references
to the Syrian is “large” when compared with other commentaries – their total
sum only amounts to 32, out of which the great majority are concerned with
single words and phrases.
It should also be emphasized that such comments on linguistic and histori-
cal matters, as can be found in this commentary are, generally speaking, in
accordance with those to be found in other commentaries.
In my opinion the material does not allow a clear answer to the question
of whether Theodoret might have acquainted himself with the Syrian through
a Greek source; it may be most natural to assume that he has, at least, had a
certain knowledge of what was characteristic of the Bible in Syriac, but the
“large amount” of quotations in the Commentary on Jeremiah does not dem-
onstrate that their author had a deep knowledge of the Syriac language, but a
“working” knowledge sufficient to permit him to have an informed opinion
about variant readings.
Concluding remarks – The question of Theodoret’s

mother tongue to be reopened?
In the above Guinot has to a great extent been quoted in connection with
Theodoret’s Commentary on Jeremiah. There are two reasons for this: one, that
Guinot is the editor of Theodoret’s Commentary on Isaiah in the Sources Chré-
tiennes (295), and as there are a number of lines of connection between the
two commentaries, many notes and details in Guinot’s edition have been of
great value in the study of the Commentary on Jeremiah. Secondly, and most
important, as has appeared, the question of Syriac or Greek or Syriac versus
Greek appears to be of a particular character in this text. Passing on from the
particular questions to the more general question, Guinot has the following
argument: “… en dépit des doutes que l’on pourrait émettre, si l’on n’avait
pour en juger que les références au “Syrien” de ses commentaires, il est certain
que Théodoret connaissait le syriaque …”.71
To my mind this is an argument turned upside down. I can easily agree
to the necessity of considering, as Guinot does, in the context of the passage
quoted and related passages, what Theodoret’s instruments as an exegete were.
As we have seen, Guinot is mainly thinking in terms of literary instruments.
Above I have tried to concentrate on the textual background of an exegete
presumably bound primarily to the Lucianic recension, and on the links con-
necting the Commentary on Jeremiah with other exegetical works by Theodoret.
Admittedly, of course, the differences may still call for supplementary consid-
eration.72
From the fairly thoroughgoing presentation of the use of “the Syrian” in
this text I have already concluded that it hardly allows a conclusion about any
“deep” knowledge of Syriac. As suggested by Guinot, Theodoret’s exegetic
instruments may very well have been primarily Greek, and it should be re-
membered that his knowledge about “the Syrian” and about conditions and
phenomena in Syria always comes down to us in Greek. So, to my mind it is
difficult to see, where, in the face of the evidence of his own texts, one should
find the certainty (as maintained by Guinot as by the majority of patristic
scholars) of his knowledge of Syriac.
From the observations presented here, not only the demonstration of his
very limited use of a Bible in Syriac, but also what he reveals about his under-
standing of Syria (and Syriac) past and present, the most verisimilar picture
of his profile and career (in relation to matters of language) would be the
following: His mother tongue was Greek. Greek to him, too, was the primary
language of the church, the language of the New Testament and of his teach-
ers and predecessors; it is also the language through which he knows the
Old Testament, viz. in the form of the Septuagint. He knows of the status
of Hebrew as a “holy” and “didactic” language, but also as the language of
Canaan. The Christian message can be proclaimed in any language, in Syriac
too, of course, even if that is a language of rustic people and heretics. As an
exegete he realizes that the Septuagint as a translation contains a number of
problems; they should – and can – in principle be solved, primarily through the
use of other Greek versions, or through a direct inspection of the Hebrew text
71 L’exégèse, p. 195.
72 Cf. below (p. 236f) about my suggestions concerning the possible origin of the Commentary on Jere
miah in Theodoret’s “school” or “workshop”.
or consulting a Syrian version. The two last mentioned possibilities, however,

are used to such a limited extent that it seems that the conclusion about his
modest knowledge of Hebrew could be valid for Syriac, too.
However, he has, of course, as bishop of Cyrrhus acquainted himself with
a number of features of Syrian culture and ecclesiastical matters of his own
time as well as of the past, most vividly presented in his Hist. rel. Apart from
the feature mentioned already, that this language can have a negative side,
culturally speaking, as the language of the countryside, and in the perspective
of the church as the language of heretics – in particular the Marcionites – a
number of narratives, too, are about figures of Syria, speaking in the name of
orthodoxy. This is true, not least, of features of the past, whereas in his own
time the bishop rather often takes upon himself the role of supervisor of one
or another ascete and enthusiast of the desert.
Both from the historical considerations and from the modern parallel of a
pastor claiming a “working knowledge” of the Hebrew of the Old Testament,
I would plead for a positive formula of Theodoret’s knowledge of Hebrew and
Syriac as being a “working knowledge”, modest, but sufficient for his work
as an exegete and a bishop.
That his point of departure is Greek, however, is maybe illustrated all the
more clearly when examining in detail the work in which most knowledge
of Syriac (at least of “the Syrian”) is revealed, i.e. the Commentary on Jeremiah,
even if we have left that text with a number of open questions.
On the whole, a number of questions are left unresolved. In approaching
the figure of Theodoret in church history, perhaps the most rewarding perspec-
tive would be the question of how to keep together the rational scrutinizer
of linguistic and exegetic details in the text of the Bible with the narrator tell-
ing tales about religious and ecclesiastical figures of his own time, including
himself – named θεοδώρητος as the child of a miracular birth.
Severian Cento No. 2 in MS Galata 54
The 14th century Armenian manuscript, No. 54 of the Galata Collection, has
already rendered great services to Patristic scholarship.1 For the study of
Severian of Gabala two centones, consecutively placed on pages 363-375 and
375-384 of the manuscript, are of particular importance.
I have given a description of the first cento attributed to “Severian of Emesa”2
in my contribution to the Berberian Memorial Volume (L 1986). The second
cento is briefly presented in my Philadelphia paper of 1979.3 The purpose of
what follows is to give a more extensive presentation of the contents of this
collection of quotations from Severian and their original contexts as well as
their contribution to the definition of a final corpus Severiani and the character
of the cento as a “florilege homily” concentrating on collecting – and separat-
ing – biblical testimonies to the economy of salvation and to the divine nature
and dignity of Christ.
It should be noted at the outset that the collector of this cento takes great
care in quoting both title and incipit of each homily and also in separating the
single quotations from one another through such formulae as: եւ յետ յոլովից
(and after a long passage); եւ յետ սակաւուց (and after a short passage); եւ յետ
այլոց (and after other things). I use the siglum G to designate the manuscript,
MS Galata, No. 54.
1 I am mainly referring to the Irenaeus fragments published by Charles Renoux, in: Irénée de Lyon,
Nouveaux fragments arméniens de l’Adversus Haereses et de l’Epideixis, Patrologia Orientalis 39,1,
Turnhout 1978. For a description of the manuscript, see Renoux’s introduction, pp. 13ff. A full de-
scription (in Armenian) is found in Babgēn’s catalogue: Babgēn [Kiwlēsērean], Ցուցակ ձերագրաց
Ղալաթիոյ ազգային Մատենադարանի Հայոց, Catalogue of the Manuscripts of the Armenian National
Library of Galata, Antelias 1961.
2 On this attribution, see L 1986, 480 [above, pp. 97f], and Per Piscatores, 147ff.
3 L 1982,2, 117ff [above, pp. 72ff].
1. One quotation from the homily

De matre filiorum Zebedaei (CPG 4249)
This first quotation in the cento derives from the homily published in 1913
by Hermann Jordan as the “29. Stück” in his collection of Armenian Irenaeus
fragments.4 The title of the homily is identical with that given by Jordan, but
the incipit is different. The first word in the Galata text is վկայիցն; in the
second word there is a lacuna; supposedly the word is արիւն, the two words
together meaning “the blood of the martyrs”. I take the difference of the incipits
to mean that Jordan’s text is incomplete, but I have not been able to trace any
homily, whether in Greek or Armenian, with an incipit corresponding to that
of G. Therefore I have no basis for determining, how large the initial lacuna
of Jordan’s text is, let alone filling it.
The text comprises 34 lines of Jordan’s edition (p. 32, l. 9 – p. 33, l. 19; Ger-
man translation: pp. 182-184, ll. 102-142). Briefly summarized the quotation
concentrates on the economy of salvation being the reason why Jesus exercises
human will and reacts to human manifestations of will. In particular, this is
illustrated through references to Jesus’ question to the mother of the sons of
Zebedee: What is it you wish? (Matth. 20,21) and the leper’s words to Jesus:
If you will… (Matth. 8,2).5
There are only few variant readings in G compared with Jordan’s text. I
shall here list the three most important ones:
1. After the first five words: եւ նա ասէ, զինչ կամիս6 (And he said: What is
it you wish?) G in addition to Jordan has an extra line emphasizing the paradox
inherent in the economy of salvation: որ զխորհուրդսն տեսանէ, հարցանէր,
զինչ կամիս. (He who sees the thoughts (of men) asked: What is it you wish?).
I presume the reason of the variation to be omission through homoioteleuton in
Jordan’s text.
2. The Armenian word for οἰκονομία is in Jordan the “classical” word
տնտեսութիւն (32,20), whereas G has տնաւրէնութիւն, which must – as a
“calque” of the Greek word divided into its “elements”: οἶκος = տուն + νόμος
= աւրէնք – be considered a “phil-hellene” feature. However, there is otherwise
in G no particular relation with what is traditionally included under “phil-
hellene” translation technique. I would therefore take this element rather to
4 Hermann Jordan, Armenische Irenaeusfragmente, Texte und Untersuchungen zur Geschichte der altchrist-
lichen Literatur, 36,3, Leipzig 1913.
5 In both cases the verb of the Greek New Testament is θέλω.
6 Jordan, op. cit., 32,9f.
S e v e r i a n C e n t o N o . 2 i n M S G a l a t a 5 4 219
be an interesting example of a translational choice illustrating that the pas-

sage from one stage or school of translation to another in old Armenia was
a gradual one, or that a copyist’s habit of rendering a word could have been
the decisive factor for the choice of word in the text, as it now appears.
3. Finally, it should be mentioned that G supports Jordan’s conjecture: լոյս
(= light) instead of յոյս (= hope).7
Jordan was well aware of the impossibility of maintaining that Irenaeus
should be the author. To some extent inspired by Nerses Akinian, the well-
known Mechitharist scholar, he considers Severian as a more probable choice.8
A few years after the publication of Jordan’s book G. Dürks argued more
extensively for Severian’s authorship,9 and this attribution has been generally
accepted.10

In natale Domini Iesu Christi (CPG 4657)
The quotation corresponds to the following lines in the Greek version: PG
61,764, l. 3 from the bottom – 765, l. 16: αἱ μαῖαι ἐψηλάφων … μὴ φθαρῆναι
(18 lines in all). In the Armenian parallel to be found in MSS Jerusalem 1 and
154, however, there is a supplementary paragraph inserted after the first seven
lines of the Migne text (i.e. after ἀνθρωπίνης, PG 61,765, l. 3). G does not have
the whole paragraph of the Jerusalem witnesses, but it has one extra phrase
here, viz. the following: Քրիստոս ծնեալ ի կուսէ առանց մարդկային սերման
(= Christ (was) born by the virgin without the semen of a man).11
The supplementary paragraph in the Jerusalem mss adds references to
Eve’s “birth” out of Adam, Moses’ striking the rock to make water pour out,
and Habakkuk bringing food to Daniel as forebodings of the miracle of the
immaculate (conception and) birth. Where the text is present in both G and
the Jerusalem mss, the amount of variations is very small, mainly of an or-
thographic kind.
Because of the surplus G could be seen as representing the same line of
7 Cf. Jordan, op. cit., 183, note 8.

8 Jordan, op. cit., 190ff.
9 G. Dürks, “Eine fälschlich dem Irenäus zugeschriebene Predigt des Bischofs Severian von Gabala”,
ZNW 21, 1922, 64-69.
10 See e.g. Zellinger, Studien, 73, and Geerard on CPG 4249.
11 Cf. L 1982,2, 114 [above, pp. 68f].
transmission as the Jerusalem witnesses. On the other hand there is no formula

of the type referred to above to divide the quotation in two and ascertain
an omission, and as the surplus is of so modest volume, and as there are on
the whole only few and small variations between G and Migne, it might be
considered, whether G belongs rather to the tradition behind the Greek text
as known to-day.
Briefly summarized the fragment preaches the miracle of the incarnation
seen as the incorruptible entering of the Son of God into the earthly life of
mankind.
For questions about authorship etc. I refer to my comments elsewhere.12
In my view the presence of the quotation in the Galata cento is an additional
and fairly strong external testimony of Severian’s authorship.

De Iesu dormiente in navicula (CPG 4699)
This fragment finds its parallel in PG 64,21, ll. 26-37: ἐκάθευδεν ὁ κύριος
οἰκονομικῶς … πρὸς τὴν προσκύνησιν ἐπιστρέφοντα. In Migne’s edition the
homily has a very long, double title: Quod mari similis sit haec vita, et in illud,
cum Iesus ascendit in naviculam cum discipulis suis, et obdormivit. The very short
quotation (corresponding to 11 lines in Migne) emphasizes Jesus’ knowledge
about the tempest to come and his power to still it (Matth. 8,23-27). However,
the incarnation as an element in the economy of salvation implies that both
Jesus’ disciples and nature (wind and sea) are allowed to act on the basis of
either belief or doubt, after which the disciples can learn their lesson, and
nature can be brought from its state of unrest into the harmony of worship.
B. Marx proposed Proclus of Constantinople as the author of this text.13 F.
J. Leroy, however, showed that the attribution to Proclus of a great number
of pseudo-Chrysostomic homilies suggested by Marx is untenable in many
cases.14 As far as this homily is concerned, Marx only refers to the comparison
between Jesus asleep in the boat and the holy child at rest in Mary’s womb.
For such a short text as this homily (very probably an abbreviation of the
12 L 1982,2, 114f and 119, and L 1995, 221f and 225ff [above, pp. 68ff, 74, 155ff, 166ff].
13 B. Marx, Procliana: Untersuchungen über den homiletischen Nachlass des Patriarchen Proklos von Konstan-
tinopel, Münsterische Beiträge zur Theologie 23, Münster i. W. 1940, 73.
14 F. J. Leroy, L’homilétique de Proclus de Constantinople, Studi e Testi 247, Città del Vaticano 1967, chap-
ter 6 and passim.
original) the amount of internal criteria must necessarily be small. Yet, a few
elements that point in the direction of Severianic authorship should be men-
tioned. In PG 49,328 and 52,819 (CPG 4186 and 4188, resp.) there are parallels
to the way in which David’s sin is described in PG 64,21f. The way in which
the author refers to an imminent risk of barbarian invasion might be paral-
lelled with corresponding elements in genuine Severian texts pointed at by
Zellinger.15 In particular, however, it would be appropriate to underline that
the repeated description of Jesus’ sleep being οἰκονομικῶς and not κατὰ τὴν
τῆς θεότητος ἀξίαν is in line with the principal argument in the other Severi-
anic texts of the Galata cento. This motif in its relationship with Jesus having
authority over the elements of nature was in general singled out as belonging
to the homiletic-exegetic thematics of Severian already by Zellinger, who refers
to parallels both in the Genesis-homilies and in the Aucher collection.16
To such internal arguments should now be added the appearance of the
quotation in the Galata cento adding an external argument, which to my mind
bears a considerable weight.
4. Four quotations from the homily

In illud: Quomodo scit litteras (CPG 4201)
These fragments, three very short quotations (nos. 1, 2 and 4), one somewhat
longer (no. 3), correspond to the following sections in Migne:
1. (PG 59,645, ll. 58-61): Οὐ πάντα γὰρ … ἀπίστοις.

2. (PG 59,645, ll. 64-74): Πρὸς Ἰουδαίους … μαχομένους.
3. (PG 59,646, l. 42 – 647, l. 20): Ἀλλ’ ἵνα μὴ νομισθῇ … τὰ νοήματα.
4. (PG 59,647, l. 72 – 648, l. 3): Ἐλέγχει Ἰουδαίων … Θεῷ μάχεται.
In content these quotations are closely related to those quoted in the earlier
sections of the cento. Here the centonist first (in section 1) quotes Severian’s
statement that Jesus’ words must be distributed so as to take some of them
to be according to his dignity, others to have been meted out corresponding
to our weakness. Then he chooses (as section 2) the paragraph about Jesus
hiding his dignity to the Jews, to whom he only spoke about the dignity of
the Father. Yet (section 3), “Son” is a name pointing towards both the divine
15 Zellinger, Studien, 81ff.

16 Zellinger, Studien, 170f
and the human in Christ. 1. John 1,7 points out that the Son participates in the
sufferings of the body. Human death is death of the body, but in Matth. 10,17f
it is not specified that it is only the body that dies. Correspondingly He, who
knows everything because of his divinity, can because of his human nature be
described as being in need of knowledge, when the word is uttered according
to the οἰκονομία (Arm. տնաւրէնութիւն), not according to τὸ τῆς θεότητος
ἀξίωμα (Arm. աստուածութեան պատիւ). This then is the textual key to the
fact (section 4) that the glory of the Son remained hidden to the Jews.
This homily on John 7,15 was attributed to Severian by B. Marx.17 This
attribution was supported by H.-D. Altendorf,18 and to the best of my knowl-
edge no objection has been raised to Severian’s authorship. Thus, with the
supplementary external evidence of the Galata cento this homily can to-day be
considered one of the texts most securely belonging to the bishop of Gabala.
The amount of variants in relation to the Migne text is very modest.

In crucem Domini et in Spiritum Sanctum
et in Trinitatem (CPG 4196)
This very short quotation finds its parallel in the following lines in Migne:
(PG 56,504, l. 61 – 505, l. 6): Πλὴν ὅταν ἀκούσῃς σταυρὸν … ἀξία δὲ Θεοῦ
δι’ ἑαυτόν.
Once again the principal topic of the cento is presented; here the double
formula of οἰκονομία and ἀξία (ἐξουσία) is applied to the awe which we owe
the suffering and crucified Lord. The final phrase unites the two elements: the
cross is (there) for our sake, God’s dignity is (there) for his own sake.
External evidence for Severian’s author’s rights to this homily was not
available until 1960, when J. Kirchmeyer pointed to the name of the author as
given in MS Sin. gr. 493.19 On the other hand internal arguments at an early
time of Patristic scholarship led to the right attribution. The historical names
to be mentioned are Sirmond and Montfaucon, whose arguments were taken
17 B. Marx, “Severiana unter den Spuria Chrysostomi bei Montfaucon-Migne”, Orientalia Christiana
Periodica 5, 1939, 281-367, see esp. 309-315.
18 H.-D. Altendorf, Untersuchungen zu Severian von Gabala, Tübingen 1957, 146-148.
19 J. Kirchmeyer, “L’homélie acéphale de Sévérien sur la croix dans le Sinaïticus gr. 493”, Analecta Bol-
landiana 78, 1960, 18-23.
up and expanded in the 20th century, first by J. Zellinger,20 and later on by A.

Wenger21 and H.-D. Altendorf.22 Thus, here the Galata cento can be said only
to strengthen a strong case.
The few lines of this fragment exhibit very few variations between the
Greek and the Armenian, but the difference of title should of course be noted,
the Galata cento having the title quoted above – as compared with the tradi-
tional one: De serpente, quam Moyses in cruce suspendit in deserto, deque divina
trinitate.
6. Two quotations from the homily

In verbum: Confiteor tibi, Pater Domine caeli et
terrae, et in visionem Danielis (CPG 4295,17a)
Quantitatively speaking, these two quotations represent more than half of the
text of the cento. As, in addition, only a small fraction of this homily is known
beforehand, viz. from a quotation in Syriac to be found in Severus of Antioch’s
Liber contra impium Grammaticum23 and covering only a small fraction of the
text of the Galata cento, G is here a witness of particular importance.
The centonist of G harmoniously unites these quotations with the preceding
half of the cento. Having used earlier on quotations concerned with Christ’s
sufferings and “ignorance”, and having shown that these do not detract from
his divinity, the question is now about Jesus praising the Father as Lord of
Heaven and Earth. Is such a praise (or confession – cf. the confiteor of the title,
Arm. խոստովանիմ, Gr. ἐξομολογοῦμαι, – Matth. 11,25; Luke 10,21) not sign
and proof of subordination?
The answer is no, because again this subject must be seen in the light of
the economy of salvation. First, Jesus’ praise of the Father must be understood
as an example set for man(kind). Secondly, it must be remembered that the
economy of salvation is prepared for the benefit of man. The source of im-
mortality did not need a resurrection; the Saviour did not need a human body,
20 Zellinger, Studien, 27-34.

21 Cf. A. Wenger, “Hésychius de Jérusalem”, Revue des Études Augustiniennes 2, 1956, 461.
22 Altendorf, op. cit., 253f.
23 Severi Antiocheni Liber contra impium Grammaticum 3,2, CSCO 101, Scriptores Syri 50, 1933 (Textus),
CSCO 102, Scriptores Syri 51, 1933 (1965) (Versio). Cf. A. Wenger, “Une homélie inédite de Sévérien
de Gabala sur le lavement des pieds”, Revue des Études Byzantines 25, 1967, 219-234. Apart from the
identification and publication of the Greek homily from which one of Severus’ quotations is taken,
Wenger comments on the four Severian fragments quoted by Severus on pp. 222ff.
nor a baptism of purification. It was man(kind) and nature who were in need;
baptism was needed to purify, first the river of Jordan, and then (Christian)
man.
The next question raised concerns Jesus’ word in John 11,25: “I am the
Resurrection”. Is there not a problem in Jesus’ uttering this word before his
resurrection, and is there not a contradiction with Paul’s word in Gal. 1,1 about
the Father having raised the Son from the dead?
In fact, heretics have argued to that effect. The hermeneutic clue to the
understanding of Paul must, however, be found in 2. Cor. 13,3; and as ac-
cording to that apostolic word Christ speaks through apostles, we are once
again led to the understanding that in the Bible certain passages are coined
with relation to the incarnation (Arm. ի մարմնալոյն), whilst other passages
elevate us to the level of the divine glory (Arm. ի պատիւ աստուածութեանն
(G, p. 381, ll. 5f).
The perspective, in which Jesus’ praise and confession should be seen, is
the right distribution of the sayings of Jesus – between the divine and the hu-
man (Arm. տես զիարդ բաժանէ որպէս զինքն որոշէ, G, p. 381, l. 13). Neither
should Jesus’ prayer about knowledge of God (cf. CPG 420124) induce us to
forget his power (Arm. զաւրութիւն).
The second quotation starts out from testimonials of the Old Testament.
First Ps. 22(21),10 is seen in combination with Is. 49,1, the sequence of which,
viz. Is. 49,3 is then combined with the New Testament key text of Phil. 2,6f. The
texts quoted are all about God, the Father, and about the divine birth which
does not take place by virtue of a law of nature or a process of nature; neither
does it add to or subtract from the eternal divinity of the Father – and the Son;
and when, in Is. 49,1-3 and Phil. 2,6f, the word “servant” is used, this is about
a function, which the Son takes upon him – in order to fulfill the economy of
salvation.
The next composition of Old and New Testament testimonials is taken from
Heb. 1,7-9 and the Psalm quotations and allusions used there.25 The words
about ἔλαιον ἀγαλλιάσεως (Ps. 45(44),8; Heb. 1,9) refer to the Holy Ghost, but
the rest of the passage of Hebrews is concerned with the Son being superior
24 Cf. above, p. 221f.

25 It seems that this is the text to be preached about on the Sunday in question. This would be the
natural understanding of Arm. այլ ի յառաջի կայ եկեսցուք (but let us get to what is in front (of us),
G, p. 381, l. 33).
to any creature, including the angels; and both David and Isaiah testified to
the eternal supremacy of the Son (Ps. 74(73),12; Is. 6,1.5).
Heretics have voiced doubts as to whether the Biblical testimonies (in
particular Is. 6, maybe) should be taken about the Son. This is repudiated
through a reference to John 12,37f.40f. The supremacy of the Son is from eter-
nity, but he can also be said to be given the supremacy, viz. that belonging
to the human economy of salvation (Arm. զարքայութիւնն … զմարդկային
տնաւրէնութեանն), and that is what the Jews do not understand.
Priorities can also be set in terms of age and time. A particular comple-
mentarity between such priorities and simultaneity relevant for language to
be used about God can be taken from Gen. 2 when explaining, how Adam is
older than Eve, because he was created first, and at the same time simultane-
ous with his woman, as she was made from a rib of his body.
Transferred to the relationship between the Father and the Son, on one
hand the Son described as Λόγος and Χριστός is “younger” or “later” than
the Father. But it should be noted that in the logos-text par excellence: the pro-
logue of the Gospel of John, the “simultaneity” is the first thing to be shown:
the Son was God from eternity, being “with the Father”. The image of Adam
who could not possibly have had an offspring without his rib in the shape of
a woman can support our understanding of the word that the Father did not
create one thing without the Son (John 1,3).
The image of Adam can be used in three further ways:
1. Adam and Eve being two persons of one nature (or essence) (Arm.
երկու անձինք եւ մի բնութիւն) corresponding to the central christological and
trinitarian doctrine.
2. The unity as expressed through the woman being with Adam can be
seen in parallel to the formula “God with God”.
3. The “superiority” of the Father being the head of the Word can be seen
in parallel to the description of Adam as the head of the woman.
According to the body Christ is the head of man and of the same substance
as man; he is therefore like us, because he is “of us”, and he is like the Father,
because he is “of the Father”.
Below, Severus’ quotation (in Lebon’s Latin translation = L) shall be set out in
parallel with the Armenian text of G.26
26 Words in G abbreviated according to usual practice in Armenian manuscripts are rendered in their
unabbreviated form.
G L
(p. 379, l. 33 – 380, l. 14) (p. 237, l. 21-33)
եւ եթէ ոք հաւաստեաւ կամիցի ասել եւ լռել, Et si quis accurate velit et loqui et audire:
Քրիստոս վասն իւր ոչ եկն, ոչ չարչարեցաւ ipse Christus pro seipso neque venit, neque
եւ ոչ յարեաւ, քանզի ոչինչ ի յայսցանէ պետս passus est, neque surrexit: nullo enim ex
անէր աղբեւր անմահութեանն եւ ոչ ինչ նմայ his indigebat fons immortalitatis; neque
ամբարի, այլ ամենայն մեզ շնո(ր)հի. երեւեցաւ sibi quidquam lucratur, nam omnia nobis
յաղագս մեր իբրեւ զմարդ. ոչ մարդկային donat. Apparuit propter nos ut homo,
մարմնոյ կարաւտեալ, այլ զի զմարդկան non humana carne indigens, sed naturam
բնութիւն ըստ անայցէ վերստին, մկրտեցաւ humanam recuperans. Baptizatus est in
ի յորդանան. եւa) զմեղս ի բաց դնելով, այլ զի Iordane, nona) peccata ponens, sed naturae
ի բնութիւն ջրոցն զսրբութիւնն խառնեսցէ. եւ aquae sanctificationem immiscens; neque
ոչ զի յայլմէ ի վեհագոյն զաւրութենէ սրբեսցի. praestantiori virtute sanctificabatur, sed
այլ իւրով իսկ յեւով մարդկութեանն, թեպէտ եւ in ipsa humanitatis specie, etsi baptizarib)
թուէր մարդկութիւն ընդունելb) ինքն ինքեամբc) videbatur, ipse seipsumc) sanctificabat. Ipsius
սրբէ, քանզի նորին է ձայն մեծի թագաւորին, enim magni Regis vox est: “Et pro eis ego
եւ յաղագս սոցա սրբեմ զանձն, զի իցեն սոքայ sanctifico meipsum, ut sint et ipsi sanctificati
սրբեալք ճշմարտութեամբ նորին աղագաւ in veritate”. Ideo etiam patitur, ut passionem
եւ ճարճարի, զի զմեր ճարճարանսն լուծցէ, nostram solvat; ideo reviviscit, ut mortem
յաղագս նորին եւ յարեաւ, զի զմահ լուծցէ, solvat; ideo resurgit, ut nos suscitet.d'
վասն նորին յարեաւ ի մեռելոց. զի զմեզ
յարուսցէ.
In general, the Syrian and the Armenian texts are very closely related. How-
ever, the following details and variant readings should be noted:
(a) Here it would be natural to correct the Armenian according to the Syrian,
adding: non.
(b) Here – as compared with baptizari – the Armenian has “assume (accept)
humanity” (մարդկութիւն ընդունել).
(c) Instead of the accusative case: seipsum (object), the Armenian has the in-
strumental case: ինքեամբ.
(d) Here Lebon adds the note: Locum non repperi (presumably referring to the
“quotation” marked a couple of lines before the end of the passage). I sup-
pose that further searching for the “source” would yield no result, as I take
the “quotation” to be one of the examples, where Severian’s rhetoric style
contains an element of “dramatizing” through the use of “direct speech”,
even where there is no source for the “line” in question.27
Neither of these differences points towards two different Greek texts, one
27 Cf. Per Piscatores, e.g. pp. 13, 292ff, 336ff.

behind each of the versions, but it would be particularly interesting to know

the Greek lying behind the difference in (b).
As to internal arguments for Severian’s author’s rights I would mainly
point at the following features:
First, it has been mentioned already that the passages quoted fit in very
well in the main theme of the whole cento, and we have also pointed to the
lines of connection with the homily CPG 4201.
Secondly, e.g. John 2,19 with its temple imagery is one of Severian’s favou-
rite “Antiochene testimonials” for the relation between divine and human in
Christ, cf. e.g. PG 52,782 (CPG 4187); 52,830 (CPG 4189); 56,501 (CPG 4196);
59,672 (CPG 4203); Zell., Stud. 14f (CPG 4215).
Correspondingly, 2. Cor. 13,3 is Severian’s favourite hermeneutical key
to opening the door between the apostolic message and Jesus’ own words in
the Gospels, referred to e.g. PG 49,323 (CPG 4186); 56,423 (CPG 4193); 56,473
(CPG 4194); 56,556 (CPG 4198); 59,700 (CPG 4204).
The combination of Is. 6,1ff and John 12,37ff (G, p. 382) taken to demon-
strate that the Son was king of eternity, i.e. even before the incarnation, can
be found in a close parallel, PG 55,605f (CPG 4190), whereas PG 56,510 (CPG
4196) is related, but slightly different, in so far as it takes Ps. 74(73),12 (cf. G,
p. 382, ll. 13ff) to be about the kingdom of the Father, Luke 1,32 about the
kingdom of the Son, and Is. 6,1ff to be about the kingdom of the Holy Ghost.
The difference between the two Greek parallels is not uncommon in Severian;
it seems to belong to the “homiletic freedom” to be expected,28 and the general
character and contents of the passage in the Armenian quotation to my mind
strengthens the case for Severian’s paternity.
Therefore, besides the external evidence in Severus and the Galata cento,
there are a number of internal testimonies in the quotations of the cento in
the form of lines of connection with well-known Severian homilies. On the
other hand, the fascination of this text, of course, lies in the additional mate-
rial it contains. Thus e.g., to the best of my knowledge the interesting use
of an Adam-Eve typology to describe how in the Trinity the relation can be
determined both in terms of “before-after” and of simultaneity – in parallel
with the sequence of the elements in the prologue of John, is not to be found
in any of the Severian texts known so far.
28 Cf. e.g. L 1995, 227 [above, p. 168], n. 40.

reek and Syrian
under the Aspects of some Syrian Seats of Learning
G
1. Introduction
It is well-known that, whilst from the 2nd century and through later centuries
an important specific institution existed in Alexandria, a “school” in which
much of the theological work of the Christian church took place, the so-called
“School of Antioch” cannot be identified as a parallel institution in the patri-
archal city of Syria. Here, when we use the word “school”, we are speaking
rather about a “movement”, a “tendency”, perhaps a “method”, more or less
common to a number of theologians, identifiable especially in the 4th and
5th centuries.
On the other hand, we are told that the disciples of Lucian the Martyr, when
writing to one another, used the greeting “dear Syl-lucianist”, which might be
translated “dear schoolmate of the school of Lucian”. But it should be borne in
mind that it is still, historically, a matter of dispute whether a certain Lucian
(Lucian the Martyr or another figure) could rightly be called “founder of the
School of Antioch”.
We shall not go further into questions of origin of Antiochene theology and
exegesis, but recollect that Diodore of Tarsus is said to have established a so-
called asceticon, frequented by John Chrysostom and Theodore of Mopsuestia,
and it should also be noted that through research of the later decades of the
20th century it has become ever more evident that there was some kind of
teacher-student-relation between Eusebius of Emesa and Diodore of Tarsus.
So even though it may be difficult, or even impossible, to draw one single
line of “school tradition” from the Antiochene authors of the early 4th century
to the great representatives of “Antiochene theology” of the 5th, there are a
number of reports about teaching relations in various contexts. At the outset,
however, it seems to me to be necessary to emphasize two points: 1) It might
be appropriate to use the plural form: “schools” or “seats of learning” rather
than the singular. 2) Such “schools” or teachers, as can be identified, are not
necessarily located in the city of Antioch.
The majority of the authors to be referred to in this article used the Greek
language as their vehicle of communication and preaching. On the other hand,
it seems true to say that during the last couple of decades more attention than
previously has been paid by researchers to the character of the links connecting
them with Syrian culture and language, the churches of Syria, and the Bible
in Syriac.
In accordance with the linkage made in the titles of two contributions to the
relevant discussions, by R. ter Haar Romeny and Lucas Van Rompay, respec-
tively, viz. A Syrian in Greek Dress (about Eusebius of Emesa) and “Antiochene
Biblical Interpretation – Greek and Syriac” (about the exegetic “School of
Antioch” in general), I have chosen the broad title “Greek and Syrian” for the
considerations presented here, adding the plural about the “schools” (or seats
of learning), from which aspect I shall try to consider some of the important
scholarly contributions, mainly those of the 1990s.
Instead of undertaking to give a full overview of the history of the research
of later years, I have taken the book and article mentioned above (both pub-
lished in 1997) as my starting point for the following remarks about charac-
teristic features of teachers in the “schools” of Antioch (or Syria). Because of
Haar Romeny’s important book, Eusebius of Emesa will necessarily call for
a particular amount of attention; and because of Van Rompay’s article, the
question of Ephraem’s relation to Antiochene exegesis must be touched upon
(even if in a more secondary place); in addition I shall add some observations
on Theodoret of Cyrrhus, often referred to as the last great representative of
Antiochene theology (briefly, also on matters Greek and Syrian after 450).
Before turning to these main topics, however, it might be appropriate to
touch upon two supplementary, introductory themes: 1) a particular sugges-
tion of a formula to be used to describe the relation of Antiochene theologians
to Greek culture in general; 2) a particular discussion about lines connecting
Antioch with the Latin speaking world.
2. Alexandria, Antioch, and Greek culture

An interesting paradigm to be used to describe the differences between Alex
andria and Antioch has been suggested by Chr. Schäublin and Frances Young.
They advocated the idea that the Antiochians have taken over much of their ex-
egetic apparatus from the Greek schools of rhetors, whereas the Greek premises
G r e e k a n d S y r i a n 231
of the School of Alexandria should rather be found in schools of philosophers. I

have two reasons to refer to this discussion without extending it or pretending
to be able to balance the matter. One reason is of a general kind, viz. that here
the notion of “school” and actual schools are given distinctive roles in more
than one context – relevant in the analysis of Antiochene exegesis. The second
reason, more particular as it were, is that the discussion referred to has played
a role in works by such authors as Haar Romeny (already mentioned) and
Adam Kamesar, to whom I shall return in order, not least, to call attention to
his notion of a “Greek approach” to the Old Testament – as opposed to other
approaches. This formula has been used by other authors as well, trying to
come to grips with the use of the Septuagint in the Old Church, where atten-
tion is directed both towards its imperfections, considered as a translation, and
towards its authority as part of the canonical Scriptures in Greek, the main
language of the Christian church.
3. Antioch and the Latin speaking world

The main reason for including considerations about authors writing in Latin
is the fact that the topic of Adam Kamesar’s principal monograph is a work
by Saint Jerome, viz. his Quaestiones Hebraicae in Genesim. Before turning to
Jerome and Kamesar I should like to refer to one further line of connection
that has been drawn between Antioch and the West. Here the author of the
Old Church is Junillus Africanus, and the modern writers are Michael Maas
and Edward G. Matthews. I am referring to the book Exegesis and Empire in
the Early Byzantine Mediterranean, bearing the subtitle Junillus Africanus and the
Instituta Regularia Divinae Legis (2003).
The title of the book already suggests that, according to the authors, Junil-
lus’ (or traditionally: Junilius’) Instituta should be viewed as belonging to the
framework of Justinian’s ideology or even theology of the Empire. Heinrich
Kihn, in his more than 100 years old book on Theodore of Mopsuestia and
Junillus (1880), maintained that the author of the Instituta was strongly influ-
enced by Theodore of Mopsuestia. Now Maas (and Matthews) maintain that
Kihn overemphasized the role of Theodore in this respect. On the other hand,
they still see Instituta as built upon the basis of Antiochene hermeneutics in
general; they do not, in any way, try to detract from the picture that Instituta,
mainly through Cassiodorus, had an enormous influence on medieval Latin
exegetes, who were thus – indirectly – to a certain extent “students” of the
“School of Antioch”.
Adam Kamesar’s book, Jerome, Greek Scholarship and the Hebrew Bible (1993)
is of a different fibre. Here, among other things, in the analysis of Jerome’s
exegetical work on Genesis we are taken into a thoroughgoing scrutiny of his
sources. Chapter 5 of the book has the title: “Jerome and his Greek Exegetical
Sources”, and Kamesar here refers to three categories: 1) Origen’s writings,
2) Liber nominum, and 3) the Antiochene fathers.
Here I shall just quote the most important part of Kamesar’s conclusion as
to Jerome’s relation to “Antiochene” fathers:
At about the time when Jerome began issuing volumes of the new translation, he pub-
lished his Quaestiones Hebraicae in Genesim. This text is not an experimental work, un-
dertaken by a man gradually moving away from the Hexaplaric LXX in the direction
of the ‘Hebraica veritas’. Nor is it merely an attempt to demonstrate the ‘utility of the
Hebrew text’, which had in fact already been recognized by most Greek scholars. Rather,
Jerome goes beyond this objective, and puts forward and justifies the system by which
he interprets the Hebrew text. That is, he defends the philological foundations of IH
[IH is Kamesar’s abbreviated reference to Jerome’s translation iuxta Hebraeos]. Jerome’s
system may be termed a ‘recentiores-rabbinic’ philology. However, this system can only
be appreciated when seen in the light of the other major systems of the time. These are
the ‘Aramaic/Syriac approach’, which may be associated with Eusebius of Emesa, and
the ‘Greek approach’, present in the works of Eusebius and Diodore of Tarsus, but more
fully developed by Theodore of Mopsuestia. If Jerome’s ‘recentiores-rabbinic’ philology
is understood against the background of these other systems, it is possible to see what
he meant by describing QHG as an ‘opus novum’, and to acknowledge that by and large
he was justified in making such a claim. (Kamesar 1993, pp. 193f.)
Obviously, Kamesar considers Jerome to be very well acquainted with various

Greek and Oriental traditions as to exegetical method. From the Alexandrians
and some Antiochene exegetes he knows “the Greek approach”, where the
Septuagint is considered to be (or used as) the basic text. In Eusebius of Emesa
he has met the idea that Syriac – as a “neighbouring language” to Hebrew –
must play a particular role in identifying the textual foundation, “the Aramaic/
Syriac approach”. But Jerome prefers an approach based on Rabbinic philology
and the three recentiores Greek versions: Aquila, Symmachus, and Theodotion
– in order to create the most satisfactory, “new” basis for the translation and
interpretation of the Old Testament.
To my mind these analyses of the sources used by the learned Jerome, cele
brated as vir trilinguis, are very important to keep in mind when approaching
a couple of “cases” from Syria (including Antioch).
4. Three cases from Syrian seats of learning

– whether Antiochene or not
The purpose of the following lines is to sketch the ways in which two authors,
who are both usually considered to represent Antiochene exegesis, approached
and used the Syrian Bible, and what their linguistic background was, espe-
cially with regard to their knowledge of Syriac. Historically speaking, the
two authors are at a distance of about 100 years from one another: Eusebius
of Emesa and Theodoret of Cyrrhus. For these two authors I refer to some
of my own studies as my starting point, whilst for the third case, Ephraem, I
refer to Van Rompay’s “experiment” as to the question of whether he could
be labelled an “Antiochene” (in exegetic matters).
4.1. Eusebius of Emesa – a Syrian in Greek dress

Eusebius was born in Edessa (presumably about 300 A.D.); Syriac, no doubt,
was his mother tongue; he is referred to as bishop of Emesa, but it seems that
he has been in office for a very short span of years, if at all. He is reported
to have been chaplain to the court and forces of the emperor (Constantius II)
and to have paid visits to both Alexandria and Antioch (maybe for reasons of
study). Obviously he belongs to a “school” that is reluctant towards allegorical
interpretation. The year 359 is given as the latest possible date of his death.
In my book Per Piscatores (1975) I made an attempt at evaluating Eusebius’
use of the Syriac Bible. However, as the texts examined were the Armenian
translations of a number of homilies, and as suitable material for comparison,
if at all available was in Latin, there were of course considerable problems of
method and technique in the process of defining a possible Syriac basis for
his quotations from the Bible. To this should be added that the principal goal
of the book was to determine who were the authors of the Armenian series
of homilies in question.
My research found that eight of the thirteen homilies belong to Eusebius,
the rest to Severian of Gabala and, according to my conclusions, a stratum of
Syriac biblical readings was much more evident in Eusebius than in Severian.
(In later articles I have shown further evidence of Syriac readings, not only in
Eusebius, but in Severian as well, – cf. especially L 1996.)
One topic was only touched upon briefly in the Per Piscatores, i.e. the possi-
bilities of identifying further elements of Eusebius’ Syriac Bible and the original
Greek text of his Commentary on Historical Writings of the Old Testament. The
Armenian translation of this text was published by Vahan Hovhannessian in
1980, and in a couple of articles of the early 1980s (L 1984 and 1987) I pointed
out some of the perspectives, that might be opened through further study of
this text.
The honour of having gone deeper into the investigation of this text, how-
ever, should be attributed to R. ter Haar Romeny. As can be seen, I have
borrowed the title of his 1997-monograph as the heading of this paragraph.
However, entering into any thoroughgoing presentation or discussion of
this highly important book would take us too far here. I shall therefore very
briefly emphasize only four elements in order to point out some of the results
gained here as well as some of the perspectives and desiderata opened by Haar
Romeny’s monograph.
1. Of course, for one thing the very title of the book, A Syrian in Greek
Dress points at Eusebius’ origin: born in Edessa, and as mentioned already,
therefore no doubt had Syriac as his language of origin. On the other hand,
his works were written in Greek. Through the ill fortune, however, of the his-
tory of transmission very little of Eusebius’ oeuvre is handed down to us in
its original language. As pointed out above, Latin and Armenian translators,
especially, have the merit of making it possible to-day to sketch, at least, some
of the literary and theological characteristics of Eusebius.
2. The possibility, to which I pointed in 1975 and 1984, of identifying in
catenae considerable parts of the Greek original of Eusebius’ commentary, has
now however been realized by Haar Romeny – in collaboration with Françoise
Petit, the expert on Greek catenae. This is one of the very important results to
be acknowledged in Haar Romeny’s book.
3. As appears from the subtitle of Haar Romeny’s book, he has chosen to
work within two restrictions: 1) he concentrates his efforts on the Genesis part
of the commentary, and 2) in particular, his comments are concerned with
the versions of the Old Testament as used by Eusebius. So, regrettably, even
to-day, more than 25 years after Hovhannessian’s edition, important parts of
Eusebius’ work are still – in practice – unknown to patristic scholars without
a knowledge of Armenian, and the contents and underlying principles of his
exegesis are so far mainly examined in a text-historical perspective. So, how-
ever important and valuable Haar Romeny’s book is, it still leaves us with a
number of desiderata.
4. In a context asking questions about school traditions it should be men-
tioned – as a merit of the book – that it contains a number of references to

some repercussions of Eusebius’ text. In the context of the traditional concept
of the “school of Antioch” the amount of references to Eusebius in Diodore
of Tarsus might be the most important element here, but also the aftermath
which is to be found in Syrian authors, such as e.g. Ishodad of Merw, is of value
for understanding the lines of tradition represented by Syrian theologians.
Presumably further studies could elaborate on that – and on the aftermath in
medieval Armenian theologians and exegetes, as well.
Without going into further detail about Haar Romeny’s (and others’) stud-
ies of Eusebius of the 1990s (and the 2000s) it might be concluded that to-day
we know much more about Eusebius’ “Syriac approach” to the Bible text; but
there are still a number of open questions about the processes that took place
on the way to the investment in Greek clothes and the subsequent appearances
in Latin and Armenian costumes.
4.2. Theodoret of Cyrrhus – a Greek in a Syrian office

In passing, it might be worth noting as a symptom of how broad the range of
cultural and linguistic questions may be, that Theodore of Mopsuestia – well-
known as the “teacher” of the Nestorian Church, which has East Syriac as its
vernacular – was, to Kamesar, one of the obvious representatives of “the Greek
approach” to the Old Testament. Here we shall be concerned with Theodoret
of Cyrrhus who, to my mind, deserves a new reading as to the question of his
vernacular and about his way of mastering problems that might be brought
together under our heading “Greek and Syrian”. It should be noted that Van
Rompay – inspired by Kamesar’s terminology – coins the formula “a more
varied approach” for Theodoret, but his points of view will be commented
upon, most naturally, in the context of Ephraem, where his discussion of Greek
and Syriac in the School of Antioch is treated.
Theodoret was born in Antioch. His date of birth is not certain, but 393 A.D.
is often quoted as the most probable year. He was educated in the monasteries
in and around Antioch. Apart from a brief discontinuation, he was bishop of
Cyrrhus from 423 until his death (466? – again there is a certain amount of
doubt about the exact year). Undoubtedly, Syriac was the primary language
of the region of Cyrrhus, and there is a strong scholarly tradition of assuming
Theodoret’s vernacular as having been Syriac, too. I shall not resume here my
arguments for dissociating myself from this tradition (cf. L 1999), but only
briefly recapitulate a couple of points relevant for the discussion of “schools”
and traditions and for Theodoret’s particular views on matters of languages

and history.
1. In many instances Theodoret shows a keen interest in the “history” and
characteristics of Syriac – and Greek. He considers Syriac to be the oldest lan-
guage (cf. above, pp. 202f and Van Rompay 1986, p. 88f), and he often seeks
the explanation of biblical words for measure and weight and geographic
names etc. in Syriac. But the historical priority of Syriac is not part of any
legend of a “golden age”; to Theodoret Greek is the primary language of the
Christian church, and differences as to language do not reveal any difference
as to “essence” (see above pp. 201f).
2. In the History of the Monks of Syria (or Historia religiosa) Theodoret expresses
a great amount of veneration for the Fathers of the Desert, many of whom
(as is well-known to the author) had Syriac as their vernacular. However,
in his book there is also a strong undercurrent of his consciousness of the
necessity of holding the free-growing, even anarchistic movements of asceti-
cism and enthusiasm under the control and discipline of the church – and of
course quelling any kind of hereticism (the language of which would often
be Syriac).
3. References to “the Syrian” are infrequent in Theodoret. Compared with
Eusebius of Emesa, e.g., the Syriac Bible plays a very restricted role, which
is the more striking, when Theodoret’s numerous references to “the three”
recentiores Greek versions are taken into account.
In this respect his Commentary on Jeremiah (and to a certain extent that on
Ezekiel) stands out. There are no references to the recentiores, and about half
of Theodoret’s references to “the Syrian” (32 out of 61) are found in this work
(cf. above pp. 204f, 206ff).
As, however, there are strong and obvious links between this work and the
rest of Theodoret’s oeuvre, it does not seem advisable to explain the particulari-
ties through a verdict of “spurious” about the Jeremiah-commentary (a pos-
sibility taken into account, but finally rejected (it seems) by J.-N. Guinot).
I have – very tentatively – suggested two possibilities of explaining the
characteristics of this text. Should the explanation be found in a particular
situation at a certain moment of Theodoret’s career, I asked in 1989 (cf. above,
p. 128), and ten years later I asked whether one should rather look for par-
ticular data in the history of the text of the Book of Jeremiah as available to
Antiochenes of the 5th century (cf. above, pp. 208ff). In this context, where
questions of schools and seats of learning are brought into the foreground, I
venture into that realm for a third explanation, once again in the form of an
open question: Could the author of the text be a student of Theodoret, on the
one hand well versed in the exegesis of Theodoret’s “school”, but, on the other
hand, writing under circumstances where the usual instruments of the school
were not available?
So, indeed, there are open questions about Theodoret and the impact of
his teaching, and maybe there are also, still, in spite of the studies by Canivet
and Guinot, lessons to be learned from Theodoret about Syrian matters. But
this should not obscure the picture of this late Antiochene theologian as an
author primarily rooted in Greek language and culture, including a Greek
approach to the Bible.
4.3. Ephraem the Syrian – an “Antiochene” from Nisibis?

As has been suggested already, this is not a “case study” in the same sense
as the above soundings into Eusebius and Theodoret. The only reason to add
these remarks is the fact that Lucas Van Rompay in his analysis of Ephraem
in the 1997 article draws up some lines of perspective highly relevant for the
topics of “Antiochene exegesis” and “schools in Syria”.
Ephraem, a contemporary of Eusebius and celebrated as “der grosse
Klassiker der syrischen Kirche” (to use Altaner’s expression), was born in
306 in Nisibis and active in this city as a deacon and poet (writing in Syriac),
until he was expelled because of the Persians’ capture of Nisibis. During his
last ten years his activities took place in Edessa. His name is thus connected
with the two famous cities of schools in Syria, Nisibis and Edessa. He died in
373.
Ephraem’s curriculum and fate could thus invite us to consider the notion
of “schools of Syria” in a broader perspective. Here we shall restrict ourselves
to a rendering of some of the important points of Van Rompay’s remarks in
the paper referred to above.
Concerning the context of this paper it might be relevant to mention that
it appeared in a report from a Jerusalem conference of 1995 about the inter-
pretation of Genesis by both Christian and Jewish writers in late Antiquity.
In the congress volume there are, indeed, valuable studies of themes that are
of direct relevance to our considerations here, such as Haar Romeny’s treat-
ment of Eusebius of Emesa and his relations to the “origins of the Antiochene
School”, and J. J.S. Weitenberg’s examination of the language in the Armenian
translations of Eusebius; also Judith Frishman’s paper about Gen 1-5 in early
East Syrian exegesis is worth attention for the understanding of early Nestorian
interpretation of the Bible.
As space forbids me to expand on such topics, we shall concentrate on

Van Rompay’s article which is of interest for two reasons: first because of
his well-informed and well-balanced introductory overview of Antiochene
exegesis, second, because of what I should like to call his “experiment” or
“test” attempting to identify Antiochene elements in Ephraem.
His way into this experiment is determined by the topic of the Jerusalem
conference. As mentioned already this conference concentrated on the inter-
pretation of Genesis. However, presumably for accidental reasons within the
history of transmission, little has survived from the later “great” Antiochenes
about this part of the Bible. Therefore Van Rompay chooses to “focus on two
fourth-century commentaries that are more loosely connected with Antiochene
exegesis, i.e. those by Ephrem the Syrian and Eusebius of Emesa”. He con
tinues: “These writings – in both their similarities and differences vis-à-vis other
Antiochene works – may help us to understand what is typically Antiochene
and what is not. Moreover, any study of the Syrian and the Jewish connections
of Antiochene exegesis must include an assessment of these commentaries”
(Van Rompay 1997, p. 104).
Now, Van Rompay compares Ephraem and “proper” Antiochenes on a
number of parameters, in particular concerned with christological and typo-
logical interpretation of narratives of the Old Testament, but also about types
of commentaries etc. Space unfortunately forbids me from discussing singular
analyses, however interesting and important, therefore I shall immediately
proceed to Van Rompay’s conclusion, where he accentuates, that
the distance between Ephrem’s Commentary and the works of the Antiochenes is quite
considerable. On the basis of Origen’s philological work, and in reaction against the al-
legorical interpretation of that same Origen, the Antiochenes developed a strict method
of exegesis which concentrated upon the historical events of the Bible. They introduced
the idea that there is a historical development within the history of salvation, governed
by God’s all-encompassing plan (or “Economy”). Within this Economy, they viewed the
Old Testament history as one phase in this progressing plan, a phase which does not
derive its meaning from a subsequent phase, but rather should be evaluated within its
own limits and in its own right. Very little of this concept, which is characteristic of the
Antiochene tradition of exegesis, is to be found in Ephrem’s exegesis (op. cit., p. 122).
Instead of declaring the “experiment” to have failed, it might be a more ap-

propriate and precise conclusion to emphasize – with Van Rompay and others
– that Antiochene exegesis is not monolithic, and what is true about Antiochene
exegesis, might – per definition – be all the more true about exegesis in the
“schools” of Syria, more broadly speaking.
Van Rompay balances his conclusions in the following statement:
Ephrem, Eusebius of Emesa, and Diodore of Tarsus were contemporaries, and all three
had connections with the Syrian world, a bilingual region with diverse cultural traditions.
It was in this context that Christianity had to define its identity, vis-à-vis Greek pagan
culture as well as Judaism. Ephrem on the one hand, and the Greek Antiochenes on the
other, represent two different models of response to this challenge. In the fluctuating
world of the fourth and fifth centuries, neither of these models was very successful. The
strict Antiochene line, reaching its climax in Theodore of Mopsuestia, became linked to
the christological discussions and quickly lost ground, making way for a more varied
approach, whose representative may be seen in Theodoret of Cyrrhus, among others
(op. cit., p. 122f).
A quotation from R. P. C. Hanson, the patristic scholar, is added. He describes

the exegesis of the early centuries as “the Church’s dance with the Bible, fanci-
fully perhaps, but not irresponsibly, perhaps erratically, but at least gaily”, on
which Van Rompay remarks that “with regard to the Syro-Antiochene region
of the fourth and fifth centuries, the music dictating their movements and steps
was coming from different directions” (op. cit., p. 123).
To remain in this imagery it might be appropriate to add that differences
of melodies and instruments contribute to the fascination of the investigation
into various musical themes, and that e.g. the contrapuntal relation between
Greek and Syrian still leaves the researcher with open and intriguing ques-
tions.
5. After 451. Concluding remarks

My concluding remarks will only consider a couple of desiderata and questions
left open, and refer to a couple of lines of perspective for the centuries after
Theodoret – or after the Council of Chalcedon, where an attempt was made
to settle the theological strife between Alexandria and Antioch.
Two very concrete desiderata should be mentioned: 1) that of a full transla-
tion of Eusebius’ Commentary on Historical Writings of the Old Testament into a
language that would make it possible for patristic scholars in general to include
this important text in their studies; 2) a new critical edition of Theodoret’s
Commentary on Jeremiah and thorough investigation into the textual forms of

the biblical lemmata and quotations.
Around the time after Theodoret, when Antioch was perhaps not remem-
bered as a central seat of learning in the same manner as had been the case for
the previous 150 years, it might be important to recall that bilinguality in the
region of Syria, of course, did not cease. About the topic “Greek and Syrian”,
just two important features of the following centuries should be mentioned
– in great brevity. First, I am thinking of the poetic inspiration brought from
Ephraem into Greek ecclesiastical poetry, as represented above all by Romanos
Melodos. Second, it might be appropriate to mention the phenomenon of phil-
hellenism on Syrian ground (and the parallel phenomenon in Armenia).
It is the more appropriate to be brief here, as these fields of study have
been covered most brilliantly by Sebastian Brock. I can refer to such titles as
“Greek into Syriac and Syriac into Greek”, “Aspects of Translation Technique
in Antiquity”, and “From Ephrem to Romanos” – and to the titles included in
the list of literature below, where e.g. the topic of “Syriac attitudes to Greek
learning” is treated under the heading “From Antagonism to Assimilation”.
I have added a few notes on some of the relevant topics under a heading
about “the (absent?) Middle Ages of the Middle East” (Lehmann 2005,3 – in
Danish, cf. below), taking my point of departure in Theodoret’s (and earlier
Christian) version(s) of the antagonism between Greek and Barbarian (cf.
above, pp. 199ff). I shall not prolong the discussion here, only add that the
questions of bilinguality, especially when one of the languages is that of the
occupying empire, is highly important for the understanding of the establish-
ment of churches, “nation”-building and theology in the centuries from 451
onwards.
In this perspective – and in the imagery of Hanson and Van Rompay – the
cases of the fourth and fifth centuries could perhaps be described as having
the function of preludes to the richly developed music of later times, where
it turned out to be important for nations and churches from Georgia in the
north to Ethiopia in the south to emphasize the fact that they had a language
of their own.
Literature
Brock 1982: Sebastian Brock, “From Antagonism to Assimilation: Syriac Attitudes to Greek Learning”,
Nina Garsoïan, Thomas Matthews and Robert Thomson, eds., East of Byzantium: Syria and Armenia
in the Formative Period, Washington D. C. 1982, 17-34.
Brock 1983: Sebastian Brock, “Towards a History of Syriac Translation Technique”, Orientalia Christiana
Analecta CCXXI, Rome 1983, 1-14.
Brock 1985: Sebastian Brock, “The Christology of the Church of the East in the Synods of the fifth to
early seventh centuries”, Aksum-Thyateira: a Festschrift for Archbishop Methodios, ed. G. Dragas, Lon-
don 1985, 125-142.
Frishman 1997: Judith Frishman, “Themes on Genesis 1-5 in Early East-Syrian Exegesis”, Frishman &
Van Rompay 1997, 171-186.
Frishman & Van Rompay 1997: Judith Frishman and Lucas Van Rompay, eds., The Book of Genesis in Je-
wish and Oriental Christian Interpretation, Traditio Exegetica Graeca 5, Louvain 1997.
Guinot 1995: Jean-Noël Guinot, L’exégèse de Theodoret de Cyr, Théologie historique 100, Paris 1995.
Haar Romeny 1997,1: R. B. ter Haar Romeny, A Syrian in Greek Dress. The Use of Greek, Hebrew, and Syriac
Biblical Texts in Eusebius of Emesa’s Commentary on Genesis, Louvain 1997.
Haar Romeny 1997,2: R. B. ter Haar Romeny, “Eusebius of Emesa’s Commentary on Genesis and the
Origins of the Antiochene School”, Frishman & Van Rompay 1997, 125-142.
Hovhannessian 1980: Eusèbe d’Emèse. I. Commentaire de l’Octateuque, préparé par P. Vahan Hovhannes-
sian, Mekhitariste, Bibliothèque de l’Académie arménienne de Saint Lazare, Venise-St. Lazare 1980.
Kamesar 1993: Adam Kamesar, Jerome, Greek Scholarship, and the Hebrew Bible. A Study of the Quaestiones
Hebraicae in Genesim, Oxford Classical Monographs, Oxford 1993.
Kihn 1880: Heinrich Kihn, Theodor von Mopsuestia und Junilius Africanus als Exegeten, Freiburg im Breis-
gau 1880.
Lehmann 1975: Henning J. Lehmann, Per Piscatores. Studies in the Armenian version of a collection of homi-
lies by Eusebius of Emesa and Severian of Gabala, Aarhus 1975.
Lehmann 1982-2005: Cf. above: L 1982-2005 [se p. 11f].
Lehmann 2005,3: Henning Lehmann, “Havde Mellemøsten en middelalder?”, Dansk Teologisk Tidsskrift
68, Frederiksberg 2005, 46-64.
Maas 2003: Michael Maas, Exegesis and Empire in the Early Byzantine Mediterranean. Junillus Africanus and
the Instituta Regularia Divinae Legis, Studien und Texte zu Antike und Christentum 17, Tübingen 2003.
Pedersen 2004: Nils Arne Pedersen, “Kristendom og skole i oldtidens Antiokia”, Dansk Teologisk Tids-
skrift 67, Frederiksberg 2004, 1-14.
Schäublin 1974: Chr. Schäublin, Untersuchungen zu Methode und Herkunft der antiochenischen Schule, Theo
phaneia 23, Köln-Bonn 1974.
Schäublin 1992: Chr. Schäublin, “Zur paganen Prägung der christlichen Exegese”, J. van Oort und U.
Wickert, eds., Christliche Exegese zwischen Nicaea und Chalcedon, Kampen 1992, 148-173.
Van Rompay 1986: Lucas Van Rompay, ed., Le Commentaire sur Genèse-Exode 9,32 du Manuscrit (olim)
Diyarbakir 22, CSCO 484, Scriptores Syri 206, Louvain 1986.
Van Rompay 1997: Lucas Van Rompay, “Antiochene Biblical Interpretation. Greek and Syriac”, Frish-
man & Van Rompay 1997, 103-123.
Weitenberg 1997: J. J. S. Weitenberg, “Eusebius of Emesa and Armenian Translations”, Frishman & Van
Rompay 1997, 163-170.
Young 1989: Frances Young, “The Rhetorical Schools and their Influence on Patristic Exegesis”, R. Wil-
liams, ed., The Making of Orthodoxy. Essays in Honour of Henry Chadwick, Cambridge 1989, 182-199.
Indices
Biblical texts
Genesis 12,17 82. 88

1-2 23 14,18.20 88
1‑5 237. 241 15,2 88
1,1 115f 15,8f 88
1.2 23ff. 87. 113 15,15f 88
2 225 17,5 89
2,6 111 17,14 89. 109. 116
2,8 111. 116 17,17‑19 89
2,8f 109 18,19 89. 116
2,21f 167f 18,21 89
2,23 115f 18,27 89. 116
2,23f 116 19,21ff 109
3,5 116 19,22 116
3,22 111. 116 20,2 89
4,1 116 20,3ff 89
4,4 114ff 20,17f 89
4,4f 116. 118 21,14 89
4,5 116 21,22 89
4,7 116 22,1 90
4,12 116 22,12 90. 116
4,15 116 22,13 116
4,24 116 23,4 90
4,26 116 23,6 90. 116
5,3 109. 116 23,15 90. 116
6,3 25 24,2 90. 109. 116
6,5 116 24,5 90
6,6 116 24,31 116
6,13 116 24,49 90
6,14 116 24,50 90. 116
6,19f 116 24,63 90. 116
8,4 113. 116 25,22 90
8,7 116 25,26 90
8,21 116 25,27 90
9,4f 116 25,28 91
11,3 109. 116 25,31 91
11,5 116 26,31 91
11,7 116 26,33 116
11,10 116 26,35 109. 116
12,8 116 27,27 109. 116
27,40 116 17 167f

31,7 116 20,7 116
31,7f 109 23,19 116
31,47 116 29,40 206
32,29f 116
33,13 116 Numbers
36,24 116. 122f 7,3 116
36,31ff 116
37,21 116 Deuteronomy
37,36 116 26,14 116
38,18 116
38,29 116 Joshua
39,2f 108. 116ff. 124f 24,29 115
41,16 116
41.38 25 Judges
41,45 116 6,15 116
43,23 116 12,6 107f. 119. 121f. 124f. 189
45,10 116 15,8 116
45,18 116 20,16 31
45,22 116
49,3f 116 1 Samuel
49,5f 112. 116 2,5 116
49,8f 116 4,8 148
49,23 31 4,21 116
49,27 116 15,11 116
19,13 116
Exodus 20,20 116
1,12 91. 109 20,41 116
1,12f 116 21,5 116
1,20f 91. 116
1,22 91 2 Samuel
2,1ff 91 1,21 116
2,14 91 5,6 116
2,18‑22 82 8,18 116
2,24f 91 20,18 116
3,14 114. 116 20,19 116
3,18 116 24,10‑17 160
4,14 116
4,25 85. 116 1 Kings
4,26 85. 116 1,9 123
5,21 116 1,38f 116
6,3 116 2,5 116
8,19 25 12,10 123
15,10 25
B i bl i c a l t e x t s 245
2 Kings 53,4 52
2,3 123 53,12 52
2,14 116
3,4 116 Jeremiah
18,26.28 203 1,5 150f
1,11 210
2 Chronicles 3,2 210
10,10 123 4,15 212
8,6 127. 208ff
Psalms 12,3 210
2,11f 115f 12,5 127. 208. 210
8,3 15. 160 12,12 210
18,12 116 14,8 210
18,46 116 15,10 127. 208. 210
19,5 115f 17,6 210
22,10 224 21,13 209f
45,8 224 23,4 209
49,2.3.5 168 23,5f 209f
71,6 168 24 206
74,12 225. 227 25,23 212
81,16 206 25,25 209f
118,25 14. 17 26 206
132,7 116 27,19 212
141,7 116 29,26 209f
148,4 206 30,20 210
31 206f
Ecclesiastes 31,2 127. 208. 210
3,17 115 31,21 209f
31,38 210
Isaiah 31,40 212
1,19 151 35 206
5,2 115 37 206
6,1 225 38,14 210
6,1ff 227 39,5 212
6,5 225 41 206
7,14 205 41,5 209f
9,11 213 43,13 212
13,21 211 46 206
13,22 211f 46,14 210. 213
19,18 203 46,17 209f
19,23 203 47,4 213
23,14 205 48,31 210. 213
36 203 48,32 213
49,1‑3 224 48,33 210. 213
49,13 212 3,3 168

49,18 212
49,27 209f Matthew
50,8 211 2,1ff 169
50,39 211f 2,11 52
51,20 127. 208. 211 4,24 52
51,27 212 5,23 52
52,18 211 5,24 52
7,3 43
Lamentations 8,2 218
1,12 211 8,4 52ff
1,15 206. 213 8,16 52
1,22 211 8,23‑27 220
3,16f 211 8,24 74
3,29f 211 9,21 136
4,18 211 9,22 136
9,32 52
Baruch 10,17f 222
1,10 212 10,22 136
11,25 75. 223
Ezekiel 12,22 52
4,11 206 14,35 52
27,12 205 16,25 136
40,7ff 129 17,1 52
45,24 206 17,16 52
18,11 136
Daniel 18,17 149
2,31 15 18,18 72
4,33 15 18,19 72
8,13 129 18,24 52
14,34ff 168 19,53 52
20,21 218
Hosea 21,8f 15
11,1 115f 21,9 19
21,15 19. 160
Joel 21,16 147. 160
2,19 160 22,19 52
22,29 149
Jonah 23,37 26. 149f
1,3 205 24,13 136
3,4 205 25,20 52
26,39 97
Habakkuk 28,19 199
3,2 168
B i b l i c a l t e x t s 247
Mark 6,63 65
1,44 52ff 7,15 222
2,4 52 10,8 178
5,34 136 11,25 224
8,35 136 12,12f 17
9,2 52 12,13 15
10,13 52 12,37f 225. 227
10,52 136 12,40f 225
13,13 136 16,2 52
19,29 52
Luke
1,32 227 Acts
5,14 52ff 1,10 43f
6,41 43 1,16 60
7,50 136 3,12 43
8,48 136 3,18 60
9,24 136 6‑7 169
9,56 136 7,31 43
10,21 223 7,32 43
12,24 43 7,42 52
12,27 43 8,18 52
17,19 136 11,6 43f
18,15 52 13,27 60
18,42 136 17,28 184
19,10 136 21,26 52
19,38 16 27,39 43
20,23 43
23,14 52 Romans
23,36 52 2,17 48
24,51 52 2,23 48
4,17 65
John 4,19 43
1,1 180 5,2 48
1,1‑2 175. 177ff 5,3 48
1,1‑5 181 5,11 48
1,1ff 177ff 8,11 65
1,3 175. 178. 225 9,16 150
1,3‑4 171ff. 178. 181ff 11,18 45
1,4 174f. 177. 183f
1,6ff 181 1 Corinthians
1,10 181 1,29 48
1,29 52 2,31 48
2,19 227 3,21 47f
5,21 65 14,11 201
15,22 65 2 Thess
15,36 65 2,14 65
15,45 65
Hebrews
2 Corinthians 1,7‑9 224f
3,6 64f 3,1 43
5,12 48 5,1 52
7,14 47f 5,3 52
9,2 47f 5,7 52
9,11 50 7,25 64
10,8 48 7,27 50. 52
10,13 48 8,3 52
10,13‑17 47 8,4 52
10,15 48 9,7 52
10,16 48 9,9 52
10,17 48 9,14 52
11,6 200 9,25 52
11,12 48 9,28 52
11,18 48 10,1 52
11,30 47f 10,2 52
12,5 48 10,8 52
12,9 48 10,11 52
13,3 224. 227 10,12 52
10,24 43
Galatians 10,39 65
1,1 224 11,4 52
3,21 65 11,17 52
6,13 48 12,7 52
6,14 48 13,15 52
Ephesians James
1,14 65 1,1 46
1,3 49f
Philippians 1,9 46f. 49
2,6f 224 1,10 46f. 49
3,3 49 1,21 61ff. 134
1,23 42ff. 50
Colossians 1,24 43f
1,16 184 2,3 45
2,11 56
1 Thess 2,13 45. 50
5,9 65 2,14 64
2,21 50. 52f
2,23 57ff
B i bl i c a l t e x t s 249
4,12 64
4,16 47. 49
5,15 64
5,20 64
1 Peter
2,5 52
2,24 52
3,18 65
1 John
1,7 222
Patristic and medieval names and sources
Ambrose 16. 19. 26. 184f Homer 201

Anastasius 15f
Aphraates 28. 132. 149. 196. 199f Irenaeus 67f. 74. 97. 217ff
Apollinarius 77. 81 Išo bar Nun 28
Apostolic Constitutions 132 Išodad of Merw 28. 33f. 235
Augustine 11. 17. 23ff. 87. 113. 184
Jacob Cyrrhesticus 197f
Bardaisan 33. 184 Jacob of Edessa 27
Basil of Caesarea 11. 23. 26ff. 33. 68. Jerome 15ff. 30f. 79. 84. 87. 172. 231f.
87. 96. 104f. 113. 156 241
John Chrysostom 18. 67f. 70f. 137.
Cassiodorus 231 141f. 149f. 155ff. 176. 200. 229
Clement of Alexandria 15f. 18f John Mayragomec̒ i 104
Constantius 79. 233 John of Ōjun 102ff
Cyril of Alexandria 30. 77f. 80. 109. Josephus 123. 126. 205f
212 Julian 199
Cyril of Jerusalem 132f Junillus Africanus 231. 241
Justin 21. 132
Damasus 16 Justinian 231
David (translator) 83
Didache 20 Komitas 102
Didymus 93
Diodore of Tarsus 21. 29ff. 84. 93. 113. Lucian 208. 229
117. 122f. 229. 232. 235. 239
Macedonius 196. 199f
Elia 83 Maësymas 196. 199f
Ephraem 26f. 33f. 77. 84f. 132. 182ff. Mani 184
230. 233. 235. 237ff Marcion 184. 197f
Epiphanius 70. 132. 156. 178 Melito of Sardes 109. 119
Eucherius of Lyons 18
Eusebius of Caesarea 79 Nicetas 15. 21
Eusebius of Emesa 7ff. 13. 21f. 30ff. 67f.
77ff. 95f. 98. 104f. 107ff. 125. 129f. Origen 15. 77. 86. 93. 107. 114. 232.
137. 141. 147. 149f. 158. 162. 176ff. 238
185. 188f. 205. 229f. 232ff. 241
Eusebius of Samosata 26f Patrophilus of Scythopolis 79
Eustathius 26. 29 Philo of Alexandria 15. 86. 93
Eznik 78 Proclus 70. 74. 220f
Procopius of Gaza 30. 32. 34. 77. 81ff.
Hesychius of Jerusalem 75. 223 88ff. 110. 114. 116f
Hilary of Poitiers 16. 19. 109. 182 Publius 199
P a t r i s t i c a n d m e d i e v a l n a m e s a n d s o u r c e s 251
Romanos 240 Theodore of Mopsuestia 34. 126. 130.

184f. 229. 231f. 235. 239. 241
Severian of Gabala 7f. 11ff. 21f. 67ff. Theodoret of Cyrrhus 7. 9ff. 21f. 28. 70.
78. 95ff. 109. 137. 141ff. 155ff. 176ff. 81. 84. 108. 119ff. 125ff. 156f. 187ff.
185. 217ff. 233. 241 230. 233. 235ff. 239. 241
Severus of Antioch 70. 75. 77. 156. 166. Theophilus of Antioch 27f
223. 225ff Theophylactus 18. 184f
Simeon Stylites 196f
Step̒anos (scribe) 83 Vardan Arewelc̒ i 83f
Suidas 18
Symbolum Athanasianum 105. 162 Xenophon 201
Tatian 183. 185
Theodore bar Koni 28
Modern writers
Aalen, S. 15 Canivet, Pierre 187ff. 196. 198. 200.

Adler, A. 18 237
Agaësse, P. 29 Carstens, Pernille 12
Akinian, N. 13f. 20. 67f. 79. 95f. 105. Carter, R. E. 72
137. 141ff. 147ff. 158ff. 176ff. 219 Castro, Leo de 18f
Aland, Kurt 38. 41. 134. 172ff. 182f. Chadwick, Henry 241
185 Chaix-Ruy, J. 23
Aldama, J. A. de 68 Clarke, Ernest G. 28
Altaner, B. 18. 21. 26. 237 Cramer, J. A. 71
Altendorf, H.-D. 68. 70. 75. 96. 104f. Cross, F. L. 67. 141
141. 156f. 222f Crum, W. E. 70. 156
Amand de Mendieta, E. 29
Anger 19 Dalman, G. 19
Aucher, J. B. 11. 13f. 67f. 72. 75f. 95ff. Datema, C. 8. 149. 152. 157
142f. 150. 158ff. 177f. 221 Deconinck, Joseph 31. 84. 113. 122f
Awetik̒ ean, G. 32 Deroy, Louis 62
Awgerean, M. 32 Devreesse, Robert 28f. 30. 81f. 86. 90.
Azéma, Y. 194 109. 111. 116. 125
Dietz, Søren 12
Babgēn, see Kiwlēsērean Dorival, G. 82
Bach-Nielsen, Carsten 12 Dragas, G. 241
Bahrdt, C. F. 107 Drusius, J. 107
Bak, Jørgen Friis 12 Dupin, L. E. 176
Bardenhewer, Otto 27. 29f. 128. 187 Duplacy, Jean 41
Barr, James 57. 61 Dürks, W. 22. 68. 70. 157f. 219
Baumstark, A. 55
Bedrossian, M. 32. 45 Elliott, C. J. 195
Bek, Lise 11 Ensslin, W. 71
Berberian, Haïg 11. 72. 163. 219 Erasmus 16. 174
Blake, R. P. 55 Esbroeck, Michel van 68
Boman, Thorleif 57 Ettlinger, Gerhard H. 70
Brock, Sebastian 40. 240f Eynde, Ceslas van den 34
Brooke, A. E. 86
Buhl, F. 21 Faller, O. 184
Burchard, Ch. 10 Fernández Marcos, N. 193. 195. 202.
Burkitt, F. C. 37f 206
Buytaert, E. M. 31. 78f. 81. 84. 96. 109. Festa, N. 188. 192ff. 201f
112. 179ff Field, F. 86. 107f. 114. 117f. 121. 124f
Fraipont, I. 24
Frishman, Judith 237. 241
Funk, F. X. 21
M o d e r n w r i t e r s 253
Garnier, Jules 26 Kiwlēsērean, Babgēn 73. 97. 217

Garsoïan, Nina 240 Kodal, Dorthe Maria 12
Geerard, Maurice 67. 141. 156. 219 Kohlbacher, M. 10
Gennaro, J. 172 Kouymjian, Dickran 11. 163
Gesenius, W. 21 Kroll, W. 71
Giet, Stanislas 26f. 29. 87
Grabowski, St. J. 24 La Croze 37
Green, G. M. 17 Lagrange, M.-J. 55
Guillaumont, Antoine 28. 34 Lake, Kirsopp 55
Guinot, J.-N. 187ff. 203. 205. 207. 211ff. Lamarche, P. 172f. 175f. 178f
236f. 241 Langkammer, H. 172
la Potterie, I. de 172f. 176
Haar Romeny, R. B. ter 9f. 188f. 205. Laurentin, R. 70
207. 230f. 234f. 237. 241 Lebon, J. 21. 75. 96. 102. 104. 225f
Hakopyan, V. 76 Ledet Christiansen, Jørgen 12
Hanson, R. P. C. 239f Lehmann, Else 12
Harnack, A. von 21 Lehmann, Henning J. 11f. 30. 67. 72.
Helm, R. 31. 79 78f. 87. 95f. 98. 105. 109f. 112f. 125.
Herding, W. 79. 84 129. 134. 137f. 141. 144. 147ff. 152.
Hespel, Robert 70 156ff. 166. 176. 182. 188f. 202. 204.
Heyer, Friedrich 10 207f. 217ff. 233ff. 236. 240f
Hilberg, J. 16. 18 Leloir, Louis 38ff. 44ff. 49f. 53. 56f. 60f.
Hilgenfeld, A. 19 63f. 134. 182f. 191
Hoffmann, G. 21 Leroy, F. J. 220
Holm, Kjeld 202 Leroy-Molinghen, Alice 188. 192. 194.
Hovhannessian, Vahan 30. 77f. 86. 200
109f. 125. 234. 241 Lesinski, M. 10
Hvidbak, Claes 12 Lindhardt, Jan 202
Hvidberg-Hansen, F. O. 12 Lint, Theo van 12
Livingstone, E. A. 11
Inglisian, Vahan 83 Lund, Mary Waters 12
Lyonnet, Stanislas 39f. 55. 57. 60.
Jansens, Cornelius 16. 18ff 148f
Jansma, T. 27. 33f
Jellicoe, Sidney 113 Maas, Michael 231. 241
Jensen, Hans 47 Macler, Frédéric 39f. 54
Jordan, Hermann 67. 73f. 218f Martin, Ch. 97
Joüon, P. 55 Marx, B. 68. 70. 74f. 141. 156f. 220.
Jugie, Martin 105. 162 222
Matthews, Edward G. 231
Kamesar, Adam 231f. 235. 241 Matthews, Thomas 240
Kautzsch, E. 19 McClellan, W. H. 35
Kihn, Heinrich 231. 241 McLean, N. 86
Kirchmeyer, J. 75. 222 Mehlmann, J. 172
Mercati, G. 109 Rahlfs, Alfred 9. 107f. 117ff. 124ff.

Merk, A. 54f 189
Merx, A. 19 Renoux, Charles 97. 217
Metzger, Bruce M. 37ff Rhodes, Erroll F. 40
Miller, Ed. L. 171ff. 178 Richard, Marcel 21
Mitchell, C. W. 184 Richards, W. L. 41f
Molitor, Joseph 57. 61. 64. 134. 191 Rompay, Lucas Van 28. 34. 126. 129.
Montfaucon, B. 75. 107f. 117f. 124f. 202. 230. 233. 235ff. 241
222 Rudberg, S. Y. 29
Morin, G. 18
Morville, Elsebeth 12 Sáenz-Badillos, A. 202
Mussolini, Benito 56 Samuelian, Th. J. 11. 98. 110. 156. 182
Mühlenberg, E. 92 Sargisean, B. 13
Møller, Henriette 12 Schnackenburg, Rudolf 175
Møller, Lars Kærulf 11 Schulze, J. L. 122
Møller Larsen, John 12 Schweizer, Eduard 29. 31
Schäublin, Chr. 230. 241
Nestle, Eberhard 38. 171 Siegfried, C. 15. 19
Nestle, Erwin 38. 174 Sirmond, J. 75. 222
New, Silvia 55 Siwrmelean, X. 32
Nobilius, Fl. 107 Sløk, Johannes 202
Nørager, Troels 12 Smoroński, K. 23. 27. 35
Smulders, P. 109. 182
Olivier, J. 85 Solignac, Aimé 24. 29
Oort, J. van 241 Sparre, Jørgen 12
Opelt, Ilona 15. 18 Sprenger, H. N. 126
Ortiz de Urbina, I. 27. 183 Spuler, B. 84
Outtier, B. 84f Stone, Michael E. 11. 110. 182
Stählin, O. 15
Pamboukdjian, K. 97 Sørensen, Dan Enok 12
Parmentier, Léon 187. 190. 192f. 201f
Pauly, A. F. 71 Teilmann, Katja 12
Payne Smith, J. 20f Teselle, E. 23
Pedersen, Nils Arne 241 Testa, E. 23
Peeters, Paul 187. 191f. 195f. 198 Thodberg, Chr. 12
Peradze, G. 70 Thomson, R. W. 76. 240
Perl, Carl Johann 27 Thyssen, Peter 12
Petit, Françoise 30. 81f. 87ff. 92f. 234 Tillemont, Lenain de 26. 176
Pilgaard, Aage 12 Tonneau, R. M. 27
Porcher, E. 70 T̒ornean, T̒. 80
Quasten, Johannes 187. 194 Uthemann, K. H. 8. 149. 157
Valois (Valesius), H. de 190

M odern writers 255
Vardanian, A. 96 Williams, R. 241

Venables, E. 195 Winkler, Gabriele 132f
Vermès, G. 85 Wissowa, G. 71
Vigouroux, F. 14 Wotke, C. 18
Villadsen, Holger 149 Wutz, F. 15f. 21
Voicu, Sever J. 149. 157
Vosté, J.-M. 185 Young, Frances 230. 241
Vööbus, Arthur 39
Zahn, Th. 15f. 19ff. 172f. 185
Weinreich, Uriel 62 Zakayan, D. 12
Weitenberg, J. J. S. 10ff. 237. 241 Zanolli, Almo 30ff. 76ff. 83f. 110
Wenger, A. 70. 75. 223 Zellinger, Johannes 22. 68. 70. 75. 96f.
Westcott, B. F. 185 102. 104f. 156ff. 168. 176f. 219. 221.
Westermann, Cl. 23 223. 227
Wickert, U. 241 Ziegler, Joseph 208f. 213
Wiles, M. F. 175 Zycha, I. 24
Willems, R. 17. 184
Abbreviations
AJ: Acta Jutlandica
ByZ: Byzantinische Zeitschrift
CCSG: Corpus Christianorum. Series Graeca
CCSL: Corpus Christianorum. Series Latina
CSCO: Corpus Scriptorum Christianorum Orientalium
CSEL: Corpus Scriptorum Ecclesiasticorum Latinorum
GCS: Die griechischen christlichen Schriftsteller
HA: Handes Amsorya
JEH: Journal of Ecclesiastical History
OC: Oriens Christianus
OCP: Orientalia Christiana Periodica
PG (or “Migne”): Patrologiae Cursus Completus. Series Graeca, ed. J.-P. Migne
PL (or “Migne”): Patrologiae Cursus Completus. Series Latina, ed. J.-P. Migne
PO: Patrologia Orientalis
RAC: Reallexikon für Antike und Christentum
RBib: Revue Biblique
REArm: Revue des Etudes Arméniennes
RHE: Revue d’Histoire Ecclésiastique
SeT: Studi e Testi
SSL: Spicilegium Sacrum Lovaniense
TU: Texte und Untersuchungen zur Geschichte der altchristlichen Literatur
ZKG: Zeitschrift für Kirchengeschichte
ZNW: Zeitschrift für die neutestamentliche Wissenschaft und die Kunde der älteren
Kirche

(Henning Lehmann) Students of The Bible in 4th and PDF

Uploaded by

Copyright:

Available Formats

(Henning Lehmann) Students of The Bible in 4th and PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Henning Lehmann) Students of The Bible in 4th and PDF

Uploaded by

Copyright:

Available Formats

Students of the Bible

in 4th and 5th Century Syria

seats of learning, sidelights and syriacisms

© The author and Aarhus University Press 2008

Layout and cover: Jørgen Sparre

ISBN 978 87 7934 991 9

Aarhus University Press

Published with grants from Carlsbergfondet

23 The Spirit οf God upon the Face οf The Waters

37 Some Questions Concerning the Armenian

77 An Important Text Preserved in MS Ven.

107 The Syriac Translation of the Old Testament

125 Evidence of the Syriac Bible Translation in Greek

131 The Question of the Syrian Background of the

141 What Translators Veil and Reveal

171 The So-Called “Absurd” Punctuation in John 1,3-4

187 What was Theodoret’s Mother Tongue?

217 Severian Cento No. 2 in MS Galata 54

229 Greek and Syrian

with the names of C. Datema and K. H. Uthemann) determined to a great

3. Occasion, justification, and disposition of the book

7 In a communication delivered in Oxford at the Fifth International Conference on Patristic Studies in

nection, is to be found in Clement of Alexandria’s Paidagogos, I,V,12,511, where

11 Ed. Stählin, GCS 12, p. 97.

not go into details in Jerome’s further exposition which he substantiates by

advances an interesting etymology for the word osanna. De Castro’s line of

of Gabala. A conclusive settlement of the authorship of the homily – whose

49 G. Dürks, De Severiano Gabalitano, Kiel 1917, p. 60-64.

In his unfinished Commentary on Genesis St. Augustine has a fuller treat-

materni corporis etiam formandis pullis quodammodo adminiculatur per quendam in

on which it is very difficult to give any judgment, as no texts have survived,

17 Cf. e.g. Bardenhewer, IV, 388.

tion of the supposedly well-known and recognized orthodoxy of Diodore’s

36 Op. cit., passim.

Still concentrating on the meaning of the verb Eusebius now goes on to

48 Cf. above, notes 40 and 45.

Another question, of course, immediately suggests itself: that of Eusebius’

If I may be forgiven an exordium that is perhaps slightly too rhetorical, I should

1 Cf. the quotation below (from Metzger, see note 3).

of a purely Greek background of the Armenian Bible9 – is that Lyonnet is

11 Op. cit., p. 306.

work done by the Institut für Neutestamentliche Textforschung in Münster,

16 Richards, op. cit. (cf. note 13), p. 269.

Rendering Texts Number of cases

նշմարեմ Mt. 7,3; Lk. 6,41; Acts 7,31; 27,39 4

նայիմ Lk. 20,23; Hebr. 3,1 2

զգուշանամ Hebr. 10,24 1

պշուցեալ հայիմ Js. 1,23 1

հայեցեալ տեսանեմ Js. 1,24 1

Text Greek phrase Armenian phrase

Acts 3,12 τί ἀτενίζετε զի էք պշուցեալ

Acts 11,6 εἰς ἥν ἀτενίσας κατενόουν յոր պշուցեալ հայէի

Js. 1,23 ἀνδρὶ κατανοοῦντι մարդոյ որ պշուցեալ հայիցի

18 L. cit., note 11.

19 Op. cit., p. 309.

21 Op. cit., p. 312.

Text Greek construction Armenian construction

2,23 ἐν + dat. instr.

5,2 ἐπί + dat. instr.