Anda di halaman 1dari 17

On the Influence of the Septuagint

on the Peshitta
HEIDI M. SZPEK
5522 East Burns Street
Tucson, AZ 85711

WHAT ARE WE DOING when we compare the Peshitta to the Masoretic


Text? E. G. Mathews, Jr., raises this fundamental question in a recent review
of a book of mine.1 Mathews states that

In evaluating a translation of a biblical work one must be reasonably sure of


dealing with a translated text and of having a reasonably close idea of its immediate Vorlage. In the application of her model to the Peshitta Book of Job,
S[zpek] compares the Syriac text of Job prepared by L. G. Rignell (The Old
Testament in Syriac according to the Peshitta Version 2/la: Job [ed. the Peshitta
Institute; Leiden: Brill, 1982]) to the Hebrew MT, presumably as it is printed in
BHS. That the sixth-century Syriac manuscript B.21 Inferiore of the Ambrosian
Library in Milan utilized the early eleventh-century Leningrad Hebrew manuscript B19A is manifestly impossible on chronological grounds. While this observation does not deny that both manuscripts preserve earlier textual traditions,
it serves, nevertheless, to highlight what this reviewer sees as the major weakness
of S[zpek]'s study. (Emphasis my own)
Mathews's further comments clearly show that he gives priority to the
influence of the LXX on the Peshitta as an explanation for congruent readings
that depart from the MT, as against internal considerations.
1
E. G. Mathews, Jr., review of Translation Technique in the Peshitta to Job: A Model
for Evaluating a Text with Documentation from the Peshitta to Job, by Heidi M. Szpek, CBQ
56 (1994) 343-44.

251

252 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


Mathews has thus raised two important questions:
1. What are we doing when we compare a Peshitta manuscript of the
sixth century CE. (found in the Leiden edition) and a MT manuscript
of the early eleventh century CE. (found in the BHS)1
2. Can we justify similar departures from the MT in the Peshitta and the
LXX as the result of anything but the LXX's influence on the Peshitta?
Our response to the first question depends on our understanding of what
a critical edition is, the significance of a manuscript's date, the textual tradition of a manuscript, the value of intra-Syriac variants within this tradition,
and the Hebrew Vorlage which may have been used by the translator(s) of the
Peshitta. A response to the second question requires an argument that can
counter the traditional assumption that similar departures from the MT in the
Peshitta and the LXX are the result of the LXX's influence on the Peshitta,
especially in light of the Greek Church's influence on the Syriac Church from
the fifth century on. This response must also include our understanding
whetherand howthe LXX was used.

I
A CRITICAL EDITION may be diplomatic or eclectic. Both the BHS and the
Leiden Syriac are diplomatic, while the Gttingen Septuagint is eclectic. The
goal of these current critical editions was to correct the major flaw of earlier
editions, namely, that they were so often based on late medieval manuscripts.
The Leningrad Codex B19 A , of 1008 CE., serves as the basis for the
BHS. The reason for the use of the Leningrad Codex was that it "is still the
oldest dated manuscript of the complete Hebrew Bible."2 The critical apparatus "represents a complete revision [of the BHK] abandoning slight variants and less important items of information" while noting "real textual
changes and other more significant matter."3 In the critical edition of the
Peshitta prepared by the Peshitta Institute at Leiden, MS. B.21 Inferiore of
the Ambrosian Library in Milan (= 7al) was used as a base text. This is a
manuscript of the sixth century CE., and it was selected because, like L for
the BHSt it is the oldest complete manuscript of the OT.4
2

K. Eiliger and W. Rudolph, BHS, xi.


Ibid., xii.
4
In the case of Codex Ambrosianus, the date 1006/7 also appears on the manuscript,
but this is most probably the year when the manuscript was presented to a monastery; see
F. E. Deist, Towards the Text of the Old Testament (2d ed.; Pretoria: N. G. Kerkboekhandel
Transvaal, 1981) 147-48.
3

INFLUENCE OF THE SEPTUAGINT ON THE PESHITTA 253


In spite of these dates, we can hardly supposeas Mathews doesthat
when we compare the BHS and the Leiden edition of the Peshitta we are
comparing an eleventh-century MS to a sixth-century one. The date preserved
on a manuscript does not necessarily coincide with the date of its composition; rather, it provides us with its terminus ad quern. The sixth-century date
of the Syriac Codex Ambrosianus and the eleventh-century date on the Hebrew
Codex Leningradensis should be used as evidence not of a late date of composition but only of the time of copying. These manuscripts represent textual
traditions that passed from century to century and, in the case of the Hebrew,
often began with a phase of oral tradition. Thus, more important than a
manuscript's date is the validity of the textual tradition that it preserves.5
While the critical edition of the Peshitta text of Job prepared by
L. G. Rignell for the Peshitta Institute in Leiden has MS. B.21 Inferiore of the
Ambrosian Library in Milan (= 7al) as the base text, a total of forty-four MSS,
including 7al, were collated with variants in MSS as late as the twelfth century
included in the second critical apparatus. In the first critical apparatus are
preserved any original readings differing from the base text where it was
deemed to be erroneous or in obvious need of emendation. This manuscript
evidence spanning the sixth to the twelfth centuries is described as "good" by
Rignell, meaning that
there are no significant differences between the later and the earlier manuscripts.
Furthermore, the P[eshitta] of the Book of Job does not show any real difference
between the West Syrian and Nestorian type of text. [The Peshitta] certainly
received its final form long before the Nestorian schism [c. end of the fifth
century].6
Before this time we have no manuscript evidence for the Peshitta of the
Book of Job. By contrast, for the books of Exodus and Numbers, for example,
we do have a representative of an earlier phase, namely, from the time of the
schism. Using this evidence, Koster was able to distinguish three consecutive
stages in the development of the Peshitta text of Exodus:
1. The text of the earliest attainable stage (fifth century)
2. The text of the stage represented by the other old MSS (sixth to eighth
centuries),7 and designated the BTR (basic textus receptus), which
corresponds closely with the base text of the Leiden edition
5
See E. Wrthwein, The Text of the Old Testament: An Introduction to the Biblia
Hebraica (Grand Rapids: Eerdmans, 1979) 13-14; and more recently Emanuel , Textual Criticism of the Hebrew Bible (Minneapolis: Fortress; Assen/Maastricht: Van Gorcum, 1992) 301$2.
6
L. G. Rignell, "Notes on the Peshitta of the Book of Job," ASTI 9 (1974) 103.
7
The ninth century was determined to be the dividing line between "old" and "young"
manuscripts; see P. B. Dirksen, "East and West, Old and Young, in the Text Tradition of the Old
Testament Peshitta," VT 35 (1985) 468$84.

254 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


3. The standard text, or TR (textus receptus), which emerges in the
ninth century8
This same development pattern was confirmed, for example, by A. P. Hayman for Numbers,9 and by S. P. Brock for Isaiah.10
While it might be tempting to assume this same pattern of development for
other books in the Peshitta of the OT, we must remember that the Peshitta is a
heterogeneous work. Individual books were translated at different times, by
different authors, with a variety of external influences exhibited in varying
degrees within individual books.11 Thus, there is no a priori reason to suppose
that books such as the Peshitta of Job which lack manuscript evidence from
this early but "wild" stage (to use Brock's description) would likewise follow
this pattern.
What information do the textual variants that abound within the manuscript tradition of all books of the Peshitta provide about the Hebrew Vorlage
used by the translator(s) of the Peshitta? Current research has indeed shown
that while textual variants are important for reconstructing the textual history
of a particular book, their value has been overestimated in reconstructing the
Hebrew Vorlage or in determining the translational character of a particular
book. A. Gelston noted this overestimation in reference to the Peshitta of the
Dodekapropheton,12 Dirksen for the Peshitta of Judges,13 Mulder for the
Peshitta of Ezekiel,14 and van der Kooij for the Peshitta of Genesis-Exodus.15
8
M. D. Koster, The Peshitta of Exodus: The Development of Its Text in the Course of
Fifteen Centuries (SSN 19; Assen/Amsterdam: Van Gorcum, 1977) 164-97,527-35; see S. P. Brock
("Text History and Text Division in Peshitta Isaiah," The Peshitta: Its Early Text and History:
Papers Read at the Peshitta Symposium Held at Leiden, 30-31 August 1985 [Monographs of the
Peshitta Institute, Leiden 4; ed. P. B. Dirksen and M. J. Mulder; Leiden: Brill, 1988] 50-51), who
more succinctly discusses Koster's stages set forth here.
9
A. P. Hayman, review of The Peshitta of Exodus: The Development of Its Text in the
Course of Fifteen Centuries, by M. D. Koster, JSS 25 (1980) 263-70.
10
Brock, "Text History and Text Division," 49-80.
1
' The heterogeneity of the Peshitta has long been recognized. See, for example, M. J. Mulder
("The Use of the Peshitta in Textual Criticism," La Septuaginta en la investigacin contempornea: Quinto Congreso de la IOSCS [Textos y estudios "Cardenal Cisneros" 34; ed. . Fernndez
Marcos; Madrid: Instituto "Arias Montano"/Consejo Superior de Investigaciones Cientficas,
1985] 37-53), who reviews past literature from 1685 (Richard Simon) through the early 1980s.
12
A. Gelston, "Some Readings in the Peshitta of the Dodekapropheton," Peshitta: Early
Text and History, 81.
13
P. B. Dirksen, "The Ancient Peshitta MSS of Judges," Peshitta: Early Text and History, 145.
14
M. J. Mulder, "Some Remarks on the Peshitta Translation of the Book of Ezekiel,"
Peshitta: Early Text and History, 171.
15
A. van der Kooij, "On the Significance of MS. 5b 1 for Peshitta Genesis," Peshitta: Early
Text and History, 183.

INFLUENCE OF THE SEPTUAGINT ON THE PESHITTA 255


The variant readings of the Peshitta of Job have likewise been overestimated. The nature of these variants is mainly orthographic (e.g, piene
and defective forms, uncontracted and contracted forms, presence or absence
of syam, interchange of tenses and obvious errors).16 Thus the nature of the
intra-Syriac variants bears witness to the fact that the manuscript tradition
is generally stable between the sixth and twelfth centuries.
What of the Vorlage utilized by the translator of the Peshitta? When
Mathews states that one must have "a reasonably close idea" of a text's
immediate Vorlage before embarking on any type of comparative studies, his
statement clearly implies his belief that the text of the Hebrew Vorlage utilized by the translator of the Peshitta is radically different from the MT. The
prevailing opinion, by contrast, is that the Hebrew Vorlage used by the translator of the Peshitta is quite close to the MT.17 The Peshitta of Job likewise
belongs to this consensus. While departures do exist between the Peshitta
and the MT of Job, anyone who compares these texts will readily see that
most departures are due to confusion of roots, or assumption of a different
vocalization, or such techniques as explicit exegesis. This means that if the
Peshitta used a Hebrew Vorlage different from the one we have today, divergent readings must be truly divergent, not explicable by internal considerations. On those few occasions when departures cannot be explained on internal
grounds, appeal can, and should be made to the possibility of a different
Hebrew Vorlage or of influence from another version, but this is rare in the
Peshitta of Job.
When we compare the two critical editions, the BHS and the Leiden
Peshitta, we are not comparing a text of the eleventh century to a text of the
sixth century, just as we are not comparing an edition printed in 1982 to one
printed in 1974: we are comparing textual traditions by means of the best
available manuscript evidence at hand, manuscript evidence that in the case
of the Peshitta of Job reflects a tradition whose internal variants are not so
significant as to suggest that the text of the Vorlage used by the translator was
so distinct from the MT of the BHS today.
16

Heidi M. Szpek, Translation Technique in the Peshitta to Job: A Modelfor Evaluating


a Text with Documentation from the Peshitta to Job (SBLDS 137; Atlanta: Scholars, 1992) 4;
The Old Testament in Syriac according to the Peshitta Version 21 la: Job (ed. L. G. Rignell for
the Peshitta Institute, Leiden; Leiden: Brill, 1982) viii-xv.
17
See most recently P. B. Dirksen, "The Old Testament Peshitta," Mikra: Text, Translation, Reading and Interpretation of the Hebrew Bible in Ancient Judaism and Early Christianity (CRINT 2/1; ed. Martin Jan Mulder; Assen/Maastricht: Van Gorcum; Philadelphia:
Fortress, 1988) 255-97 esp. 258-59; Mulder, "Use of the Peshitta in Textual Criticism," 40-45,
esp. 44-45; , Textual Criticism of the Hebrew Bible, 152.

256 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


II
ON THE QUESTION of the influence of the Septuagint on the Peshitta,
these observations can be found in current literature:18
1. The influence of the LXX on the Peshitta has been overestimated in
the past (Cook, Magary, Szpek).
2. Congruences may be accidental in nature (Gelston, Morton, Szpek).
3. Congruences may be the result of a common treatment or similar
response to a textual difficulty in the MT (Dirksen, Lane, Magary,
Morton, Schoors, Szpek).
4. Congruences between the Peshitta and the LXX are sporadic (Brock,
Eriksson, Gelston, Szpek, Weitzman).
5. The LXX was often not consulted in very difficult passages (Eriksson, Szpek, Weitzman).
6. Congruences can in many cases be explained by internal considerations (Lund, Magary, Szpek).
7. Only exclusive agreements between the LXX and the Peshitta should
be considered possible evidence of direct influence of the LXX on the
Peshitta (Dirksen, Lund).
How can we use these observations in asking (a) whether the Peshitta is
dependent on the LXX, and if so, (b) how the LXX was used? On the basis
of these observations, I have established six "arguments" that can be used as
18
The observations are collected from Brock, "Text History and Text Division," 49-80;
J. Cook, "The Composition of the Peshitta Version of the Old Testament (Pentateuch)," Peshitta:
Early Text and History, 147-68; P. B. Dirksen, The Transmission of the Text in the Peshitta
Manuscripts of the Book of Judges (Monographs of the Peshitta Institute, Leiden 1 ; Leiden:
Brill, 1972); idem, "The Peshitta and Textual Criticism of the Old Testament," VT 42 (1992)
377-90; Jan-Erik Eriksson, "The Hymns of David Interpreted in Syriac: A Study of Translation
Technique in the First Book of Psalms (Ps 1-41) in the Peshitta" (Ph.D. diss., Uppsala University, 1989); A. Gelston, The Peshitta of the Twelve Prophets (Oxford: Clarendon, 1987);
idem, "Some Readings in the Peshitta of the Dodekapropheton," Peshitta: Early Text and
History, 81-98; David J. Lane, The Peshitta of Leviticus (Monographs of the Peshitta Institute,
Leiden 6; Leiden: Brill, 1994); Jerome A. Lund, "The Influence of the Septuagint on the Peshitta:
A Re-Evaluation of Criteria in Light of Comparative Study of the Versions in Genesis and
Psalms"(Ph.D. diss., Hebrew University, Jerusalem, 1988); Dennis Robert Magary, "Translation
Technique in the Peshitta of the Book of Micah" (Ph.D. diss., University of Wisconsin-Madison, 1995); Antoon Schoors, "The Peshitta of Kohelet and Its Relation to the Septuagint," After
Chalcedon: Studies in Theology and Church History Offered to Professor Albert Van Roeyfor
His Seventieth Birthday (Orientalia lovaniensia analecta 18; ed. C. Laga, J. A. Munitiz, and
L. Van Rompay; Leuven: Peeters, 1985) 347-57; Szpek, Translation Technique in the Peshitta to
Job; Richard Andrew Taylor, "An Analysis of the Syriac Text of the Book of Daniel" (Ph.D.
diss., Catholic University of America, 1990); Donald Morton Walter, "The Peshitta of II Kings"
(Ph.D. diss., Princeton Theological Seminary, 1964); M. P. Weitzman, "The Peshitta Psalter
and Its Hebrew Vorlager VT 35 (1985) 341-54.

INFLUENCE OF THE SEPTUAGINT ON THE PESHITTA 257


a system of "checks and balances" to determine whether the Peshitta used the
LXX.19 What results from applying these arguments may then explain how
the LXX was used. The first four of these arguments are based on the presence of a congruent reading, while the remaining two are based on the absence of a congruent reading.
Argument 1. The value of a congruency must be taken into consideration before
one suggests that versional influence from the LXX is responsible for a parallel
reading in the Peshitta. Just as a manuscript's tradition is more important than
its date, so too the nature and significance of a parallel outweigh the mere
presence of a parallel.
The value of a congruency can be insignificant or significant in terms of
its nature. An insignificant parallel refers to a plus, minus, or substitution
that is somewhat trivial in nature, or that can readily be explained by reasons
within the text. So, for example, in Job 1:5 we find the explicit expression of
past continuity intimated in the Heb. imperfect nfcw\ "he would do," in both
the Peshitta (rVocn .TJS^.) and the LXX () ; in Job 4:4 the addition of
an implicit relative pronoun is found in the Peshitta, the LXX, and the Tg:
in Job 4:4 a collective singular in the MT is rendered by a plural in the
Peshitta and the LXX.
Significant parallels are those similar departures from the MT in the
Peshitta and the LXX that are so unique in terms of grammar, semantics,
syntax, or style that they suggest some degree of influence. So, for example, in Job 1:15 the MT's ann *DV, "by the mouth of the sword," is
rendered "by the sword" in the Peshitta ( n ^ i u s ) and in the LXX (
) ; in Job 3:10 MT rea nVr, "the doors of my womb," is rendered K^JrV.T criA^D'Oi ^Ji\ "the doors of the womb of my mother" in
the Peshitta and , "the gates of my mother's
womb," in the LXX; in Job 22:3 both the Peshitta and the LXX lack a
translation for the MT's pTsn 'D Dnn, "Is it a delight to Shaddai that
you are proven right?"
Argument 2. When the immediate environment does not continue the "parallel,*'
this may be further evidence suggesting that a congruent reading, in particular
an insignificant one, is not the result of versional influence.
Consider, as an example, chap. 1 of Job. This chapter contains twelve
readings where the Peshitta and the LXX preserve congruent departures
19
Here I am following the terminology used by Lund in his dissertation (see the preceding
note) and Dirksen's concept of "possibilities" ("Peshitta and Textual Criticism of the Old Testament," 377$79).

258 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


from the MT: w. 1, 5, 7, 9, 10, 12, 13, 14, 15, 16, 18, 22. If we consider the
value of these parallels as discussed in "argument one," eight are insignificant
(in vv. 5, 7, 12, 14, 18, 22 explicit exegesis; in w. 15 and 16 removal of
figurative language). Four remain. In 1:1 both the Peshitta and the LXX add
an indefinite particle to the MT: "there was a man" (tf"x)becomes "there
was a certain man." The explanation for this congruency is difficult to determine. The added particle may be implicit in the MT. Not just "a man" or "any
man" is referred to, but a particular man who lived in the land of Uz. This
is made clear by the following "whose name was Job." In 1:9 the syntax of
the MT, "and the adversary answered the Lord and said," is altered to "and
the adversary answered and said to the Lord" in the Peshitta and the LXX.
It is noteworthy, however, that wherever this introductory formula occurs in
Job, the Peshitta makes a similar adjustment in syntax, while the LXX does
not. The fact that in 1:10 MT pD, "to break through," becomes t$\fc>, "to
increase, multiply," in the Peshitta and , "to make many,
increase," in the LXX can readily be explained as the result of substituting
a contextually obvious lexeme for a rare term, a phenomenon common elsewhere in the Peshitta to Job. In 1:13 both the Peshitta and the LXX add the
qualifying "(the sons and daughters) of Job" where the MT reads "and his
sons and his daughters." Note, however, that the syntax of each is distinct:
the Peshitta reads "the sons and daughters of Job," whereas the LXX reads
"the sons of Job and his daughters."
Argument 3. A parallel reading may be deemed coincidental if the same technique
or adjustment is found elsewhere in the Peshitta without the LXX parallel.
Consider, by way of example, the congruency found in 1:9 (noted in
argument two above) in the syntax of the introductory formula "and Nl
answered N2 and said."20 This formula occurs nine times in Job (1:7,9; 2:2,4;
38:1; 40:1,3,6; 42:1). In two cases (38:1; 40:6) the Hebrew also contains the
prepositional phrase "from the whirlwind" after the direct object. By contrast, the Peshitta presents the formula as "and Nl answered and said to N2
(from X)" in all nine cases. It is significant that of the thirteen appearances
of this formula elsewhere in the Hebrew Bible, the Peshitta makes a similar
syntactic adjustment in all but four.21 Thus, of the twenty$two occurrences
of this introductory formula the Peshitta makes a syntactic adjustment in
eighteen; the LXX contains a congruent reading only once, in Job 1:9. The
lack of any other congruencies in the LXX and the frequency with which the
20

See Szpek, Translation Technique in the Peshitta to Job, 109$11, where this issue was
first discussed.
21
In two cases (2 Sam 4:9; 2 Kgs 7:2) the lack of adjustment is readily apparent; in two
(1 Sam 21:5; 22:14) the lack of rearrangement is peculiar.

INFLUENCE OF THE SEPTUAGINT ON THE PESHITTA 259


Peshitta makes this alteration strongly suggest that in the case of 1:9 the
congruent reading of the Peshitta and the LXX is the result of coincidence,
not of the LXX's influence on the Peshitta.
Argument 4. When a congruent reading in the Peshitta and the LXX is also
present in another version, this weakens the possibility that the Peshitta's reading
is dependent on the LXX.
We see this argument demonstrated, for example, in Job 31:9b, where
the Peshitta, the LXX, and the Vg all add the conditional particle " i f to the
second stich; in Job 34:18 the MT's "ibxn is rendered in the Peshitta, the
LXX, the Vg, Theodotion, as if the MT were lkn; or in 36:7 KODV O*DVD nxi,
"and with kings upon (lit. to) the throne," the Peshitta, the LXX, the Tg, and
the Vg all provide the equivalent of "upon," clearly implicit in the Hebrew
preposition V.
The final two arguments are based on the absence of Septuagintal congruences in the Peshitta. Both question why the LXX was not used by the
translator of the Peshitta, and why the two translations differ at difficult
junctures in the text.
Argument 5. The absence of Septuagintal influence on very difficult passages
might be used as an argument in absentia that elsewhere the Peshitta's translator
did not consult the LXX.
In Job 3:5 (uv nnD innitt*), for example, the enigmatic nnoD is translated in the Peshitta (as well as the Tg and the Vg) as if it were derived from
the root n o , "to be bitter," + the preposition D. The LXX, by contrast, lacks
a translation for this lexeme and joins uv and inn^rr to v. 6 with the mnn n^Vn
serving as the joint subject (with uv) of the verb inny:r:
, "let that day and night be cursed." In Job 12:5, the syntax
makes this verse a crux interpretum: pxtf rontfvV ro TDV, "the one whose
thoughts are at ease has contempt for disaster [?]." The Peshitta's paraphrase
rdicL^o rVAicu^T. c i o i s o i A , "in order to take away contempt and
iniquity" is based on the terms rn and TDV. An infinitive is supplied, but no
equivalent of pxtf mntf. The LXX's reading, $
, "for it had been ordained that he should fall under
others at the appointed time,"22 clearly did not serve as a basis for this
reading of the Peshitta. One further example (of many) is in Job 28:4. The
words of this verse in the MT are all understandable, until they are placed
in relation to one another:
22
The LXX's reading is difficult. Here, in the remainder of v. 5 and in v. 6, the LXX tends
to combine and paraphrase elements of the MT. See E. Dhorme, A Commentary on the Book
of Job (New York: Thomas Nelson, 1967; French original, Paris: Lecoffre, 1926) 169.

260 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


:W3 tfuxa IVT br\ * OTiDtfan n* DS Vm pD
Far from any sojourner they open a shaft,
Far from human feet they are forgotten folk,
Far from people they swing suspended there.23
(or)
He breaketh open a shaft away from them that sojourn in the light;
They that are forgotten by the foot (that passeth by);
That hang afar from men, that swing to and fro.24
(or)
He sinks a shaft far from habitation,
Forgotten by the foot of man,
Suspended remote from men they sway.25
(or)
A foreign people has pierced shafts,
Forgotten of human feet;
They swing to and fro, they are poised far from men!26
If we compare the translations of this verse in the Peshitta:

They inherited a channel from a strange people,


they were forgotten (or "led astray") from feet,
and they were cut off from men.
and in the LXX:
,
,
.
There is a cutting off of the torrent by reason of dust,
so they that forget the right way are weakened;
they are removed from (among) men.
It is obvious that the Peshitta did not use the LXX as a crib.
23

Norman C. Habel, The Book of Job: A Commentary (OTL; Philadelphia: Westminster,


1985) 388.
24
S. R. Driver and G. B. Gray, A Critical and Exegetical Commentary on The Book of
Job (2 vols.; Edinburgh: T. & T. Clark, 1921) 2. 238.
25
Marvin H. Pope, Job: A New Translation with Introduction and Commentary (AB 15;
Garden City, NY: Doubleday, 1983) 197.
26
Dhorme, Commentary on the Book of Job, 401.

INFLUENCE OF THE SEPTUAGINT ON THE PESHITTA 261


Argument 6. The presence of divergent departures from the MT may suggest
that elsewhere the Peshitta did not consult the LXX.
In Job 27:19, for example, the MT's Vi astf* Ttf, "the rich man will
lie down, and he will not be gathered," is rendered in the Peshitta "because
the rich man will lie down and will not again rise (JOLO^IA C U HAO)"
and in the LXX "the rich man has lain down, and he will not continue" (
). Both the Peshitta and the LXX understood the Hebr. niphal
as a qal active, but the Peshitta adds a complementary infinitive to complete
the thought. Another example occurs in Job 31:10, n^sn *n#x i n i jnon
53\ "Then let my wife grind for another, and let others crouch upon
her." Both the Peshitta and the LXX substitute a euphemism for the second
stich. Each, however, is semantically distinct: "Then my wife ground for
others, and she baked in another place (r*jiurt' r ^ o x s \)" ;
"Then let my wife also please another, and let my children be brought low
( )."
Ill
IN THE PESHITTA TO JOB I counted 244 cases where the Peshitta and the
LXX are congruent against the MT. 27 Application of arguments 154 yields
the results that follow.
Argument I (value of a congruency). Of the 244 congruencies, 56 are
significant,28 and 188 are insignificant.29
27
The late nineteenth century witnessed the publication of three studies on the Peshitta
to Job, those of Mandi, Stenij, and Baumann. According to A. Mandi (Die Peschittha zum
Hiob, nebst einem Anhang ber ihr Verhltniss zu LXX und Targum, 1 [Budapest: Leo Dropper, 1892] 33), the presence of additions and doublets leads us to conclude that "the Peshitta was
altered and often corrected on the basis of other translations, especially the LXX" (translation
my own); then (pp. 34-35) he lists sixteen places where the Peshitta and the LXX are parallel,
but he cautiously labels them "congruenz mit MT." E. Stenij (De Syriaca Libri lobi Interpretatione quae Peschita Vacatur [Helsingfors: Frenckell, 1887]) did not deal with this issue but
covered intra-Syriac variants in the Peshitta of Job. By contrast, E. Baumann's study of the
Peshitta of Job ("Die Verwendbarkeit der PeSita zum Buche Ijob fr die Textkritik," AW 18
[1898] 305$38; 19 [1899] 15$95,287$309; 20 [1900] 177$307) included consideration of the Peshitta's
relation to the LXX. He provided 169 occurrences where the Peshitta and the LXX were
parallel, but noted (ZAW 19 [1899] 65$69) that "where a parallel is found, it is often sporadic,
and it is difficult to imagine that the translator of the Peshitta would have consulted the LXX
in such insignificant matters" (translation my own).
28
The nature of these congruencies involves the removal of figurative language, generalization, contextual translation, interpretation, syntax, major omissions, and major semantic adjustments.
29
The nature of these congruencies concerns adjustments involving conjunctives, language difference, explicit exegesis, error, redundancy, and ambiguity.

262 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


Argument 2 (evidence from the immediate environment). Of the fortytwo chapters in the book of Job, thirty follow a pattern of sporadic versional
parallels, both significant and insignificant in nature. In twelve chapters,
then (chaps. 4, 7, 9, 10, 11, 12, 14, 18, 29, 30, 32, 33), there is sustained
congruence with the LXX; however, the sustained parallels extend over only
three or four verses at most (in chaps. 18, 29, 30, 32).
Argument 3 (same technique or adjustment elsewhere in the Peshitta
without the LXX parallel). Of the 244 congruences of the Peshitta with the
LXX, the same technique or adjustment is found elsewhere in the Peshitta
without the LXX parallel in 114 cases.
Argument 4 (nonexclusive readings). Of the 244 congruencies between
the Peshitta and the LXX, 106 are also present in one or more of the other
versions (Tg, Vg, Aquila, Symmachus, Theodotion).
It is noteworthy, too, that in 210 of the 244 congruencies of the Peshitta
with the LXX more than one argument is pertinent. In the remaining 34
congruencies, argument 1 (value of a congruency) alone is applicable. Of
these 34 congruencies, 12 are insignificant, while 22 are significant. In all but
4 of these 22 cases, however, there are internal reasons that might explain
how two versions could arrive independently at the same translation.30 Consider each of the four cases in which such an internal reason is not present.
/. Job 4:16.
-.mtx Vipn nooi "naV ruion inm TDX XVI l o r
(And) it stood still and I could not recognize its appearance;
a likeness (was) before my eyes,
a whisper then a sound I heard.
. *03fV.T \*, rlao

fVoo^JO

And I stood still and I did not recognize (it),


and there was no appearance before my eyes;
then a soft whisper and a voice I heard which said . . .
,
,
' .
I arose and perceived (it) not;
I looked and there was no form before my eyes;
but I only heard a breath and a voice (saying) . . .
The significant congruency of the Peshitta and the LXX ("and there was no
appearance/form before my eyes") is the contradiction to the MT's "a likeness (was) before my eyes." How do we explain this negation? Perhaps they
30

Examples of such internal reasons are confusion of roots (10:20; 12:6; 12:23; 16:7),
explicit exegesis (33:8), and interverse influence (12:16b).

INFLUENCE OF THE SEPTUAGINT ON THE PESHITTA 263


both extended the preceding negation, "I could not recognize its appearance," to this phrase, believing that only a voice was to be heard, with no form
present.
2. Job 7:15.
My soul chose strangulation,
death rather than my bones.

:*mo mo *tfD3 pano mam

And you tried my soul more than (or "from," or "by") destruction,
and my bones more than (or "from," or "by") death.
' ,
.
You will separate my life from my spirit;
and my bones from death.
The syntax of the MT has long been a crux interpretum, in particular the
function and position of the preposition }D in both stichs. The Peshitta's
syntax clearly parallels that of the LXX, but the result in the Peshitta is
nonsensical. The Peshitta literally adopted the MT's West Semitic ina, but
in Syriac ina means "to test, try," not "to choose" as in Biblical Hebrew.31
Thus, if we assume that the Peshitta's translator consulted the LXX for the
syntax of this verse, why did the translator not consult it for the meaning of
the verb man?
3. Job 12:14.
: ^ xVi tfn< Vv i ^ ma? xV onn*
If he tears down, then it will not be rebuilt.
(If) he shuts [the door] in the face of a man,
then it will not be opened.
. HJU 1271 Auffr ^x ncVn
,jj<k

cu^zj

june*

,J&n3

.lurt*

.jG

Behold, if he demolishes, who can rebuild?


And if he shuts [the door] in the face of a man, who can open?
, ;
' , ;
If he should cast down, who will build up?
If he should shut up against man, who will open?
The congruency of the Peshitta and the LXX in departing from the MT in
the second part of each stich is striking. Is direct influence the only answer?
31

See Szpek, Translation Technique in the Peshitta to Job, 41$42, 51.

264 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


Perhaps the Peshitta and the LXX both read these niphal forms as qal
imperfects: "If he tears down, then he (or "one") will not rebuild! (If) he shuts
[the door] upon a man, then he (or "one") will not open!" Finding the
impersonal subject or the subject of the preceding clause inappropriate, both
transformed these second clauses into rhetorical questions: Who can rebuild?
Who can open? (= None can rebuild! None can open!)
ID mn \ ntf*?tf an
nay ixxo xV itfx VJ?
:a*px nx vrvm
And against his three friends his anger was kindled,
because they had not found an answer;
yet (lit. "and") they had declared Job guilty.

4. Job 32:3.

.(BOIIHIO

ocvrA

And he was angry at his three friends,


because they were not able to give (him) an answer to Job;
yet (lit. "and") they condemned him.
,
,
.
And he was also very angry with (his) three friends,
because they were not able to return answers to Job,
yet they set him down for an ungodly man.
Does the MT itself contain the catalyst which prompted the makers of both
versions to add the indirect object and then substitute a pronominal suffix for
the nominal object? Or was a different Vorlage used? "They had not found an
answer" (for Job) may be implicit in the text, but why replace "yet they had
declared Job guilty" with "yet they had condemned him" or "yet they set him
down for an ungodly man"? Perhaps to avoid redundancy after having just
added "Job" explicitly.
Application of arguments 1 through 4 to the Book of Job has demonstrated that 98.4 percent of the congruencies of the Peshitta with the LXX
can be explained as the result of something other than direct versional influence of the LXX on the Peshitta. This evidence is strengthened further by the
final two arguments, that of the absence of Septuagintal influence on the
Peshitta in difficult passages (argument 5) and the presence of divergent
readings in the LXX and in the Peshitta (argument 6). In presenting these
arguments I supplied ample illustrations showing that in many difficult passages the translator of the Peshitta of Job did not consult the LXX and that

INFLUENCE OF THE SEPTUAGINT ON THE PESHITTA 265


at times the Peshitta of Job and the LXX depart from the MT in ways that
diverge from one another.
The Peshitta's translator did not use the LXX as a constant companion
in translation. Even to say that the translator consulted the LXX when there
was need to resolve a difficulty in the text is clearly inaccurate (argument 5).
The quest of a solution to the problem of Peshitta readings congruent with
those of the LXX must ihen take a different path. Perhaps congruencies are
due to the "memoriter influence" suggested by Eriksson.32 Perhaps congruencies entered the text through the lectionary system, as the weekly readings
were adjusted to the Greek text and these adjustments of select passages
eventually worked their way into the text of Job in a later copy of the
Peshitta. We do have lectionaries from the ninth century and later that contain a number of these unique congruencies.33 Perhaps the manuscripts of the
Peshitta text of Job which we have today were in some way "cleansed" of
Greek influence by a later editor who overlooked the few unique readings
that remain. Perhaps one should not discount the possibility that no one yet
has found the explanation of these few sporadic, unique congruencies with
the LXX that are preserved in the Peshitta of Job.34 Or perhaps the Peshitta
and the LXX are translated from two Hebrew Vorlagen.
Current research has brought forth many fresh ideas on the cause of
congruent readings in the Peshitta and the LXX. We can no longer simply
attribute such concurrence to direct dependence of the Syriac on the Greek
as Mathews would have us believe. The relation of the Peshitta to the LXX
is a very complex one. While no single hypothesis, theory, or response can
be applied to the entire corpus of such a heterogeneous work as the Peshitta
of the Old Testament, perhaps the arguments put forward here can be used
to determine the LXX's degree of influence on each individual book, so that
we can then better understand the use of the LXX in the translation of the
Peshitta as a whole.35
Postscript. After this article was already in the editor's hands, the posthumous work of Gsta Rignell, The Peshitta to the Book of Job, Critically
Investigated with Introduction, Translation, Commentary and Summary (cd.
32

Eriksson, "Hymns of David Interpreted in Syriac," 184.


See The Old Testament in Syriac 2/ la: Job, xv-xviii.
34
Mulder ("Use of the Peshitta in Textual Criticism," 43-45), notes how doubtful it is that
one can adequately respond to the question how the LXX was used by the translator of the
Peshitta.
35
I wish to express my thanks to Michael V. Fox of the Department of Hebrew and
Semitic Studies, University of Wisconsin-Madison, for his many suggestions and comments
throughout the preparation of this article.
33

266 THE CATHOLIC BIBLICAL QUARTERLY I 60, 1998


Karl-Erik Rignell; Kristianstad [Sweden]; Monitor, 1994), came to my attention. On p. 5, its editor questions whether Rignell, who died in 1981, intended
publication of this work. It does seem likely that, had he lived to publish it,
he would have included parallels with the LXX and the other versions in his
notes, where he deals mainly with the Peshitta's departures from the MT.
His summary is general and instinctual in nature, but his instincts are
well grounded on his research. His final comments on the Vorlage used by the
Peshitta translator of Job and on the influence of the LXX are highly pertinent. He states that "on the whole the Syriac text corresponds to the Hebraic, at least to some extent" (p. 370). When the Peshitta departs from the
MT it is because of "the translator's prejudicial attitude to the meaning of
the text" (p. 376). He believes, furthermore, that subsequent work on the
Peshitta of Job was based not on the Hebrew text but on a defective form of
the original translation (p. 375). He thinks that the Syriac translator of Job
was "in all probability entirely independent" of the LXX, although in the
case of the first translator, "who often, according to one hypothesis, only
stated the meaning of the Hebraic word," the LXX "could have influenced
his choice of expressions"; in later stages of translation, no influence of the
LXX on the Peshitta's text of Job is verifiable (p. 379).

^ s
Copyright and Use:
As an ATLAS user, you may print, download, or send articles for individual use
according to fair use as defined by U.S. and international copyright law and as
otherwise authorized under your respective ATLAS subscriber agreement.
No content may be copied or emailed to multiple sites or publicly posted without the
copyright holder(s)' express written permission. Any use, decompiling,
reproduction, or distribution of this journal in excess of fair use provisions may be a
violation of copyright law.
This journal is made available to you through the ATLAS collection with permission
from the copyright holder(s). The copyright holder for an entire issue of a journal
typically is the journal owner, who also may own the copyright in each article. However,
for certain articles, the author of the article may maintain the copyright in the article.
Please contact the copyright holder(s) to request permission to use an article or specific
work for any use not covered by the fair use provisions of the copyright laws or covered
by your respective ATLAS subscriber agreement. For information regarding the
copyright holder(s), please refer to the copyright information in the journal, if available,
or contact ATLA to request contact information for the copyright holder(s).
About ATLAS:
The ATLA Serials (ATLAS) collection contains electronic versions of previously
published religion and theology journals reproduced with permission. The ATLAS
collection is owned and managed by the American Theological Library Association
(ATLA) and received initial funding from Lilly Endowment Inc.
The design and final form of this electronic document is the property of the American
Theological Library Association.