ARAMAIC-LIKE FEATURES IN THE PENTATEUCH

Gary A. Rendsburg
Rutgers University

The term “Aramaic-like features” is to be distinguished from the term

“Aramaisms.” The former refers to linguistic traits found in pre-exilic texts,
whose presence can be explained by one of two reasons: either the texts are
northern in origin, or the settings of the texts have an Aramean flavor. The lat-
ter refers to those features, found primarily in the post-exilic corpus, which re-
flect clear Aramaic influence over Hebrew. Aramaic-like features occur with a
significant concentration in five sections of the Torah: Genesis 24, Genesis
30–31, Numbers 22–24, Genesis 49, and Deuteronomy 33. Style-switching
explains the first three texts, since the first two are narratives geographically
set in Aram, while the third portrays a prophet from Aram in the plains of
Moab. Regional dialectology explains the remaining two sections: the sayings
about the individual tribes must originate in those specific locations, which is
why one finds words like MOwrDd ; , MRrR…g, and so forth, in the blessings to Issachar,
Naphtali, Joseph, and Gad. If the Pentateuch were the product of Persian-
period Jewish scribes, as claimed by some scholars during the last several
decades, one would expect Aramaisms or Aramaic-like features to appear
throughout its 187 chapters in significant concentrations, and not, as per the
main conclusion of this essay, in select chapters for specific purposes.

As is well known, a major trend has been noticeable in the field of bibli-
cal studies during the past twenty years or so. More and more scholars are
dating more and more biblical texts to the Persian period and even to the
Hellenistic period. Typically, scholars who take this approach base their
dating of biblical texts on social, political, and theological concerns deemed
to be present in the texts. In so doing, they regularly ignore the most objec-
tive criterion available for the dating of texts, namely, the linguistic
evidence.1

1
This point has been made by other scholars; see, for example, J. Joosten, “The Distinction between
Classical and Late Biblical Hebrew as Reflected in Syntax,” HS 46 (2005): 327–339, in particular the
following statement: “historical linguistics has been invoked only marginally, and only a posteriori.
Linguistic data are no longer expected, it seems, to play a part within the historical-critical approach” (p.
328). I (and presumably Joosten et al.) have in mind, for example, the many published works of N.-P.
Lemche, T. L. Thompson, and P. R. Davies. The last named scholar recently penned an essay in which he
purports to deal with the linguistic evidence, but he speaks only in generalities, and one looks in vain for
the discussion of a single specific lexical or grammatical issue; see P. R. Davies, “Biblical Hebrew and the
History of Ancient Judah: Typology, Chronology and Common Sense,” in Biblical Hebrew: Studies in
Chronology and Typology, ed. I. Young (JSOTSup 369; London: T & T Clark, 2003), pp. 150–163.
Contrast every other article in this volume, each of which is replete with Hebrew data (regardless of
whether or not one agrees with the conclusions).
Hebrew Studies 47 (2006) 164 Rendsburg: Aramaic-like Features

It is thus refreshing to find, from time to time, certain scholars who do

engage with the linguistic evidence. Here I may cite, most prominently,
Alexander Rofé, Michael Barré, Marc Brettler, and Michael Waltisberg, all
of whom have written detailed studies of specific chapters considered to be
Persian-period compositions (Genesis 24, 1 Kings 21, Psalm 116, 1 Samuel
2, Judges 5).2 Their treatments include a listing of Aramaisms and other
late features, which, to their minds, support their conclusions.
In two previous studies, I have countered the conclusions of these schol-
ars, for to my mind they have neglected to note the settings of the texts
under consideration.3 Four of the texts, I argued, were northern Israelite
compositions, and thus one could expect to find Aramaic-like features in
them—without attributing their presence to a late dating. The Aramaic-like
features in the one remaining text, Genesis 24, were explained as elements
of style-switching—the scene occurs in Aram, to which Abraham’s servant
has traveled in order to obtain a bride for Isaac—and thus the language of
this story reflects the environment.
The present article builds on my previous studies, though it limits its fo-
cus to the Pentateuch. I have tried to gather as many features that could pos-
sibly be labeled as Aramaisms, or as I prefer to call them, Aramaic-like
features. I differentiate these two terms as follows: True Aramaisms are
those features, mainly lexical items, found primarily in the post-exilic
corpus—clear borrowings from Aramaic; while Aramaic-like features are
those features, again mainly lexical ones, which are found in pre-exilic texts
and carry a different explanation. As indicated in my two articles, they ap-
pear in pre-exilic texts for one of two reasons: either the texts are northern
in origin, or the setting of the texts has an Aramean flavor.

2
The first two chapters were treated by Rofé; then each of the other texts by the other three scholars listed,
in respective order. Bibliography: A. Rofé, “An Enquiry into the Betrothal of Rebekah,” in Die Hebräische
Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, ed. E. Blum,
C. Macholz, and E. W. Stegemann (Neukirchen: Neukirchener Verlag, 1990), pp. 27–39; A. Rofé, “The
Vineyard of Naboth: The Origin and Message of the Story,” VT 38 (1988): 89–104; M. L. Barré, “Psalm
116: Its Structure and Its Enigmas,” JBL 109 (1990): 61–79; M. Z. Brettler, “The Composition of
1 Samuel 1–2,” JBL 116 (1997): 601–612; and M. Waltisberg, “Zum Alter der Sprache des Deboraliedes
Ri 5,” ZAH 12 (1999): 218–232.
3
G. A. Rendsburg, “Some False Leads in the Identification of Late Biblical Hebrew Texts: The Cases of
Genesis 24 and 1 Samuel 2:27–36,” JBL 121 (2002): 23–46; and G. A. Rendsburg, “Hurvitz Redux: On
the Continued Scholarly Inattention to a Simple Principle of Hebrew Philology,” in Biblical Hebrew:
Studies in Chronology and Typology, ed. I. Young (JSOTSup 369; London: T & T Clark, 2003), pp.
104–128. Needless to say, my research in this area is heavily indebted to the pioneering article of A.
Hurvitz, “The Chronological Significance of ‘Aramaisms’ in Biblical Hebrew,” IEJ 18 (1968): 234–240.
See also his more recent treatment: A. Hurvitz, “rqjmb ‘Mymzyamra’h tyygws :arqmh tpwqtb tymraw tyrbo
tyarqmh tyrboh” in Studies in Hebrew and Jewish Languages Presented to Shelomo Morag, ed. M. Bar-
Asher (Jerusalem: Bialik, 1996), pp. 79–94.
Hebrew Studies 47 (2006) 165 Rendsburg: Aramaic-like Features

To my mind, the vast majority of the Torah is Judahite in origin. We

will identify one Israelian passage at the end of this article, but by and
large, the five books reflect both Standard Biblical Hebrew and a
thoroughly Judahite agenda. The one exception might be the book of
Deuteronomy, and a full investigation of the language of that book remains
a desideratum, but to my eye, a cursory glance does not reveal the kind of
concentration of Israelian Hebrew traits that one finds in, say, the material
about the northern kings in the book of Kings.4
Within this decidedly Judahite corpus, however, one does find concen-
trations of Aramaic-like features from time to time, and it is to those texts
that I now turn our attention.5 The first of these is Genesis 24, which, as
indicated above, I already have treated in a previous article. Note the
following features, all pointed out by Rofé:

1. The expression MˆyAmDÚvAh yEhølTa “God of heaven” in Gen 24:3, 7.

2. The phrase jåq; tI _aøl rRvSa “that you not take” in Gen 24:3, calquing on
Aramaic ald (instead of expected Hebrew MIa).
3. The root amg (Hiphfiil) “give drink” in Gen 24:17 (instead of expected
hqv [Hiphfiil], which is used seven times elsewhere in the story).
4. The root hro “pour” in Gen 24:20 (instead of standard Hebrew qxy or
qyr [Hiphfiil]).
5. The root hav (Hitpafiel) “gaze, watch” in Gen 24:21.
6. The particle aøl_MIa “but rather” in Gen 24:38 (instead of standard
Hebrew MIa yI;k).
7. The noun tOn∂;d◊gIm “choice gifts” in Gen 24:53.

Indeed, these are all Aramaic-like features, but they are not Aramaisms,
especially when one notices the total lack of inner Hebrew indications of
Late Biblical Hebrew in Genesis 24. At every point where the text could
have used a Late Biblical Hebrew feature, a Standard Biblical Hebrew fea-
ture is found instead. For example, the phrase hDÚvIa jåqDl “take a wife” occurs
six times (Gen 24:3, 4, 7, 37, 38, 40 [see also v. 67]) with no instances of
4
See G. A. Rendsburg, Israelian Hebrew in the Book of Kings (Occasional Publications of the
Department of Near Eastern Studies and the Program of Jewish Studies, Cornell University 5; Bethesda,
Md.: CDL Press, 2002). For my most recent and most comprehensive statement on Israelian Hebrew
(albeit largely in outline form), see G. A. Rendsburg, “A Comprehensive Guide to Israelian Hebrew:
Grammar and Lexicon,” Orient 38 (2003): 5–35.
5
In general I present the specific data in outline form only. For examples deduced in previous
publications, the reader may find full discussions therein. For examples deduced here for the first time, I
present more information, both substantive and bibliographic, though even in these instances I have not
produced thorough treatments.
Hebrew Studies 47 (2006) 166 Rendsburg: Aramaic-like Features

the Late Biblical Hebrew equivalent hDÚvIa aDcÎn; and the verb rhm is used for
“hurry” (Gen 24:18, 20, 46), instead of the Late Biblical Hebrew equivalent
lhb.6 Moreover, the noun-verb ratio of .600 places this chapter firmly
within the classical stratum,7 as per Frank Polak’s analysis of biblical nar-
rational style.8 We are led to conclude that this chapter is an exemplar of
pre-exilic Hebrew, and that the sprinkling of Aramaisms in the story con-
stitutes a literary device, to transport the reader, along with Abraham’s
servant, to the land of Aram.
There is a second narrative in Genesis that returns the reader, along with
the action, to the land of Aram—and that of course is the story of Jacob and
Laban in Genesis 30–31. Here too we find a host of Aramaic-like features
in the story, as noted first by Jonas Greenfield, with additional examples put
forward by the present author.9 Greenfield noted three examples:

1. The verbs lE…xÅ¥yÅw “he took away” in Gen 31:9 and lyI…xIh “he took away”
in Gen 31:16 (otherwise lxn Hiphfiil means “save” in Hebrew).
2. The verb qE;b√dÅ¥yÅw “he overtook” in Gen 31:23 (instead of the expected
Hebrew gE;cÅ¥yÅw—though see Gen 31:25).
3. The verb yˆnA;tVvAf◊n aøl “you did not allow me” in Gen 31:28 (calquing on
Aramaic qbv).

I then added the following examples:

4. The noun z…wl “almond” in Gen 30:37 (in place of dqv).

5. The noun dÎ…g “fortune” in Gen 30:11 (in the naming of Gad).
6. The root dbz “provide, supply” (as both verb and noun) in Gen 30:20
(in the naming of Zebulun).
7. The 3rd person feminine plural form hÎnVmAj´¥yÅw “and they were in heat” in
Gen 30:38 (standard Hebrew would be hÎnVmAjE;tÅw).

6
See R. M. Wright, Linguistic Evidence for the Pre-exilic Date of the Yahwistic Source (LHB/OTS
[JSOTSup] 419; London: T & T Clark, 2005), pp. 71–74, 82–84.
7
Genesis 24 contains 394 nouns and 263 verbs, yielding a noun-verb ratio of .600. The verbs are classified
as follows: 44 nominal verbs and 219 finite verbs, for a nominal-finite verb ratio of .167.
8
F. Polak, “arqmb trwpysh NwClb twpwqtw twrwmt,” Beth Mikra 43 (1997–1998): 30–52, 142–160; and F.
Polak, “The Oral and the Written: Syntax, Stylistics and the Development of Biblical Prose Narrative,”
JANES 26 (1998): 59–105.
9
J. C. Greenfield, “Aramaic Studies and the Bible,” in Congress Volume Vienna 1980, ed. J. A. Emerton
(SVT 32; Leiden: E. J. Brill, 1981), pp. 110–130, especially pp. 129–130; and G. A. Rendsburg,
“Linguistic Variation and the ‘Foreign’ Factor in the Hebrew Bible,” IOS 15 (1996): 177–190, especially
pp. 182–183.
Hebrew Studies 47 (2006) 167 Rendsburg: Aramaic-like Features

8. The inflected participle yItVb¨n◊…g “I was robbed” in Gen 31:39 (2 times)

(even though this morpheme is attested only later in Jewish Palestinian
Aramaic10 ).

Even more examples are forthcoming:

9. The noun MyIvÎyV;t “he-goats” in Gen 30:35, more common in Jewish

Palestinian Aramaic and other Western Aramaic dialects than in
Hebrew.11
10. The noun MyIfDh√r “troughs” in Gen 30:38, 41, which reflects Aramaic
phonology, where the root fhr corresponds to Hebrew Xwr “run”
(thus one reconstructs a semantic development “runner” >
“trough”12 ).
11. The verb PIlTjRh “exchange” in Gen 31:7, 41, occurring only here in
the Bible with reference to wages, money, and so on, paralleling a
usage well attested in Jewish Babylonian Aramaic in particular.13
12. The noun MyˆnOm “times” (in the expression MyˆnOm t®rRcSo “ten times”) in
Gen 31:7, 41—occurring only here in the Bible, and note how the
root hnm “count” is standard in Aramaic and more productive than in
Hebrew.
13. The form MRhVtRa in Gen 32:1, in place of standard MDtOwa, with the pres-
ervation of the h in this form linking it to Aramaic Mwhty / Nwhty.14

10
See G. Dalman, Grammatik des jüdisch-palästinischen Aramäisch (Leipzig: Hinrichs, 1905), p. 284.
11
M. Sokoloff, A Dictionary of Jewish Palestinian Aramaic of the Byzantine Period (2d edition;
Baltimore, Md.: Johns Hopkins University Press, 2003), p. 580. See also the references in A. Tal, A
Dictionary of Samaritan Aramaic (Leiden: Brill, 2000), p. 948. In light of the distribution of the word
vˆy;At in the Bible, one wonders if it is not always employed as a style-switching device. It occurs again in
Gen 32:15, which in many respects continues the language of Genesis 30–31 (note both MyIvÎyV;t and MyIlEj√r in
this verse, both of which occur in the preceding chapters); and later in the Bible in Prov 30:31 and 2 Chr
17:11. The latter two passages have Arabian affinities (note the use of the form M…wqVla A in the former, and
the reference to Arab tribute-bearers in the latter), which may be significant in light of the cognate tays in
Arabic.
12
Note the English words “run,” “runnel,” “runner,” and “runnet,” all meaning “small stream, rivulet.”
More significantly, see also the technical meaning of “runner” = “a channel along which molten metal runs
from the furnace to the mould” (Oxford English Dictionary, s.v. “runner,” def. II.9.c.); as well as “runnel”
in the sense of a man-made conduit, as in this 1883 citation: “Small runnels are generally chiselled for the
purpose of conducting the water into the cistern” (Oxford English Dictionary, s.v. “runnel,” def. 2).
13
M. Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the Talmudic and Geonic Periods (Ramat-
Gan: Bar-Ilan University Press, 2002), pp. 465–466.
14
See the similar approach taken by Mordechay Mishor concerning the presence of MRht V aR in Exod 18:20;
M. Mishor, “On the Language and Text of Exodus 18,” in Biblical Hebrew in Its Northwest Semitic
Environment: Typological and Historical Perspectives, ed. S. E. Fassberg and A. Hurvitz (Publications of
the Institute for Advanced Studies 1; Jerusalem: Magnes Press, 2006), pp. 225–229, especially p. 227.
Note that the present example comes from Gen 32:1—that is, one verse beyond the bounds of Genesis
30–31—on which see below, n. 18.
Hebrew Studies 47 (2006) 168 Rendsburg: Aramaic-like Features

I also would note two additional usages, which cannot be pinned down as
Aramaic-like features to the extent that I would like, but probably should be
mentioned here as part of the mix as well:

14. The root Psk “yearn, long for” (in the expression hDt ; pV sA kV nˆ POsVkˆn) in
Gen 31:30, not exactly the same as the root Psk “be ashamed, be
embarrassed, be chagrined,” used commonly in Jewish Babylonian
Aramaic15 but the closest parallel that one finds in Semitic.
15. The unique usage represented in wøl dyˆ…gIh yIlV;b_lAo “by not telling him”
in Gen 31:20—perhaps a reflection of Aramaic, especially given the
large number of compound particles based on lAo in Aramaic (lo
ypa, ybg lo, tnm lo, qyso lo, MC lo, etc.16 ), even if ylb lo itself is
not attested.

If this were not enough to bring you to the land of Aram, the author in-
cluded one final zinger as well, a pure Aramaic two-word expression,
aDt…wdShDc rÅg◊y “mound of witness” in Gen 31:47, equaling Hebrew dEoVlÅ…g, serv-
ing as an explicit reminder that Laban and Jacob have been speaking
Aramaic all along, and not Hebrew; just as Shakespeare’s single phrase et tu
Brute suffices to remind the theatre-goer that Julius Caesar and his cohorts
have been speaking Latin all along, and not Elizabethan English.
These features cannot be considered Aramaisms that penetrated Hebrew
during the Persian period. As was the case with Genesis 24, also here in
Genesis 30–31, whenever a Late Biblical Hebrew feature could have been
used, a Standard Biblical Hebrew feature is found instead. For example, the
word JK®r®;d “journey, distance” is used in Gen 30:36; 31:23, instead of the
Late Biblical Hebrew equivalent JKDlShAm; and once more we may note the
phrase hDÚvIa jåqDl “take a wife” in Gen 31:50, as opposed to the Late Biblical

15
M. Sokoloff, Dictionary of Jewish Babylonian Aramaic, pp. 592–593. In addition, note that, among the
four other attestations of this root in the Bible, the closest usage is to be found in Ps 84:3, which, as part
of the Korah collection, is an Israelian composition; see G. A. Rendsburg, Linguistic Evidence for the
Northern Origin of Selected Psalms (SBLMS 43; Atlanta, Ga.: Scholars Press, 1990), pp. 51–60. The
most glaring Aramaic-like feature in Psalm 84 is the verb rwd “dwell” in 84:11, its only occurrence in the
Hebrew sections of the Bible (the verb rwd in Ezek 24:5 means “pile, arrange,” even if one posits an
ultimate connection to rwd “dwell”; for discussion see M. Greenberg, Ezekiel 21–37 [AB 22A; New York:
Doubleday, 1997], p. 498).
16
M. Sokoloff, Dictionary of Jewish Palestinian Aramaic, pp. 406–408; and M. Sokoloff, Dictionary of
Jewish Babylonian Aramaic, p. 863.
Hebrew Studies 47 (2006) 169 Rendsburg: Aramaic-like Features

Hebrew equivalent hDÚvIa aDcÎn.17 Moreover, the noun-verb ratio of .629 also
places these two chapters firmly within the classical stratum.18
I turn now to a third Pentateuchal text, which includes a heavy dose of
Aramaic-like features. In this case, the readers are not transported to the
land of Aram, but instead an Aramean prophet comes to the land of Canaan.
I refer, of course, to the story of Balaam in Numbers 22–24 and in particu-
lar to the oracles placed in his mouth. Here too I repeat information from
my earlier study, with specific attention to the following Aramaic-like
features:19

1. The reduplicatory plural form yér√rAh “mountains” (construct) in Num

23:7.
2. The noun oAbør “dust-cloud” in Num 23:10, with a cognate in Syriac,
as well as in Akkadian.
3. The noun vAjÅn “divination” in Num 23:23 (with the plural form MyIvDj◊n
in Num 24:1 in the prose narrative).
4. The noun tUkVlAm “kingdom” in Num 24:7, the classic Aramaic form, in
place of the standard Hebrew form hDkDlVmAm.
5. The verb bDÚvAjVtˆy “be considered” in Num 23:9, with the Hitpafiel serv-
ing for passive voice, as with the T-stem in Aramaic (one expects the
Niphfial in Hebrew).
6. The verbal form …wyDÚfˆn “stretched out” (or “standing tall”20 ) in Num
24:6, with the retention of the yod in the lamed-he verb, as in
Aramaic.
7. The form MUa◊n in Num 24:3–4 (3 times), 15–16 (3 times), used of hu-
man speech, occurring elsewhere only in northern settings in the

17
See R. M. Wright, Linguistic Evidence, pp. 71–74, 95–97, 130.
18
The data include Gen 32:1–3. To some extent, starting at Gen 30:1 is somewhat arbitrary, since there is
no break indicated in the Masoretic Text at this point. But one must begin somewhere, especially since the
previous break is a setuma after Gen 28:10. On the other hand, a clear break is present at Gen 32:3, which
is followed by a petuh.a, and thus I have included the first three verses of Genesis 32 in my calculations.
This also will explain the incorporation of the form MRht V aR in Gen 32:1 into our survey (on which, see
above). In sum, the data are as follows: Gen 30:1–32:3 contains 608 nouns and 359 verbs, yielding a
noun-verb ratio of .629. The verbs are classified as follows: 39 nominal verbs and 320 finite verbs, for an
exceedingly low nominal-finite verb ratio of .109.
19
G. Rendsburg, “Linguistic Variation,” pp. 183–184. The first effort in this direction was the seminal
article by S. A. Kaufman, “The Classification of the North West Semitic Dialects of the Biblical Period
and Some Implications Thereof,” in Proceedings of the Ninth World Congress of Jewish Studies: Panel
Sessions: Hebrew and Aramaic Languages (Jerusalem: World Union of Jewish Studies, 1988), pp. 41–57,
especially pp. 54–55.
20
M. Moreshet, “harwh hnCm :(w ,dk rbdmb) ‘…wyÚDfIn MyIlDjVn;Ik,’” Beth Mikra 48 (5732): 51–56; and S. Morag,
“Molb ylCmb MyynwCl Mynwyo :twmdq ydbwr,” Tarbiz 50 (1981): 1–24, especially pp. 15–16, in particular n.
54.
Hebrew Studies 47 (2006) 170 Rendsburg: Aramaic-like Features

Bible;21 most likely related to Eblaite en-ma,22 which once more takes
us to the general region of Aram, even if the floruit of Ebla was at the
end of the Early Bronze Age.

These are not the only examples, however. Note further the following
Aramaic-like features, some of them a bit less obvious and glaring:

8. The noun MyîrUx “mountains” in Num 23:9 in the a-line of the couplet,
replacing standard Hebrew MyîrDh, paired with twøoDb◊…g “hills” in the b-line
(the only such case in the Bible); the form MyîrUx evokes Aramaic Myrwf
“mountains,” and no doubt reflects an attempt to include that Aramaic
word in the poetry, using the Old Aramaic orthography still, in which
the emphatic interdental /z./ is represented by x (before the shift to f
occurred)—in fact this orthography appears still in the Adon letter,
line 8, where “he guarded” appears as rxn (and not expected rfn).23
9. The expression MyîrDv◊y tOwm, literally “death of the upright,” in Num
23:10, is explained by Menahem Kister as the opposite of the expres-
sion hjl twm “an evil death” in Nerab tomb inscription, no. 1, line
4—and indeed this verse as a whole shares much in common with
Nerab tomb inscription, no. 2, lines 3–4.24
10. The verbal root Mrg “devour bones” in the expression MérÎg◊y MRhyEtømVxAo◊w
“and their bones he devours,” in Num 24:8, uses the denominative
verb based on the Aramaic word Mrg “bone” (2 times elsewhere in the
Bible, though as the Qal; this is the only Pifiel).
11. The suffix MRhy´- added to a plural noun ending in -ôt, as in the above-
cited word MRhyEtømVxAo “their bones,” in Num 24:8; as is well known,
Standard Biblical Hebrew prefers the form MÎ- , while Late Biblical
Hebrew prefers the form MRhy´- , with the latter occurring as a result of
Aramaic influence;25 not that every instance of MRhy´- in pre-exilic texts

21
G. A. Rendsburg, “The Northern Origin of ‘The Last Words of David’ (2 Sam 23,1–7),” Biblica 69
(1988): 113–121, especially pp. 115–116.
22
C. H. Gordon, “Vocalized Consonants: The Key to um-ma/en-ma/Man,” in The Tablet and the Scroll:
Near Eastern Studies in Honor of William W. Hallo, ed. M. E. Cohen, D. C. Snell, and D. B. Weisberg
(Bethesda, Md.: CDL Press, 1993), pp. 109–110. See also G. A. Rendsburg, “Hebrew Philological Notes
(I),” HS 40 (1999): 27–32, in particular pp. 29–30.
23
See S. Morag, “twmdq ydbwr,” p. 8.
24
M. Kister, “Some Blessing and Curse Formulae in the Bible, Northwest Semitic Inscriptions, Post-
Biblical Literature and Late Antiquity,” in Hamlet on a Hill: Semitic and Greek Studies Presented to T.
Muraoka on the Occasion of His Sixty-Fifth Birthday, ed. M. F. J. Baasten and W. Th. van Peursen (OLA
118; Leuven: Peeters, 2003), p. 325.
25
For general discussion, see A. Hurvitz, A Linguistic Study of the Relationship between the Priestly
Source and the Book of Ezekiel (Cahiers de la Revue Biblique 20; Paris: J. Gabalda, 1982), pp. 24–27. For
Hebrew Studies 47 (2006) 171 Rendsburg: Aramaic-like Features

is an example of style-switching, but in the present instance, in the

mouth of Balaam, this is most likely the proper explanation.26

To these certain examples, I add here two other items of potential value to
our treatment:

12. The noun MyIlDj◊n “palm trees” in Num 24:6, a rare usage in the Bible,
with a cognate in Arabic nah.l27 —was this lexeme used in Old
Aramaic as well? Was it known in the land of Qedem, whence
Balaam came, located at the northern reaches of the Syrian Desert?
13. The form yådDo in the expression yådDo hÎnyˆza
S hA in Num 23:18, usually
translated “give ear to me”; however, since NyˆzTaRh typically governs the
preposition lRa or -Vl (see especially Deut 1:45; Ps 77:1, both with a
pronominal suffix attached to lRa),28 due consideration should be
given to the suggestion that yådDo here means “my warnings,” closely
related to the noun ydo “covenant, testimony” occurring repeatedly in
the Sefire treaties;29 if this interpretation is correct, then we have
here another striking Aramaic-style usage in the Balaam oracles.

In short, the Balaam oracles are filled with Aramaic-like usages, which
together serve the purpose of the style-switching employed by the ancient
Israelite author.
These three sections of the Torah contain the highest concentration of
Aramaic-like features, but they do not hold a monopoly. There are two
other texts where such elements may be found, namely, the blessings con-
cerning the tribes in Genesis 49 and Deuteronomy 33.30 One is not sur-
prised to find Aramaic-like features in these two chapters, but I hasten to

the most recent treatment, see M. Bar-Asher, “Pyosb Nwyo) l″zj NwCll arqmh Nyb Narmwq NwCl
(hygwlwprwmb,” twlygm 2 (2004): 137–149.
26
For another instance of this morpheme used for stylistic purposes, note the form MRhyEtwøba S in 1 Kgs
14:15, in the mouth of Ahijah of Shiloh, addressed to the wife of Jeroboam. This is the only occurrence of
MRhyEtwøbaS in the entire narrative corpus of Genesis through Kings, the eighth century prophets, pre-exilic
Psalms, Proverbs, etc., which otherwise use MDtwøba S 82 times (according to my counting). This would be
another feature of Israelian Hebrew appearing in the material concerning the northern kings in the book of
Kings, though I neglected to include this example in my book Israelian Hebrew in the Book of Kings.
27
See S. Morag, “twmdq ydbwr,” pp. 14–16. The only other instance is Song 6:11.
28
The only other collocation of NyˆzTaRh and dAo is in Job 32:11, in which the preposition introduces the
object to be listened to, namely MRkyEtOn…wbV;t “your wise-sayings.”
29
S. Morag, “twmdq ydbwr,” pp. 10–12.
30
For the first of these texts, see G. Rendsburg, “Israelian Hebrew Features in Genesis 49,” in Let Your
Colleagues Praise You: Studies in Memory of Stanley Gevirtz (Part 2), ed. R. J. Ratner, L. M. Barth, M.
L. Gevirtz, and B. Zuckerman, Maarav 8 (1992): 161–170. I have not yet published my work on the
parallel material in Deuteronomy 33, though I plan to do so in the near future.
Hebrew Studies 47 (2006) 172 Rendsburg: Aramaic-like Features

add that we find these elements specifically in the blessings to the northern
and Transjordanian tribes, that is, the ones with the closest links to Aramaic
speakers—but not in the blessings to Judah, Benjamin, Simeon, and Levi.
Note the following from Genesis 49:

1. In line with what we noted above, regarding the verb Mrg in the
Balaam oracles: the noun MRrR…g in Gen 49:14 in the blessing to Issachar:
M®rÎ…g rOmjS rDkCDÚcˆy “Issachar is a donkey of bone,” using the Aramaic
word for “bone” (in place of Hebrew MRxRo).
2.–3. The expression rRpDv_yérVmIa “lambs of beauty” in Gen 49:21, in the
blessing to Naphtali, using two words better attested in Aramaic: rma
“lamb” and the root rpv “beauty, be beautiful.”
4. The noun r…wv “wall” in Gen 49:22, in the blessing to Joseph; true, this
lexeme appears in 2 Sam 22:30 = Ps 18:30 as well, but only in the b-
line, where Aramaisms are more likely to occur.31

If we turn to the tribal blessings in Deuteronomy 33, we see several

more Aramaic-like features, though once more they appear only in passages
addressing the northern and Transjordanian tribes.

1. The noun MyIj∂r◊y “months” in Deut 33:14 in the blessing to Joseph (as
opposed to the standard Hebrew term MyIvDdFj).
2. The plural form yér√rAh “mountains” (construct) in Deut 33:15, also in
the blessing to Joseph (see also above in the discussion about the
Balaam material).
3. The verb aEt´¥yÅw “and he came” in Deut 33:21 in the blessing to Gad,
from the root hta, the standard Aramaic equivalent of Hebrew awb.
4. The noun Mwør∂;d “south” in Deut 33:23 in the blessing to Naphtali, the
Aramaic equivalent to the various Hebrew words for “south,” includ-
ing NyImÎy, NDmyE;Et, and b‰g‰n—otherwise, Mwør∂;d appears in Ezekiel, Qohelet,
and Job.

The five texts that we have surveyed—Genesis 24, Genesis 30–31,

Numbers 22–24, Genesis 49, and Deuteronomy 33—are the only sections of
the Torah with a significant concentration of Aramaic-like features, or call
them Aramaisms, if you will. This cannot be coincidental. Style-switching

31
G. R. Driver, “Hebrew Poetic Diction,” in Congress Volume Copenhagen 1953 (SVT 1; Leiden: E. J.
Brill, 1953), pp. 26–39.
Hebrew Studies 47 (2006) 173 Rendsburg: Aramaic-like Features

explains the first three texts that we looked at, while geography explains the
second two. The sayings about the individual tribes must originate in those
specific locations, which is why one finds words like M®r‰…g, Mwør∂;d, and so
forth, in the blessings to Issachar, Naphtali, Joseph, and Gad—but not in the
blessings to Judah, Benjamin, Simeon, and Levi.
If the entire Torah were the product of the Persian period or later, one
would expect to find clusters of Aramaisms throughout the five books, and
not just in these five specific texts, comprised of 227 verses.32 Clearly this
is not the case—and yet this point has neither been recognized nor been ad-
dressed by those who posit the period of Persian rule as the background for
large chunks if not the entirety of the Pentateuch. In fact, one is surprised
that those who propose a late dating for the Torah—in whole or in part—do
not utilize this evidence (save for an occasional exception, such as Rofé, as
noted at the outset). The surprise quickly dissipates, however, when one re-
alizes that the linguistic evidence—especially the controlled use of language
in well-defined literary contexts, as demonstrated herein—is largely ig-
nored. To be perfectly honest, I feel a bit like Rab-shaqe taunting the Juda-
hites with the offer of two thousand horses, if they were able to provide
riders (2 Kgs 18:23). I here provide for those who date the Torah to the late
period the evidence that they might be able to utilize—even if they have not
been able to garner it for themselves—though I quickly supply the response
as well: these Aramaic-like features cluster in the Pentateuch in specific
texts for specific reasons, and thus cannot serve as a factor in the quest for a
date.
I do not mean to imply that a single Aramaism or Aramaic-like feature
cannot be found now and then in other sections of the Torah. Clearly there
are scattered examples, several of which I present here, each one with a
clear explanation, to my mind.

1. The verb lE;lIm spoken by Sarah in Gen 21:7 in the phrase lE;lIm yIm “who
would utter (to Abraham that Sarah would suckle sons, that I would
bear a son for his old-age).” In this case, the author reached deep into
the Hebrew lexicon, presumably plucking a word from a border dia-
lect or from Aramaic itself, in order to produce alliteration in the text.
Note how lE;lIm echoes the sounds of the roots lwm “circumcise” in Gen

32
I arrive at the number 227 in the following way: Genesis 24 consists of 67 verses; Genesis 30–31
(including 32:1–3 [see above, n. 18]) consists of 90 verses; the four Balaam oracles comprise 28 verses
(4 + 7 + 7 + 10); the material about the northern tribes in Genesis 49 includes 18 verses; and finally the
same material in Deuteronomy 33 includes 24 verses.
Hebrew Studies 47 (2006) 174 Rendsburg: Aramaic-like Features

21:4 and lmg “be weaned” (2 times) in Gen 21:8.33 Alliteration in

biblical literature has not been fully appreciated, and yet one finds
authors utilizing this technique at every turn, often in very creative
ways, as in the present instance.34
2. The noun MyIj∂r◊y “months,” which we noted above in Deut 33:14, oc-
curs again in Exod 2:2, in the birth story of Moses. In this case, the
author selected this word in order to create alliteration with the key
phrase rDmEjAb h∂rVmVjA;tÅw “and she loamed it with loam” in Exod 2:3, with
a more distant echo heard in the word rRmOj “mortar” in Exod 1:14. As
in the above example, so here: the demands of the literary device of
alliteration, which appears no less in prose than it does in poetry, sent
the author searching for the most appropriate word choice. In this
case, we actually are not dealing with an Aramaic lexeme per se, since
the word jry also occurs in Ugaritic, Phoenician, and the Gezer calen-
dar. More accurately, one should say that the author, presumably a
Jerusalemite, selected this word from a different dialect of Hebrew,
most likely one spoken in the north, which shared this lexical isogloss
with Phoenician and Aramaic. As such, MyIj∂r◊y in Deut 33:14 may not
be an Aramaic-like feature per se, but its presence in the blessing to
Joseph remains relevant nonetheless.
3. Earlier we noted that MyIfDh√r “troughs” occurs twice in Genesis 30 as
part of the Aramaic coloring of the Jacob and Laban story. The noun
occurs again in Exod 2:16, in the story of Moses in Midian, in a text
and indeed a book otherwise devoid of Aramaisms. What function
could it serve here? I would propose that this vocable serves the in-
tertextual purpose of the author, who desires to bring the reader back
to another instance of an ancient Israelite bachelor who fled his
homeland for a foreign land, who encountered a female at a well, who
watered her flock, and who came to marry her and live with his
father-in-law. The typescene studied by Robert Alter requires very

33
Elsewhere the root llm “speak, utter” occurs in the Hebrew portions of the Bible only in (late) poetic
texts: Ps 106:2; Job 8:2; 33:3. The latter two presumably are part of the overall Aramaic-like style of Job,
while Psalm 106 is a late psalm where Aramaic influence is to be expected; note, for example, the root jbv
“praise” in Ps 106:47, which also refers to the ingathering of the people from among the nations. On the
root jbv, see A. Hurvitz, NwCll NwCl Nyb (Jerusalem: Bialik, 1972), pp. 88–91.
34
I have noted instances of the employment of rare words to produce alliteration, in both prose and poetry,
in previous publications, though only now is my first sustained essay on the subject to appear: G. A.
Rendsburg, “Alliteration in the Exodus Narrative,” in Shalom Paul Festschrift, ed. C. Cohen, A. V.
Hurowitz, A. Hurvitz, Y. Muffs, B. J. Schwartz, and J. H. Tigay (Winona Lake, Ind.: Eisenbrauns,
forthcoming). The example presented in the next paragraph, MyIj∂r◊y in Exod 2:2, is among the passages
treated in my forthcoming article.
Hebrew Studies 47 (2006) 175 Rendsburg: Aramaic-like Features

little lexical linkage to allow the reader to see such connections35 —

nonetheless, the author of Exodus 2 went the extra step and included
the key word MyIfDh√r in his story, just as other key words and expres-
sions, such as bwøf_yI;k in Exod 2:2 and hDbE;t in Exod 2:3, 5, bring the
reader back to other parts of the Genesis story.
4. The particle NEh “if” occurs in Lev 25:20, clearly an Aramaic form, in-
stead of the regular Hebrew form MIa. As I will show in a forthcoming
article,36 the jubilee pericope at the end of the book of Leviticus has
three other atypical grammatical forms, all of which are to be classi-
fied as Israelian Hebrew features: a) the infinitive absolute høn∂q con-
tinuing a finite verb (Lev 25:14); b) the vocalization føoVm (Lev 25:16);
and c) the form tDcDo (Lev 25:21). For reasons that I leave for others to
explain, this section of the Torah, and only this one, I might add,
appears to stem from northern Israel.

There may be other Aramaisms lurking here or there in the Torah, but
regardless, I believe that my survey is more or less complete. Some words
that have been labeled as Aramaisms37 turn out to appear in Canaanite texts
as well, such as fyI;lAv in Gen 42:6, which occurs in Ugaritic;38 PRlEj in Numb
18:21, 31, the root of which occurs in Phoenician;39 and sRkRm in Numbers
31 (6 times), which finds a congener in the nomen agentis or participle
form skm in Punic.40 Though even if one were to accept these terms as
Aramaisms (see the information conveyed in notes 39–40), still we would
be dealing with isolated occurrences of such items, with no significant
concentration present.
In conclusion, if the Pentateuch were the product of Persian-period
Jewish scribes, as claimed in prominent places during the last several dec-

35
R. Alter, The Art of Biblical Narrative (New York: Basic Books, 1981), pp. 52–58.
36
G. A. Rendsburg, “ynwpx rwbyjk (dk–j ,hk arqyw) lbwyh tnC ofq,” in Avi Hurvitz Festschrift, ed. S. E.
Fassberg and A. Maman (forthcoming).
37
See, most prominently, in M. Wagner, Die lexikalischen und grammatikalischen Aramaismen im
alttestamentlichen Hebräisch (BZAW 96; Berlin: Alfred Töpelmann, 1966), pp. 56, 76, 114.
38
G. del Olmo Lete and J. Sanmartín, Diccionario de la lengua ugarítica, 2 vols. (Aula Orientalis
Supplementa; Barcelona: Editorial AUSA, 1996–2000), p. 440. Alternative interpretations of Ugaritic šlyt.
are possible (see the dictionary entry for the word as the Š-stem of the root lyt.), but a connection with
Hebrew fyI;lAv remains the simplest explanation.
39
C. R. Krahmalkov, Phoenician-Punic Dictionary (OLA 90; Leuven: Peeters, 2000), p. 185. One could
argue, of course, that the root Plj “exchange” and its nominal forms passed from Aramaic into both
Hebrew and Phoenician as a result of the considerable trade between and among the three groups of
speakers. If this were the case, then PRlEj in Num 18:21, 31 would be a true Aramaism, though again, one
relatively isolated in the Torah.
40
C. R. Krahmalkov, Phoenician-Punic Dictionary, p. 281. The point made above for PRlj E may be
repeated here for sRkmR ; note that once more we are dealing with a lexeme from the semantic field of trade.
Hebrew Studies 47 (2006) 176 Rendsburg: Aramaic-like Features

ades, one would expect Aramaisms or Aramaic-like features to appear

throughout its 187 chapters in significant concentrations, and not, as per the
main conclusion of this essay, in select chapters for specific purposes.41

41
My thanks to both Ziony Zevit and Clinton Moyer for their comments on an earlier version of this
paper—though I alone, of course, remain responsible for the final version.