BEYOND BABEL
Resources for Biblical Study
Steven L. McKenzie
Editor
Number 42
BEYOND BABEL
A Handbook for Biblical Hebrew and Related Languages
BEYOND BABEL
A Handbook for Biblical Hebrew
and Related Languages
Edited by
John Kaltner and Steven L. McKenzie
Society of Biblical Literature
Atlanta
BEYOND BABEL
A Handbook for Biblical Hebrew and Related Languages
Copyright © 2002 by the Society of Biblical Literature
All rights reserved. No part of this work may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopying and recording, or by means of
any information storage or retrieval system, except as may be expressly permitted by the 1976
Copyright Act or in writing from the publisher. Requests for permission should be addressed
in writing to the Rights and Permissions Office, Society of Biblical Literature, 825 Houston Mill
Road, Atlanta, GA 30329 USA.
Library of Congress Cataloging-in-Publication Data
Beyond Babel : a handbook for biblical Hebrew and related languages / edited by Steven
McKenzie & John Kaltner.
p. cm. — (Resources for biblical study ; no. 42)
ISBN 1-58983-035-0
1. Middle Eastern philology. 2. Semitic philology. 3. Middle East—Languages—
Grammar, Comparative. 4. Middle Eastern literature—Relation to the Old Testament
5. Bible. O.T.—Criticism, interpretation, etc. I. McKenzie, Steven L., 1953- II. Kaltner, John,

Peggy L. Day, University of Winnipeg
class="bi x0 y0 w1 h1"
PREFACE
The intent behind this book is to provide a general orientation to the
languages of importance for the study of the Hebrew Bible for readers who
have not had detailed exposure to those languages. We hope that the book
will be particularly useful to students who are just beginning their aca-
demic careers in the study of the Hebrew Bible. But it should also find an
audience among those who have not had detailed exposure to one or
more of the languages discussed here and who would like to cultivate at
least a rudimentary acquaintance with it or them. The chapters do presup-
pose familiarity with biblical Hebrew, although we have included a chapter
on biblical and inscriptional Hebrew that situates this material within its
broader linguistic context. Indeed, many readers may find it helpful to
begin with this chapter before moving to less-familiar territory.
The languages treated here are those that, in our estimation, are the
most significant for the study of the Hebrew Bible for purposes of com-
parative grammar and lexicography or for comparative history and
literature, or both. Other languages might have been included. We consid-
ered including a chapter on Sumerian but ultimately decided that, given
our readership, the linguistic and literary connections with the Hebrew
Bible were not strong enough to warrant a separate chapter. Greek litera-
ture is increasingly cited in recent Hebrew Bible scholarship for its
comparative value. However, we deemed it most appropriate to reserve
it—along with other languages that are especially important in textual crit-
icism (Syriac included)—for treatment in a potential companion volume
dealing with the New Testament. Failing such a volume, and granted a sec-
ond chance (or edition) of the present work, the addition of Greek and
Sumerian, and possibly other languages, may be appropriate.
As authors for each chapter we sought specialists with proven records

LBH Late Biblical Hebrew
LXX Septuagint
MA Middle Arabic
masc. masculine
MH
Mishnaic Hebr
Mishnaic Hebr
ew
ew
MT
MT
Masoretic Text
NH New Hittite
nom. nominative
NS New Script
OH Old Hittite
OS Old Script
OSA Old South Arabian
PBEH postbiblical epigraphic Hebrew
PBH postbiblical Hebrew
PIE (Proto) Indo-European
pl. plural
PN personal name
PS Proto-Semitic
SBH Standard Biblical Hebrew
sg. singular
V vowel
PRIMARY SOURCES
Ant. Josephus, Jewish Antiquities
b. Babylonian Talmud

ArOr Archiv Orientální
AUSS Andrews University Seminary Studies
BA Biblical Archaeologist
BaghM Baghdader Mitteilungen
BAR Biblical Archaeology Review
BASOR Bulletin of the American Schools of Oriental Research
BDB Brown, F., S. R. Driver, and C. A. Briggs, A Hebrew and
English Lexicon of the Old Testament. Oxford: Oxford
University Press, 1907.
BHS Biblia Hebraica Stuttgartensia. Edited by K. Elliger and
W. Rudolph. Stuttgart: Deutsche Bibelgesellschaft, 1983.
Bib Biblica
x ABBREVIATIONS
BJS Brown Judaic Studies
BO Bibliotheca orientalis
BZAW Beihefte zur Zeitschrift für die alttestamentliche Wis-
senschaft
CAT The Cuneiform Alphabetic Texts from Ugarit, Ras Ibn Hani
and Other Places. Edited by M. Dietrich, O. Loretz, and
J. Sanmartín. Münster: Ugarit-Verlag, 1995.
CBQ Catholic Biblical Quarterly
CBQMS Catholic Biblical Quarterly Monograph Series
CHD The Hittite Dictionary of the Oriental Institute of the Uni-
versity of Chicago. Edited by Harry A. Hoffner Jr. and
Hans G. Güterbock. Chicago: Oriental Institute, 1980–.
CIS Corpus inscriptionum semiticarum
CTH Catalogue des textes hittites
CTA Corpus des tablettes en cunéiformes alphabétiques décou-
vertes à Ras Shamra-Ugarit de 1929 à 1939. Edited by
A. Herdner. Mission de Ras Shamra 10. Paris: Imprimerie

JAOS Journal of the American Oriental Society
JBL Journal of Biblical Literature
JCS Journal of Cuneiform Studies
JDS Judean Desert Studies
JEA Journal of Egyptian Archaeology
JEOL Jaarbericht van het Voorasziatisch-Egyptisch Gezelschap
(Genootshhap) Ex oriente lux
JFSR Journal of Feminist Studies in Religion
JHNES Johns Hopkins Near Eastern Studies
JNES Journal of Near Eastern Studies
JQR Jewish Quarterly Review
JSOTSup Journal for the Study of the Old Testament: Supplement Series
JSS Journal of Semitic Studies
JTS Journal of Theological Studies
KAI Kanaanäische und aramäische Inschriften. H. Donner and
W. Röllig. 2d ed. Wiesbaden: Harrassowitz, 1966–1969.
KBo Keilschrifttexte aus Boghazköi
KTU Die keilalphabetischen Texte aus Ugarit. Edited by M. Diet-
rich, O. Loretz, and J. Sanmartín. AOAT 24/1. Neukirchen-
Vluyn: Neukirchener Verlag, 1976.
KUB Keilschrifturkunden aus Boghazköi
LÄ Lexikon der Ägyptologie. Edited by W. Helck, E.Otto, and
W. Westendorf. Wiesbaden: Harrassowitz, 1972–.
LCL Loeb Classical Library
Lg Language
MDAI Mitteilungen des Deutschen archäologischen Instituts
Mus Muséon:Revue d’études orientales
NAWG Nachrichten (von) der Akademie der Wissenschaften in
Göttingen
NRSV New Revised Standard Version

SubBi Subsidia biblica
TA Tel Aviv
TynBul Tyndale Bulletin
UF Ugarit-Forschungen
UT Ugaritic Textbook. C. H. Gordon. AnOr 38. Rome: Pontif-
ical Biblical Institute, 1965.
VT Vetus Testamentum
ZA Zeitschrift für Assyriologie
ZAH Zeitschrift für Althebräistik
ZÄS Zeitschrift für ägyptische Sprache und Altertum
ZAW Zeitschrift für die alttestamentliche Wissenschaft
ZDMG Zeitschrift der deutschen morgenländischen Gesellschaft
ZDPV Zeitschrift des Deutschen Palästina-Vereins
ZfE Zeitschrift für Ethnologie
ZPE Zeitschrift für Papyrologie und Epigraphik
ABBREVIATIONS xiii
class="bi x0 yc1 w2 h16"
INTRODUCTION
John Huehnergard
1. THE STUDY OF NEAR EASTERN LANGUAGES
IN BIBLICAL SCHOLARSHIP
In their quest to understand the text of the Hebrew Bible, students and
scholars have for centuries turned to other Near Eastern languages. Already
in the Middle Ages, Jewish exegetes and grammarians compared obscure
Hebrew words and roots with similar forms in the Arabic spoken in their
surroundings and with Aramaic forms with which they were familiar. The
rise of critical biblical scholarship in Europe some two centuries ago coin-
cided roughly with the beginnings of comparative and historical linguistics;
although the latter was founded on the basis of the Indo-European lan-
guages, its methods were soon also applied to the Semitic languages, and

classical forms of Aramaic and Ethiopic, all of which have been the sub-
ject of a continuous tradition of study, and unlike, say, Akkadian,
Egyptian, and Ugaritic, languages that had been completely forgotten and
that had to be recovered or reconstructed in toto when they were redis-
covered.) There are other, related difficulties in the study of biblical
Hebrew, including (1) the relatively small size of the corpus of biblical
Hebrew (so that many words that may have been quite common in the
spoken language appear only sporadically and are consequently difficult
to interpret with confidence);
1
(2) the presence in the corpus of diverse
genres, including poetry, narrative prose, aphorisms, and the like; (3) the
long chronological span covered by the corpus, nearly a millennium, dur-
ing which time the spoken language undoubtedly underwent at least
some change; (4) the likely existence in the corpus of diverse dialects in
addition to the standard Jerusalem literary dialect in which most of the
text was written. The study of other languages and of other forms of
Hebrew (especially Mishnaic, for which see the chapter on postbiblical
Hebrew) provides an awareness of these problems and, sometimes, solu-
tions, as is also abundantly illustrated in each of the subsequent chapters
of this book.
2. OVERVIEW OF THE SEMITIC LANGUAGE FAMILY
Hebrew is a member of the Semitic language family. Other members of
the family that are described in detail in the present volume are Akkadian,
Ugaritic, Phoenician, Moabite, Ammonite, Edomite, Aramaic, and Arabic.
Still other Semitic languages are Eblaite, a cuneiform language, closely
related to Akkadian, attested in third-millennium texts from the city of Ebla
(in present-day Syria); the various Old (or Epigraphic) South Arabian lan-
guages, which are attested from the eighth century B.C.E. until the sixth
century C.E. (these languages—Sabaic, Minaic or Madhabic, Qatabanic, and

Semitic in turn is further subdivided into the Modern South Arabian branch,
the Ethiopian branch, and a third branch called Central Semitic. The latter
comprises the Old South Arabian languages,
2
Arabic, and the Northwest
Semitic languages. The Northwest Semitic subbranch includes Ugaritic,
Aramaic, and the Canaanite languages, which are Phoenician (and Punic),
Moabite, Ammonite, Edomite, and, finally, Hebrew. According to this clas-
sification, therefore, Hebrew’s closest relatives, the languages with which
it most recently shared a common ancestor, are, first, the other Canaanite
languages (note that in Isa 19:18 Hebrew is called ˆ['n"K] tp'c] “the language
JOHN HUEHNERGARD 3
2
Until recently, Old South Arabian has been grouped with Modern South Arabian
and with Ethiopian Semitic. See, however, Norbert Nebes, “Zur Form der Imperfekt-
basis des unvermehrten Grundstammes im Altsüdarabischen,” in Semitische Studien
unter besonderer Berücksichtigung der Südsemitistik (vol. 1 of Festschrift Ewald
Wagner zum 65. Geburtstag; ed. W. Heinrichs and G. Schoeler; Beirut and Stuttgart:
Steiner, 1994), 59–81; Victor Porkhomovsky, “Modern South Arabian Languages from
a Semitic and Hamito-Semitic Perspective,” Proceedings of the Seminar for Arabian
Studies 27 (1997): 219–23; and Rainer Voigt, “The Classification of Central Semitic,”
JSS 32 (1987): 1–21.
of Canaan”), followed by the remaining Northwest Semitic languages
(essentially, Aramaic and Ugaritic), and then Arabic and the Old South
Arabian languages.
In addition to factors of genetic proximity, account must also be taken
of cultural and historical considerations. For example, for much of the
second millennium B.C.E., and into the first, Akkadian was a lingua franca
throughout the Near East, that is, a language used for communication
among peoples speaking different languages, and there are as a result a

they spoke in Hebrew letters (Judeo-Arabic).
No writing system records every significant feature of a language. Dif-
ferent systems are more successful in noting some features, less successful
in others. The early Phoenician alphabet, for example, presumably
recorded each of the consonants of the language discretely but gave no
indication of the vowels. Phoenician had fewer consonants than ancient
Hebrew, and when speakers of Hebrew borrowed the Phoenician alpha-
bet they had to press at least one symbol into service to represent more
than one sound, namely,
ç for what the Masoretes later differentiated as c
sx and v ss (there were probably a few other such double-duty letters in
early Hebrew; see below and the article in this volume on biblical
Hebrew). In Mesopotamian cuneiform, on the other hand, vowel quality
(and sometimes, but not regularly, vowel quantity) was indicated, but the
system was not well adapted for the clear differentiation of series of
homorganic consonants (i.e., consonants pronounced at the same place in
the mouth, such as the labials, voiced b and voiceless p; thus, the syllables
ab and ap were always written with the same sign).
Both because of the inadequacies of native writing systems and
because of their diversity, scholars find it useful to transliterate the vari-
ous languages into a common system. This allows the details of the
phonology and grammar of individual forms to be represented clearly,
and it also greatly facilitates the comparison of forms across languages.
The linguistic similarity of Hebrew
[m'v;, Syriac :amש, Arabic b0≥ØX0, and
Ethiopic SM[, all meaning “he heard,” is obviously much more transpar-
ent when those forms are transliterated, respectively, as ssaama(, ssma(,
sami(a, and sam(a. Western scholars specializing in the study of the
Semitic languages have long used a relatively uniform system for translit-
erating the sounds into the Latin alphabet, using special diacritics for

(The IPA representation of these consonants depends on their actual pro-
nunciation in the various Semitic languages. In the modern Ethiopian
Semitic languages, they are glottalic, thus IPA [t’] for the consonant that cor-
responds to Hebrew f; in Arabic, they are pharyngealized, e.g., IPA [t≥].)
(3) An underdot is also used in h˙ for j. As we will see below (sec-
tion 4, end), Hebrew j, like [, reflects the merger of two originally
distinct Semitic consonants (which also remain distinct in Ugaritic and
Arabic), the voiceless pharyngeal fricative h˙ (IPA [4]) and a voiceless velar
fricative, which is transliterated by Semitists as hH (i.e., a “hooked h”; in
IPA, this is [x]).
(4) The sound “sh” denoted, for example, by Hebrew v, is tradition-
ally transliterated by ss (i.e., s with a “wedge” or “hachek”; in IPA, the
symbol for “sh” is [S]). The Semitistic transliteration of Hebrew c is sg (i.e.,
s with acute). (Traditionally, c is pronounced the same as s, IPA [s]; the
probable ancient pronunciation of c is a voiceless lateral fricative, IPA [l].)
(5) The spirantized variants of the bgdkpt consonants are frequently
not indicated specially in transliteration; if notation of the spirantization is
important, however, this may be done with underlining (or, in the case of
g and p, an overline instead), as in kaat
t
ab
d
for bt'K;.
(6) For Proto-Semitic and for some of the Semitic languages, short vow-
els are written with no diacritic (thus, a, i, u), while long vowels are
6 INTRODUCTION
5
See International Phonetic Association, Handbook of the International Phonetic
Association: A Guide to the Use of the International Phonetic Alphabet (Cambridge:
Cambridge University Press, 1999).

For a variety of reasons, speakers do not pronounce their language in
exactly the same way as those from whom they learned it. One of the most
important—and surprising—aspects of sound change is that it is regular
and can be described by rules. As an example, consider forms of biblical
Hebrew such as kesep “silver,” kaspî “my silver,” and (ebed “servant,” (abdî
“my servant”; compare those with Akkadian kaspum “silver,” kaspıi “my sil-
ver,” and Arabic (abdun “servant,” (abdıi “my servant,” which suggest that
the original bases of these words were *kasp- and *(abd- (an asterisk, *, is
used to indicate a form that has been reconstructed for the common ances-
tral language, or protolanguage). The Hebrew “segholate” forms kesep and
(ebed show two phonological developments: the change of the original
vowel *a between the first and second consonants to e, and the insertion
JOHN HUEHNERGARD 7
(anaptyxis) of a vowel e between the second and third consonant. These
developments may be written as rules, as follows:
(1) a > e / C_CC# (that is, “a becomes [>] e in the following environ-
ment [/]: after a consonant and before two consonants at the end of a
word [the underline _ indicates the position of the sound in question;
# indicates a morpheme or word boundary]”; thus, e.g., *kasp > *kesp,
*(abd > *(ebd )
(2) ø > e / C_C# (that is, “nothing/zero becomes e [or, e is inserted]
between two consonants at the end of a word”; thus, *kesp > kesep,
*(ebd > (ebed; note that these two rules operate sequentially, rule 2
operating on the result of rule 1).
6
As noted earlier, sound rules are regular; that is, they operate without
exceptions. Thus, any base inherited into Hebrew with the form CaCC is
expected, when it has no suffix, to become CeCeC. When we do find
exceptions, such as *bayt- > bayit “house” or *ba(l- > ba(al “lord” (rather
than the forms that our rules would generate, **beyet and **be(el [a double

.
he(e´lîtuunû is the result of an analogy between the third-person masculine
singular and plural forms with a suffix, on the one hand, and the corre-
sponding second-person masculine singular and plural forms; it may be
represented as follows, using a sound verb to show the forms more clearly:
ssémaaraanû : ssémaaruunû :: ssémaartaanû : X = ssémaartuunû,
in which the simple change of aa to uu that characterizes the change of sin-
gular to plural in the third-person forms on the left is extended to the
second-person forms on the right.
A subcategory of analogical change is leveling, by which a paradigm
is made more uniform. An example in English is the generalization of /s/
to mark the plural, where once there were several ways in which plurals
were formed (preserved vestigially in forms such as oxen and geese). An
example in Hebrew is found in some verb paradigms: in the perfect con-
jugation of h˙aapeesß “to delight in,” many of the forms have patah˙ rather
than sßeerê in the second syllable, the result of a sound rule (called Philippi’s
Law: a stressed *i becomes sßeerê, ee, when in an originally open syllable, but
patah˙,a,in an originally closed syllable; thus *h˙apíˇ˛a > h˙aapeesß, but
*h˙apíˇ˛ta > h˙aapasßtaa); but in some verbs that had *í in the second syllable
originally, the third-person masculine singular also has patah˙ rather than
the expected sßeerê, as the result of leveling, for example, in qaarab “he
approached” (rather than *qaareeb; the expected ee appears in the pausal
form qaareebâ “she approached” [Zeph 3:2]). The same leveling is responsi-
ble for the examples of third masculine singular pi(el perfects with patah˙
in the second syllable, such as giddal “he made great” (Josh 4:14). Unlike
sound change, analogical change (including leveling) is not regular and
predictable; the mere availability of an analogy does not always trigger a
new development. Thus, for example, we find h˙aapeesß (Gen 34:19) rather
than **h˙aapasß and the expected giddeel (Isa 49:21) as well as giddal.
A third type of linguistic change is semantic change. The meanings of

18:4); the former has the form of a pu(al perfect, the latter of a hop(al
imperfect (jussive), yet, as common as this root is in biblical Hebrew,
there are no attested pi(el or hip(il forms that correspond to them; nor
are there any pu(al imperfects or hop(al perfects of this verb. This
unusual distribution has suggested to scholars that the forms luqqah˙ and
yuqqah˙ may in fact not be pu(al and hop(al but rather vestiges of an old
qal passive conjugation that was, perhaps, no longer recognized as such
by the Masoretes.
In comparative reconstruction, as the name suggests, cognate forms
of related languages are compared in an attempt to get at an earlier stage.
Cognates are forms, such as Hebrew kesep and Akkadian kaspum (see
above), that exhibit a similar form and meaning and that can be shown
to share a common ancestor. Hebrew kesep and Akkadian kaspum clearly
have similar meanings, and their root consonants, k-s-p, are identical. Let
us now consider Hebrew pétah˙ and Akkadian pete, both of which are
imperatives meaning “open!”; the forms seem to be cognate (they are
similar in form and meaning), but here Akkadian is lacking the third root
consonant of the Hebrew form. A similar correspondence is shown by
other pairs, such as Hebrew h˙aÅmôr and Akkadian imeerum “donkey,” and
Hebrew zebah˙ “sacrifice” and Akkadian zıibum “food-offering.” In these
word sets we see an apparently regular correspondence of Hebrew h˙ and
the lack of a consonant in Akkadian. But in other instances, Hebrew h˙
corresponds to Akkadian hh: Hebrew h˙aameess and Akkadian hhamiss “five”;
Hebrew )aah˙ and Akkadian ahHum “brother”; Hebrew mooah˙ “marrow” and
10 INTRODUCTION
8
Avi Hurvitz, The Transition Period in Biblical Hebrew: A Study in Post Exilic
Hebrew and Its Implications for the Dating of Psalms (Hebrew) (Jerusalem: Bialik
Institute, 1972), 22 and n. 25; Mark F. Rooker, Biblical Hebrew in Transition: The
Language of the Book of Ezekiel (JSOTSup 90; Worcester: Sheffield, 1990), 127–31.

trate the range of such data available to the student of Hebrew who is
interested in this field of study.
5.1. PHONOLOGY
Proto-Semitic (PS), the ancestral language from which all of the
attested Semitic languages descend, had twenty-nine consonants (all of
which remain distinct in the Old South Arabian languages). In biblical
Hebrew, which had only twenty-three consonants, some of the original
JOHN HUEHNERGARD 11
9
In fact, certain types of evidence suggest that the distinction between*hH and *h˙
was preserved in Hebrew for much of the biblical period. See the discussion in the
chapter, “Hebrew (Biblical and Epigraphic).”
10
See HALOT, 1:295–96.