Non-Roman fonts used: SPTiberian, SPIonic, SPAtlantis

This article is also available in transliteration and text-only formats.

J. Hoftijzer and K. Jongeling. Dictionary of the North-West Semitic Inscriptions. 2 vols. Handbuch der Orientalistik, part 1: Nahe und der Mittlere Osten, 21.1-2. Leiden/New York/Cologne: E. J. Brill, 1995. ISBN: 90-04-09817-8. Pp. 1266. US $406.00.

1. In 1960, E. J. Brill began publishing its landmark Dictionnaire des inscriptions sémitiques de l'ouest (DISO), ably edited by Jacob Hoftijzer and the late Charles F. Jean. Fascicles I/II appeared in 1960, fascicle III in 1962, and fascicles IV/V in 1965. Being a product of the pre-word-processor era, DISO suffered several drawbacks. For example, lexemes from texts published after the completion of fascicles I and II never made it into the later fascicles. And since there has been a virtual explosion of new material since 1960, DISO has become more and more outdated with the publication of every new inscription.

2. Aware of this problem for some time, Hoftijzer and Jongeling have been working hard to remedy this situation. Their long-awaited update to DISO is a careful, thorough, state-of-the-art reference tool, something which should meet the needs of biblical scholars, epigraphers, textual critics, and lexicographers for years to come. As in DISO, each word in DNWSI is subdivided into two broad categories: (1) an enumeration of the term's various grammatical forms, conveniently grouped by language (and sometimes dialect); and (2) a one-word translation, offered after giving special attention to contextual data (and often even cross-referenced to other lexical forms in DNWSI).

3. The languages referenced by this dictionary include the following: Old Canaanite (including the glosses found in the El Amarna tablets and other texts); Phoenician (including Official Phoenician, Byblos Phoenician, the magical texts from Arslan Tash, some Cyprus texts, and a few Phoenician texts from Egypt); Punic (including texts written in non-Semitic script, and even a few Punic words attested in some literary Greek and Latin texts); Moabite (the Mesha and Kerak inscriptions plus various inscriptional fragments and seal inscriptions); Ammonite (the Tell Siran bottle inscription plus various ostraca and seal inscriptions); Hebrew (the Lachish letters, the Gezer calendar, the Samaria ostraca and all other relevant epigraphical evidence up to c. 300 CE); Deir Alla (the plaster texts about Balaam bar Beor, whose idiosyncratic dialect "is neither strictly Canaanite nor strictly Aramaic" [p. xii]; formerly Hoftijzer held the Deir Alla texts to be Aramaic [J. Hoftijzer and G. van der Kooij, Aramaic Texts from Deir Alla (Leiden: Brill, 1976): 300-302]).

4. Other languages include Samalian (based on three texts whose linguistic peculiarities, while similar to Aramaic, set them apart from Old Aramaic texts); Old Aramaic (including the Tell Fekheriye inscription, Sefire, Zakkur, and other Aramaic inscriptions prior to 700 BCE); Official Aramaic (everything else in Aramaic up to 300 CE, including the syllabic cuneiform text from Warka, the Hermopolis papyri, and the Aramaic ideograms found in Parthian, Pehlevic, Sogdian, and Chwarezmian texts); Nabatean (all Aramaic texts written in some form of Nabatean script); Palmyrenean (all Aramaic texts written in some form of Palmyrenean script); Hatra (all texts from Hatra, as well as later Aramaic texts from Assur); Waw (selected early Syriac terms from an important collection of amulets); and Jewish Aramaic (including the Aramaic papyri from Elephantine and all other known Jewish Aramaic epigraphical texts).

5. As in DISO, DNWSI omits divine names, personal names, and geographical names. Syriac epigraphical material is, with a few exceptions, excluded from consideration, and no attempt is made to include material from Qumran, Ugarit, or rabbinic Jewish sources, since dictionaries and concordances for each of these well-worked areas are already in print. An unavoidable, yet unfortunate, problem is that no material published since 1991 is included in this dictionary. Therefore, no reference is made to the recently-discovered 9th-8th century Aramaic inscription from Tel Dan, with its haphel use of Klm (line 4, [A. Biran and J. Naveh, Israel Exploration Journal 45 (1995): 12], attested elsewhere only in the Zakkur inscription [KAI 202A.3]), its predilection for the so-called "short imperfect" verb form (lines 2, 3, 5, 6, and 9) and its clear reference to the House of David (line 9, dwdtyb; cf. also the phrase <sem>)r)l dwdh</sem> on the Mesha inscription, which should now probably be read, with J. C. L. Gibson, as "lion of David" [Syrian Semitic Inscriptions (Oxford: Clarendon, 1971) I.76, line 12]).

6. Text critics will be especially grateful for this reference tool. One example of its utility is its applicability to the famous textual crux in Ruth 2:7. Morphological discussion of this passage usually centers on the problem of what to do with htb#$. MT h@t@fb;#$i can be parsed as Qal inf + 3rd fem sg suff of b#$ayF, "to sit"; thus h@t@fb;#$i hzE might be translated something like "this, her sitting." It can also be parsed as the segholate noun tbe#$e + 3rd fem sg suff, and translated "her ceasing." Or similarly, the consonantal text can conceivably be repointed htfb;#$f and parsed as a Qal perf 3rd fem sg of tba#$f, "to stop, cease," translating "she has ceased."

7. All of these possibilities find support in the versions. The Targ ()btyd) reads a Peal fem ptc + prefixed particle d of bty, "to sit." The OG (ou0 kate/pausen) reads a Qal perf 3rd fem sg of tb#$, "to cease," then reads the hz preceding it as a negative (or simply inserts a negative and ignores hz). The Vg (<lat>reversa est</lat>) reads a Qal perf 3rd fem sg of bw#$, "to return."

8. The underlying problem, of course, is how to define terms composed of the Hebrew radicals t]h/b[w]#$[. DNWSI provides immediate access to pertinent examples of this same problem in the inscriptional data. In the Karatepe inscription, for example (KAI 26 ii 7-8, 13; cited in DNWSI, p. 473), Azitiwada recounts several reasons why a Cilician city ought to be named after him. One of these reasons is his ability to give its inhabitants a "pleasant dwelling" (<sem>$bt n(mt</sem>). DNWSI, citing this reference under the root <sem>y$b</sem>, lists the possibility that the Phoenician term should be derived from <sem>$bt</sem>, "to cease, rest," as well as <sem>y$b</sem>, "to sit, reside." In other words, <sem>$bt</sem> at Karatepe poses the same morphological and semantic ambiguities in its context as does htb#$ in Ruth 2:7. Seeing polysemantic ambivalence at Karatepe therefore better helps explain the versional options and strengthens the possibility that polysemantic variability might also be occurring in Ruth 2:7.

9. Biblical scholars, textual scholars, epigraphers, and lexicographers everywhere should be very grateful to Hoftijzer and Jongeling for producing such a fine reference work. The fact that they chose to produce it in English is an added plus to English-speaking readers!

© TC: A Journal of Biblical Textual Criticism, 1996.

Michael S. Moore
Fuller Seminary Southwest