Journal of Hebrew Scriptures - Volume 16 (2016) - Review

1. Introduction

Since the 1990s a debate has raged between Hebraists and biblicists over the history of ancient/Biblical Hebrew (BH) and the prospect of dating the writings of the Bible in their Masoretic Text (MT) dress on the basis of linguistic criteria.[1] The book under consideration, A Concise Lexicon of Late Biblical Hebrew, by the distinguished Israeli scholar Avi Hurvitz, may be regarded as the magnum opus of the scholar who has led the field for nearly half a century.[2] Our aim in this review is to evaluate critically the Lexicon on the basis of its general and specific contents and its explicit and implied objectives.[3]

2. Contents

The bulk of the Lexicon consists of eighty entries, alphabetized, on individual words and expressions (25–243).[4] Most of these are two or three pages in length, a handful are four pages, two are one page, and one is six pages (מַלְכוּת). The entries are organized according to the “Structure of the Entries” (14–7) at the beginning of the volume. This section uses the word אִגֶּרֶת as its example, so effectively this noun is discussed twice in the Lexicon (14–7, 25–7). The treatment of each word/expression is methodical. First, each entry opens with the late “Lemma” (e.g., אִגֶּרֶת) accompanied by its respective linguistic form (n.f. = noun feminine), gloss (“letter, …”), and biblical references (Neh 2:7, etc.) with brief quotations of text. With one exception (cf. §5.7.4) the biblical references for the late form/use are exhaustive. Twenty-three entries are also cross-referenced to other related entries in the Lexicon (e.g., אֲדַרְכֹּן refers to דַּרְכְּמוֹן, and vice versa). Second, when there is a cognate word in “Biblical Aramaic (BA)” it is listed with its biblical references and brief quotations of text. Thirty-one entries list Aramaic cognates. Third, since the late variants “supplant” earlier words/expressions, “Classical Biblical Hebrew (CBH) Alternatives” are listed (e.g., סֵפֶר) with a selection of references and quotations of text. The latter usually juxtapose early and late texts with the early and late variants, respectively, such as a text from Samuel–Kings with an early variant alongside a text from Chronicles with its late alternative, though the texts cited are usually non-synoptic. Fourth, after presenting the relevant biblical data, each entry shifts to “External Post-Classical Sources,” that is, extra-biblical written sources dated to the Second Temple and Rabbinic periods. These are divided into two main groups of sources, “Renderings/Paraphrases/Glosses” (e.g., Aramaic Targums) and “Independent Use” (i.e., independent of a direct biblical influence, e.g., Ben Sira, the Dead Sea Scrolls [DSS], Mishnah), and a selection of references and quotations complements each entry. Fifth, “Comments” are made on the item under consideration and as a rule they incorporate quotations from the scholarly literature. Sixth, each entry concludes with a selective “Bibliography.”

The eighty entries in the Lexicon are preceded by a “Prolegomenon” (1–13). The brief “Introductory Remarks” (1–2) stress the distinction between First and Second Temple BH, periods separated by a “critical turning point” or “watershed” in the history of the Hebrew language in the fifth century b.c.e., after which many linguistic innovations are attested for the first time in biblical writings, and the conventional periodization of BH as Archaic, Classical, and Late is affirmed. The second and longest part of the introduction, “The Linguistic and Historical Setting of LBH” (2–11), reasserts the conventional tripartite division of BH and it focuses on two principal topics. First, the distinctive linguistic characteristics of LBH books are discussed. The late writings in view are Second Isaiah, Haggai, Zechariah, Malachi, Qoheleth, Esther, Daniel, Ezra, Nehemiah, and Chronicles. (Transitional writings such as Jeremiah and Ezekiel, and “chronologically problematic” writings such as Joel, Jonah, Song of Songs, and Ruth, are differentiated from these books.) The characteristics outlined are Persian loanwords, Aramaic interference, Rabbinic Hebrew (RH) elements, and inner-BH developments. Second, the well-known Hurvitzian criteria for identifying distinctive late features and writings are described. These are late biblical distribution, linguistic contrast with classical alternatives, and extra-biblical corroboration in relation to late features, and accumulation of such features with reference to late writings. The short third and fourth parts of the introduction discuss “The Conceptual Plan of The Concise Lexicon of Late Biblical Hebrew” (12) and reject “The Non-Diachronic Approach to BH” (12–3).

The Lexicon is rounded out by a “Preface” (ix–x), “List of Abbreviations and Sigla” (18–23), and “Bibliography: Works Cited in the Lexicon” (244–70). The book does not have indexes, but the eighty lemmas are listed one by one in the “Contents” (vii–viii).

3. Objectives

A lexicon is “a word-book or dictionary” (OED) and a dictionary is

Already there are more than a few of such catalogs of the vocabulary of the Bible, for example BDB, DCH, HALOT, NIDOTTE, TDOT, TLOT, TWOT, and so on, to mention only the popular Hebrew-English lexicons or dictionaries. So why another one?

The Lexicon is not just another traditional lexicon. It is a specialized one whose unique subject matter is a selection of eighty late words and expressions which are attested exclusively or predominantly in “acknowledged”[5] late (i.e., postexilic, post-Classical) writings of the Bible (and other late sources). For this reason the first and main objective of the Lexicon is to catalog some items of distinctive late BH language.[6] But the Lexicon has several other purposes.

The second objective is to validate that the late items which are extracted from the late writings are in actual fact late.[7] To repeat, an item is considered late when it appears exclusively in late writings, that is, it is a late innovation, or when it occurs more intensely or predominantly in late writings, that is, it has experienced late diffusion. In either sense, therefore, a word or expression may be considered “characteristic” or “indicative” or “typical” of late BH. The confirmation of the lateness of any given item is determined on the basis of the criteria already mentioned above: distribution in the late writings of Esther, Daniel, Ezra, Nehemiah, and/or Chronicles, contrast with alternative words or expressions in acknowledged early (i.e., preexilic, Classical) writings, and extra-biblical corroboration in other late sources such as Ben Sira, the DSS, RH, Imperial Aramaic, and so on.

The third and final objective of the lexicon is to substantiate that the acknowledged early writings with none or less of the late words and expressions under consideration are in actual fact early. In other words, Hurvitz argues that the collective absence in certain (early) writings of innovations or diffusions that are evident in late writings indicates that those (early) writings are definitely early.[8]

Together the second and third objectives as described here suggest that the ultimate purpose of the Lexicon is to reinforce the linguistic dating of early writings to the early period and late writings to the late period. This is a reasonable conclusion in the light of Hurvitz's continous research and publication over his long career on the linguistic dating of biblical literature, the pivotal role that his four linguistic dating criteria (distribution, contrast, extra-biblical corroboration, accumulation) play in the Lexicon (9–11, and passim), and especially the recurring references in the Lexicon to the “dating” of biblical books, not features.[9] This aspect of the Lexicon renders it somewhat unconventional and idiosyncratic as a lexicon, whether speaking in general or of BH or of LBH.

4. Commendations

The main benefits of the Lexicon are that it provides a helpful entrance into Hurvitz's linguistic dating approach, his theory and method, gathers in one place many case studies that he published previously in different venues, and affords a good overview of some distinctive vocabulary of LBH. Altogether Hurvitz is cited in the bibliographies of 47 of 80 entries, but in fact he looks at some of the other items in earlier publications as well.[10] Likewise, much of the “Prolegomenon” (especially 2–11) is a modified version of another recent publication.[11] Consequently the book offers little that is new. The main exception is the richness of the Qumranic and Rabbinic citations that are given to support the lateness of the late words and expressions. The individual bibliographies are also useful, but they must be used cautiously since they cite only sympathetic voices (cf. §5.3). All things considered, Hebraists and biblicists who agree with Hurvitz's linguistic dating approach will be happy with this volume.

5. Reservations

5.1. Introduction. In the following we summarize some of our general reservations about the Lexicon.[12]

5.2. Lack of engagement with conventional historical linguistic theory and method. We argued above (§3) that the Lexicon's linguistic dating objective makes it rather unconventional and idiosyncratic as a lexicon. Furthermore, even though Hurvitz asserts, “our Lexicon is, by definition, diachronic in nature and thus constitutes part and parcel of the discipline of Historical Linguistics” (13), it is remarkable that the discipline of historical linguistics is mentioned only once in the volume, namely here. In addition, the bibliography fails to cite even a single title related to historical linguistics or language variation and change. The theory and method evinced in the publication are entirely Hurvitz's, and his objective is linguistic dating. Hurvitz may wish his volume to be historical linguistic in scope, but it is not, at least not in a conventional way. It might be that Hurvitz would reach a different conclusion about the efficacy of language for dating biblical writings if he were to embrace standard historical linguistic theory and method.

5.3. Limited engagement with scholarly literature on BH historical linguistics. We observed above (§4) that much of the Lexicon's content was already available in other publications by Hurvitz and others (Bergey, Polzin, Rooker, Wright, etc.). Furthermore, the “Comments” incorporate quotations from the scholarly literature, and indeed those citations constitute a large share of the discussion of each lemma. However, as mentioned above (§4), the comments, quotations, and bibliographies are restricted to other like-minded publications. Hurvitz explicitly refuses to engage or debate, even cite, any scholarship related to the so-called “non-diachronic school” or “minimalists” or “challengers.”[13] Yet in some cases other independent scholarship with historical linguistic or linguistic dating implications, and conclusions contrary to Hurvitz's, is dismissed or overlooked.[14]

5.4. Limited engagement with literary and textual research on the Hebrew Bible. Elsewhere we have argued that scholars in the area of BH historical linguistics often disregard literary-critical and text-critical matters or, rather, they act as if the MT is the “original” text of biblical writings. It would be inappropriate in this context to review in detail Hurvitz's non-linguistic assumptions and views.[15] It is interesting to observe, however, that in the Lexicon he seems open to editorial and scribal processes in the production of postbiblical writings (e.g., 57, 90, 129), but he is reticent to allow those in the case of biblical books (e.g., 96), except where Chronicles rewrites Samuel–Kings (passim). And while it is acknowledged that “[b]iblical literature was composed by different writers over an extended period of time” so that the books “differ from one another literarily (e.g., prose vs. poetry), historically (First Temple vs. Second Temple periods), and geographically” (3), literary and textual developments within individual biblical books are ignored.

5.5. Periodization and unexplored matters. Hurvitz's linguistic dating method is inseparable from the assumption of distinct language periods, transitional periods between those periods, and books “acknowledged” to have been written (more or less as they are found in the MT) in one or another of those periods (1–4). So, for example, books such as Jeremiah and Ezekiel have language that is transitional between early and late BH (3, 90, 149, 168, 219–20), and similarly Qumran Hebrew (QH) represents a transition between BH and RH (2). Elsewhere we have examined various theoretical and methodological problems with the notion of language periodization (states and transitions).[16] We will not repeat our earlier criticisms here. However, there is one peculiar aspect of the Lexicon in this regard that we wish to highlight. In this work Hurvitz speaks about the fifth century b.c.e. as the “critical turning point” (1) and “watershed” (2) in the history of BH. However, in previous publications Hurvitz has talked about the sixth century b.c.e. as the “turning point” between the successive stages of BH.[17] The reason for the change in Hurvitz's view is left unexplained. More importantly, the many potential historical and literary repercussions of this new perspective are passed over in silence.

5.6. Hurvitz's early data. The focus of the Lexicon is the language of LBH. But given the importance of his linguistic contrast principle (10) and his linguistic dating objective (§3), Hurvitz usually cites early alternatives to the late neologisms. There are various problems with the early data that are offered in the Lexicon.

5.6.1. Full distributional data for the early alternatives are not given. This remark applies to the reporting of early alternatives in, for example, both the early books of Genesis–Kings and the late books of Esther–Chronicles. It might be interesting and informative to know, for example, where the early alternatives for late כעס occur in early and especially late writings, since it is stated that “[t]he late distribution of kāʿas within the Bible faithfully reflects an actual linguistic situation” (144), but no information is offered regarding the distribution of those early alternatives in late writings.[18]

5.6.2. Sometimes certain early alternatives are cited in the section “CBH Alternatives” at the beginning of each entry and then additional alternatives are cited in the “Comments” section. It would have been helpful to have all of the early alternatives cited together in one place. For example, מִשְׁפָּט and חֻקָּה are cited on p. 101 as early alternatives for late דָּת and then later חֹק and תּוֹרָה are cited as other early alternatives (102).

5.6.3. Sometimes particular early alternatives are cited but other early alternatives that are suggested elsewhere, even by Hurvitz himself in other publications, are unmentioned. For example, הגה is cited as the early alternative to late דרשׁ + God's law (98), but elsewhere Hurvitz supplies שׁמר + God's law as the early alternative.[19]

5.6.4. Sometimes the early alternatives are so rare that one wonders in what sense they can be considered viable or meaningful alternatives rather than peculiarities or anomalies. For example, the hithpael of ילד is cited on p. 122 as the early alternative of the late hithpael of יחשׂ, but the former appears only once in the Bible, in Num 1:18.[20]

5.6.5. Sometimes no early alternatives are offered (31 [but cf. 32–3], 79, 92, 109, 145 [but cf. 146], 223, 231) or some or all of the suggested items are acknowledged as uncertain (60, 84, 221, 239). Given the importance of the principle of linguistic contrast to Hurvitz's method (10), it seems significant when early alternatives are unknown or unclear.

5.6.6. Semantic equivalence of early and late variants is always assumed or asserted, never argued, and in some cases the words or expressions under consideration may well not be synonymous.[21] For example, it seems doubtful that אֵין + infinitive and לֹא + imperfect are fully equivalent in meaning, even though there is overlap in certain uses of these constructions (36). Linguistic contrast is problematic without semantic equivalence.

5.7. Hurvitz's late data. At this point we come to the heart of the Lexicon: the selection, presentation, and explanation of late words and expressions in the core LBH books of Esther–Chronicles and elsewhere. In what follows we will advance some of our reservations concerning Hurvitz's choices and their underpinnings.

5.7.1. The criteria for the selection of the eighty items are unclear. We would like to know, for example, why Hurvitz selected those eighty words and expressions when many more late lexical items are suggested in the scholarly literature;[22] Hurvitz studies some other items elsewhere;[23] some of the eighty items do not occur in the undisputed late books of Esther–Chronicles;[24] 29 of 80 items occur only once or twice in BH, 48 of 80 occur five times or less in BH, and 67 of 80 occur ten times or less in BH; and so on. We comment more below on the distribution of Hurvitz's eighty lexical items in (L)BH. Are we to assume that the eighty items in the Lexicon are good representatives, or even the best representatives of lexical change from early to late BH in Hurvitz's thinking? If not, then why did Hurvitz pick these items and not others?

5.7.2. Many related items are treated separately and might have benefited from joint consideration or perhaps discussion according to roots or semantic fields.[25] We observed above (§2) that twenty-three entries are cross-referenced to other related entries. Sometimes the same information and/or quotation(s) are repeated verbatim in different entries. Of course such repetition is mainly due to the usual structure of a lexicon. In any case, a noteworthy example, which we discuss below (§6.5), is the treatment of the seven Babylonian months in the Bible, all of which are accompanied by identical information in the sections “CBH Alternatives” and “Comments.”

5.7.3. The Lexicon contains entries where the issue under consideration is perhaps better regarded as orthographical or grammatical rather than lexical.[26]

5.7.4. There are some issues with the documentation of late data, aside from the fact that late items are not thoroughly documented in postbiblical sources (QH, RH, etc.). The Lexicon gives complete biblical references for the eighty lemmas, with the single exception of דָּת where Esth 2:8 is overlooked (101). On the other hand, the Lexicon never reports when a lemma occurs two or more times in a single verse (e.g., דָּת twice in Esth 3:8; 101). Occasionally the inclusion or exclusion of a particular late example is debatable, as Hurvitz recognizes, for example, with regard to בהל in Prov 28:22 (46).

5.7.5. The most problematic aspect of the Lexicon is the treatment of the LBH corpus (mainly Esther–Chronicles) as a single, essentially unified linguistic entity. Five issues are involved here: variation of late variants across different books, concentration of late variants in individual books, “replacement” of early variants by late variants, “coexistence” of early variants and late variants, and the notion of “characteristic” language. In general one would probably not bring up such matters in a critical review of a lexicon; however, Hurvitz's Lexicon is a special case since these five issues have potential implications for his linguistic contrast principle (10) and his linguistic dating objective (§3). In this review we can only sketch the concerns on which we elaborate further in our separate article.[27]

5.7.5.1. Variation of late variants across different books. In theory Hurvitz is aware that different books use different items at different rates,[28] but in practice he rarely discusses the items in a way which illuminates the differences (and similarities) between the books.[29]

5.7.5.2. Concentration of late variants in individual books. The Lexicon gives no information about how many distinct late variants appear in any given book and how often they occur there.

5.7.5.3. “Replacement” of early variants by late variants. Hurvitz says many times that a late innovation “replaces” an early alternative.[30] Such “replacements” or (categorical) “contrasts” between early and late writings have special significance in Hurvitz's method:

5.7.5.4. “Coexistence” of early variants and late variants. Alongside many statements about “replacement” (cf. previous point) the Lexicon also repeatedly states that early alternatives persist in late writings, such that early variants and late variants “coexist” with one another.[31] Stated differently, late variants “begin to” or “began to” (73, 99, 102, 128, 186, 219) replace early variants but did not “altogether,” “completely,” “entirely,” “fully,” “instantly,” “systematically,” or “thoroughly” replace them (26, 47, 71, 82, 96, 131, 143, 200, 215). In short, “the emergence of a neologism does not necessarily stop the use of the older synonym” (189). Furthermore, Hurvitz believes, such early words and expressions are not part of “the living linguistic usage” (75) or “the living language” (87) or “an actual linguistic situation” (144); rather, they are nothing more than “ancient survivals or archaizing devices” (144) or “survivals” (186) from an earlier period of the language. In our separate article we explain how Hurvitz's ideas about “replacement” and “coexistence” actually complicate and even controvert his conception of language periodization (states and transitions), and we also comment more generally on the knotty issue of language “continuity” in historical linguistic theory and method.[32]

5.7.5.5. Notion of “characteristic” language. While Hurvitz recognizes that “[s]ome features may be idiosyncratic, not characteristic of LBH in general” (8), more often than not he talks about words or expressions that are “characteristic” or “indicative” or “typical” of LBH (3, 5, 8, 9, 10, 37, 44, 59, 64, 96, 113, 128, 143, 151, 178, 190, 194, 215, 220, 222). What, however, does “characteristic” mean? How might something that is “characteristic” be characterized empirically? And how does the idea of “characteristic” language relate to the preceding ideas of “variation,” “concentration,” “replacement,” and “coexistence”? The Lexicon does not incorporate any statistical analysis, and consequently it does not allow us to get a good grip on what is “characteristic” about the lexicon of LBH. We shed some light on this issue in our separate article.[33]

6. Discussion of specific examples

6.1. Preliminary remarks. At this point we would like to change tactics and look briefly at some individual linguistic variables in the Bible as a whole. This exercise will help us to appreciate better the way in which Hurvitz's Lexicon deals with individual linguistic items. It would be interesting and enlightening to investigate each of Hurvitz's variables in a separate study, but here we decided to look at “Damascus,” “after that,” “end,” and the seven Babylonian months as a group. It is important to underline that these four case studies illustrate Hurvitz's method as applied throughout his eighty entries.

6.2. “Damascus” (94–7). The spelling of “Damascus” as דַּרְמֶשֶׂק rather than דַּמֶּשֶׂק, the former with regressive dissimilation of the geminate consonants mem-mem, is attested six times in BH in the book of Chronicles: 1 Chr 18:5, 6; 2 Chr 16:2; 24:23; 28:5, 23. However, a glance at complete data for BH (MT) and QH enables us to get a firmer grasp on the history of spelling of “Damascus”[34]:

Hurvitz's main contention is that “the continued late usage of the Classical דַּמֶּשֶׂק in no way impugns the exclusively late linguistic status of דַּרְמֶשֶׂק. The decisive issue is not when a given Classical element ceased to be employed, but when a linguistic innovation became available as a substitute for its Classical counterpart” (96). In contrast, our main contention is that even if we grant that דַּרְמֶשֶׂק may be a late development (relatively or absolutely), it is not “characteristic” of LBH or QH and it does not serve Hurvitz's linguistic dating objective, since other late writers, such as those of Zechariah, Song of Songs, CD, and 4Q266, among many others,[35] do not make use of late דַּרְמֶשֶׂק, but only early דַּמֶּשֶׂק. Stated differently, even if the occurrence of late דַּרְמֶשֶׂק proved lateness, the occurrence of early דַּמֶּשֶׂק would not prove earliness.

6.3. “After that” (34–5). The phrases אַחַר זֶה and אַחֲרֵי (כָּל) זֹאת (“after that, following this”) appear five times in BH: Job 42:16; Ezra 9:10; 2 Chr 21:18; 32:9; 35:20. There are several similar phrases in the Aramaic of Daniel (Dan 2:29, 45; 7:6, 7). According to Hurvitz this is one of “many changes that took place between one phase and the next” (35), that is, between CBH and LBH. Talk about “phases,” however, should cause us to look at complete data for the items in question. The following figure gives the late variants and their early alternatives, וַיְהִי) אַחֲרֵ(י) כֵן) and אַחֲרֵ(י) הַדְּבָרִים הָאֵלֶּה, in BH (MT) and QH [36]:

Setting aside the Prose Tale of Job,[37] the late variants in Ezra and Chronicles (cf. references above) are accompanied also by the early variants (Ezra 7:1; 1 Chr 18:1; 19:1 [both synoptic]; 2 Chr 32:1; 33:14 [both non-synoptic]), and some late writings have only the early alternatives (e.g., אַחַר הַדְּבָרִים הָאֵלֶּה in Esth 2:1; 3:1). In other words, it is important that we document and explain differences “between one phase and the next” (35) and between writings in a single (i.e., late) phase of the language. Such differences should serve as a caution against the idea of a linguistic “phase” or “period,” as described by Hurvitz in this entry and elsewhere.

6.4. “End” (188–90). The noun סוֹף is used five times in BH: Joel 2:20; Qoh 3:11; 7:2; 12:13; 2 Chr 20:16. In addition, there are five occurrences in BA (Dan 4:8, 19; 6:27; 7:26, 28), and the cognate verb appears seven times in the Bible (BH: Isa 66:17; Jer 8:13; Zeph 1:2; Ps 73:19; Esth 9:28; BA: Dan 2:44; 4:30). Hurvitz argues that סוֹף is “a late word in BH” (189), “an element especially characteristic of the Second Temple linguistic milieu” (190), “among the distinguishing markers of LBH, late extra-Biblical Hebrew, and late Aramaic” (190). It is true, with the possible exception of Joel 2:20,[38] that the noun is attested, according to convention, in late-dated writings only. However, when Hurvitz remarks, “the emergence of a neologism does not necessarily stop the use of the older synonym” (189), we should pause to look at all the relevant data. The late noun סוֹף and its early alternatives (“אחרית ,קץ ,קצה”) have the following distribution in BH (MT) and QH:[39]

The earlier synonyms of סוֹף predominate and are “characteristic” of all of BH and QH, both early and late writings. Furthermore, some late writings like Daniel, Nehemiah, CD, and 1QS use the early words exclusively and even extensively. The distribution in Daniel is surprising. The Aramaic of Daniel has both the noun and verb of the root סוף (cf. references above),[40] but LBH Daniel was not influenced by this root and looks like CBH writings in the use of the early alternatives. In conclusion, סוֹף appears several times in two (Qoheleth, Chronicles) or three (Joel) late books, but it is not a “distinctive” marker of late writings since typically they do not have this noun at all, but rather instead the early alternative(s).

6.5. Babylonian months (28–30, 40–1, 120–1, 140–1, 182–4, 191–2, 226–7). Seven of eighty entries (almost 10%) in Hurvitz's Lexicon are Babylonian month names, for example, אֲדָר (“Adar”), a name of the 12th month. Hurvitz's main point, repeated verbatim in each entry and even with the same example, is: “Only in the LBH corpus do we find both (the older) numerical system and (the more recent) Mesopotamian names side by side in one and the same verse, e.g., Esth 3:7 לְחֹדֶשׁ שְׁנֵים עָשָׂר הוּא חֹדֶשׁ אֲדָר” (see on p. 28, 40, 120, 140, 182, 191, 226). Furthermore, he says: “[T]he numeral month names were in use in Hebrew throughout the first millennium b.c.e. They persist also in the late biblical books…. Only toward the end of the Second Temple period did the Babylonian names replace the numeral system” (29; emphasis added). The following figure gives complete data for BH (MT) and QH:[41]

There are some potentially significant issues for understanding usage and development in Biblical and Postbiblical Hebrew which we have to omit discussing in detail in the present context. For example, in Zechariah and Esther, without exception, the Babylonian month names follow and explain the numerical month names (cf. Zech 1:7; 7:1; Esth 2:16; 3:7, 13; 8:9, 12; 9:1; after the explanation is introduced in Esth 9:1 it is unnecessary to repeat it in vv. 15, 17, 19, 21).[42] Or, for instance, Nehemiah uses the first, sixth, and ninth Babylonian month names (נִיסָן in 2:1; אֱלוּל in 6:15; כִּסְלֵו in 1:1), but instead of the seventh Babylonian month name (תִּשְׁרִי), which does not appear anywhere in BH or QH,[43] Nehemiah has הַחֹדֶשׁ הַשְּׁבִיעִי (“the seventh month”; 7:72; 8:2, 14). These issues and related ones merit further study. But what we want to highlight in this context is that Hurvitz's claim about “replacement,” that “Babylonian names replace[d] the numeral system” toward the end of the Second Temple period (29), is an overstatement and, based on the surviving data, has almost no relevance to any of the writings thought to have been written in the Second Temple period![44] Note the scarceness of (late) Babylonian month names in the figure above. Note also that they are absent in transitional and late writings such as Jeremiah and Ezekiel, Haggai (vs. Zechariah), Daniel, Ezra, and Chronicles (vs. Esther and Nehemiah), and most QH writings. In contrast, the distribution of the (early) Canaanite month names is not what we would expect if there was a linear development from the Canaanite to the numerical to the Babylonian month names. Compare the use of the Canaanite month names in Exodus, Deuteronomy, and Kings, in contrast to the numerical month names in Genesis, Leviticus, Numbers, Joshua, and Samuel. It is also surprising, given Hurvitz's early dating of P, that this source (or redaction) has only the numerical month names. In short, Hurvitz's treatment of the Babylonian month names fails to give us a clear understanding of “the lexicon of Late Biblical Hebrew.”

6.6. Implications. In our judgment it is quite possible that each of the foregoing LBH lexemes could be relatively later in origin than its corresponding CBH alternative(s). So, for example, דַּרְמֶשֶׂק is a relatively later equivalent of דַּמֶּשֶׂק, as we have argued elsewhere;[45] or נִיסָן is a relatively later equivalent of הַחֹדֶשׁ הָרִאשׁוֹן and אָבִיב. Furthermore, in some cases, it may be persuasively argued that the embracing of a particular lexeme by biblical writers occurred rather late in an absolute sense, for example, sometime during or after the exile. That seems to be a reasonable explanation for the occurrence of Babylonian month names in some biblical books. Consequently, we would agree with Hurvitz that many of the LBH items in the Lexicon could be later features of BH than their CBH alternatives, relatively or absolutely. In that sense the title “Lexicon of Late BH Features” is a valid one for his book. However, the title “Lexicon of Late Biblical Hebrew,” the actual title of the book, is problematic because it gives a false impression of the language of the corpus of late writings as a whole. The reason for this, as we demonstrate in our separate article,[46] is that what is generally called late language is a very rare and idiosyncratic part of the language of the corpus of late books, and those books show very different tendencies in their use (including non-use) of late variants. In addition, the actual distribution of early and late variants in BH, which rarely conforms to Hurvitz's principle of linguistic contrast between early/CBH and late/LBH writings, undercuts the efficacy of language for dating, which is the ultimate purpose of the Lexicon (§3). Returning to a specific example, and repeating what we said above, even if the occurrence of דַּרְמֶשֶׂק proved lateness, the occurrence of דַּמֶּשֶׂק would not prove earliness (cf. §6.2).

6.7. Are the late features really late? Despite what we said in the previous section, often it is uncertain that the features discussed in the Lexicon are really late in origin. Sometimes (MT) biblical distribution, early extra-biblical attestation, or both, might raise doubts about the absolute lateness of a late variant. Examples are:

• Biblical (MT) distribution: p. 36–9, אֵין + infinitive, perhaps 2 Sam 21:4; p. 42–4, אֶרֶץ יִשְׂרָאֵל, see 1 Sam 13:19; 2 Kgs 5:2, 4; 6:23; p. 45–7, בהל, see Ps 48:6; Prov 20:21 (Qere); 28:22; perhaps Zeph 1:18; p. 88–91, דָּוִיד, see 1 Kgs 3:14; 11:4, 36; Amos 6:5; 9:11; p. 94–7, דּוּמֶּשֶׂק, perhaps a remnant of דַּרְמֶשֶׂק, see 2 Kgs 16:10; p. 98–100, דרש + God's law, Isa 34:16; p. 101–2, דָּת, perhaps Deut 33:2; p. 127–9, יְרוּשָׁלַיִם, cf. Jer 26:18; p. 179–81, נדב hithpael, Judg 5:2, 9; p. 185–7, נשׂא + אִשָּׁה, see Judg 21:23; p. 188–90, סוֹף, Joel 2:20; cf. verb סוף; Ps 73:19; Jer 8:13; Zeph 1:2, 3; perhaps Amos 3:15; p. 196–8, …עַד לְ, see 1 Kgs 18:29; p. 234–6, שַׁלִּיט, Gen 42:6.
• Early extra-biblical attestation: p. 25–7, אִגֶּרֶת, Aramaic; p. 48–51, בּוּץ, Phoenician; p. 112–4, זְמָן, perhaps Akkadian;[47] p. 115–6, זַן, perhaps Phoenician and Ugaritic;[48] p. 145–6, כִּרְצוֹן, cf. the verb רצה in Hebrew;[49] p. 147–9, כְּתָב, perhaps Hebrew;[50] p. 202–4, פִּתְגָם, cf. Homeric Greek verb;[51] p. 208–9, צָפִיר, perhaps Phoenician;[52] p. 217–20, קום piel, Aramaic.
• Both biblical distribution and early extra-biblical attestation: p. 165–70, מַלְכוּת, Num 24:7; 1 Sam 20:31; 1 Kgs 2:12;[53] Aramaic and perhaps Hebrew;[54] p. 173–5, מַעֲרָב, Ps 75:7; Aramaic and Ugaritic; קבל piel (213–6), Prov 19:20; perhaps Amarna Canaanite.[55]

Besides these points, because there is an extreme scarcity of early Hebrew documentary sources of known dates and places, and because much of what is deemed to be late based on (MT) biblical distribution can be called into question on the basis of, first, literary-linguistic circularity (i.e., book A is late, therefore feature Z in book A is late),[56] and second, the probability of lexical variation and change in the long and complex process of writing, editing, and copying the biblical writings,[57] absolute certainty about the lateness of this or that variant will remain out of reach.[58]

7. Conclusion

Hurvitz's Lexicon conveniently gathers in one place many of his previous studies of specific late BH lexical features, and those previous studies are supplemented by other helpful examples of usage in QH and RH. His book is a particularly valuable asset within the descriptive realm of the Lexicon. This said, since Hurvitz's ultimate objective is prescriptive, in the sense that he intends for his Lexicon to reinforce the dating, relatively or absolutely, of biblical writings on a linguistic basis, the work should be read and used cautiously, and also with awareness that the quirky field of linguistic dating is gradually being eclipsed among linguists in general by the more robust methods of mainstream historical linguistics and historical sociolinguistics, including corpus linguistics and variationist analysis.

Rezetko Robert, Radboud Universiteit Nijmegen & University of Sydney

Naaijer Martijn, Vrije Universiteit Amsterdam