Late Biblical Hebrew and The Qumran Pesher Habakkuk1
Ian Young
University of Sydney


The most widely held scholarly view argues that Early Biblical Hebrew (EBH) developed into Late Biblical Hebrew (LBH) during the sixth-fifth centuries BCE. It is claimed that on this basis scholars are able to date the composition of biblical books by analysis of their language. In contrast, we argue that EBH and LBH represent not successive chronological periods, but rather co-existing styles of Hebrew. This is demonstrated by the language of the Qumran Pesher-commentary on the biblical book of Habakkuk. Despite dating to the first century BCE and thus long after the period when LBH is said to dominate, Pesher Habakkuk is in EBH. It does not share the accumulation of LBH forms which characterises the core LBH books like Ezra, and exhibits a large number of cases where it prefers EBH linguistic forms against their LBH equivalents.

1. The Chronological Model

The most widely held view on the language of the Hebrew Bible considers that we can detect a clear chronological development from Early Biblical Hebrew (EBH) to Late Biblical Hebrew (LBH). EBH is the language of the pre-exilic or monarchic period, down to the fall of the Kingdom of Judah to the Babylonians in 586 BCE. The exile in the sixth century BCE marks a transitional period, the great watershed in the history of Biblical Hebrew (BH). After the return from exile in the late sixth century BCE, we have the era of LBH.2

Other scholars differ from this view mainly on the question of the date of the transition from EBH to LBH. Rather than the exile, these scholars see the decisive transition happening in the mid-fifth century BCE. This dating is possible since the core LBH books of Esther, Daniel, Ezra, Nehemiah and Chronicles are considered by them to date to after the mid-fifth century BCE.3 In contrast, the view that the exile was the decisive event is founded on the observation that among the LBH-related works, that is, works whose language causes them to be related by these scholars to the core books, are not only some psalms, the prose tale of Job, the Song of Songs, Lamentations, and Qoheleth, but also the book of the sixth century prophet Ezekiel.4

Both views, however, agree that the main explanation for the linguistic differences between, say, Genesis and Ezra, is chronological development. EBH developed into LBH. Furthermore, this development is generally considered to have continued from LBH to the language of the Dead Sea or Qumran Scrolls, Qumran Hebrew (QH), and typically is considered to have terminated in the emergence of Mishnaic or Tannaitic Hebrew (MH). The fact that a BH form can be found in QH or MH is commonly cited as evidence for the conclusion that it is therefore a late linguistic item.5

Let us summarise some key presuppositions of the chronological approach to BH. EBH developed into LBH. Biblical texts can be dated on linguistic grounds because LBH was not written early, nor did EBH continue to be written after the transition to LBH, whenever that occurred. Since QH is written long after the transition, it therefore must be even further developed along the road to lateness, presumably thus displaying clear links with LBH. LBH linguistic features were unavoidable by late writers.

In contrast to the chronological approach, a new approach has been emerging over the past few years.6 Briefly, this approach argues that EBH and LBH are styles which co-existed for much, if not all, of the biblical period. There is no sharp contrast between EBH and LBH. All EBH books contain LBH linguistic features, just not the concentration found in the core LBH books. Rather than two chronological eras with a transition between them, we see two basic authorial/scribal approaches to language use—conservative and non-conservative. Conservative (EBH) authors/scribes mainly relied on a limited core of linguistic forms, while non-conservative (LBH) authors/scribes were more open to utilizing the variety of linguistic forms available to them. Between these two poles there was a continuum of openness to linguistic variety.

2. Pesher Habakkuk

We now turn to a detailed study of the language of the Qumran Pesher Habakkuk, 1QpHab (henceforth PHab).7 It is a work whose historical allusions make certain that it is later than the composition of any of the biblical books. If the chronological theory is correct, therefore, PHab should show the unmistakeable signs of LBH. If it does not, the chronological approach is severely challenged.

It would of course be logical to avoid the dichotomy proposed in the previous paragraph by recourse to the concept of “archaising”. It is entirely reasonable to suggest that some later authors could have mastered the style of older works well enough to successfully imitate their style. However, this move has been explicitly rejected by proponents of the chronological model of BH. Thus, recently W. M. Schniedewind has argued that “archaizing is quite transparent because later authors did not have the linguistic tools and training to replicate the classical language.… It seems that Qumran Hebrew represents an attempt in this direction, but evidently it was not possible for them”.8 The necessity of this rejection is not hard to understand in the context of linguistic dating of biblical texts. Once it is admitted that late authors could successfully imitate early language, then our ability to tell early from late compositions on the basis of language is negated. The dating of the biblical books on non-linguistic grounds is widely debated in modern biblical scholarship. If EBH could be produced in any chronological period, there is no basis for the theory of chronological development. Thus, whether the language of PHab is produced by imitation or not, if it is not LBH, the chronological theory is struck a heavy blow.9

1QpHab, the pesher-commentary on Habakkuk from cave 1 at Qumran, is generally held to refer to the Roman invasion of Judea in the first century BCE, and the sole manuscript copy is also dated to the first century BCE.10 Berrin defines “Pesher” as “a form of biblical interpretation peculiar to Qumran, in which biblical poetic/prophetic texts are applied to postbiblical historical/eschatological settings through various literary techniques in order to substantiate a theological conviction pertaining to divine reward and punishment”.11 Thus the ancient prophecy of Habakkuk is understood to refer to events of the author’s own day. The work is structured so that the biblical text of Habakkuk (chapters 1–2) is quoted one section at a time, followed by an interpretation introduced by a formula such as “its interpretation (pesher) is …”. Thus we have a first century BCE work commenting on an older work. Due to the mention of the Chaldeans (Hab. 1:6), the Book of Habakkuk, on which the commentary is based, is commonly dated to the late pre-exilic period, c.600 BCE.12 It is thus universally assumed to be an example of EBH.

3. Lbh Features In Pesher Habakkuk

3.1 Methodological Problems

According to the chronological approach to BH, we should expect PHab to exhibit clear features of LBH. At this very first point, however, we run into severe problems. What exactly are the features of LBH? The classic methodology of Avi Hurvitz for identifying LBH has three criteria. A fourth criterion is then used to decide whether a particular text is to be considered late on linguistic grounds.13

The first criterion is linguistic distribution: the linguistic feature in question must occur exclusively or predominantly in biblical books which are indisputably post-exilic in date, that is the core LBH books of Esther, Daniel, Ezra, Nehemiah and Chronicles, whose historical setting indicates without doubt a post-exilic date. Thus, the form מַלְכוּת‎ for “kingdom” occurs in each of the core LBH books, a total of 78 times out of 91 occurrences in the Hebrew Bible. A further 6 occurrences are in LBH-related psalms and Qoheleth.14

The second criterion is linguistic contrast: there must be expressions in earlier biblical books which express the same meaning as the linguistic item in question. This is intended to rule out of discussion linguistic forms that may appear in LBH sources simply because there was no opportunity to use them in EBH texts. Thus מלכות‎ can be considered in linguistic opposition to other BH words for “kingdom” such as ממלכה‎.

The third criterion is extra-biblical attestation: the linguistic form in question must appear in post-exilic sources, whether Hebrew or Aramaic, from outside the Hebrew Bible. This is intended to demonstrate that the form was indeed current in the post-exilic period. Thus מלכות‎ is widely used in later Aramaic dialects and Tannaitic literature and is also found at Qumran.15

Fourth and finally, there is the criterion of accumulation: if a particular biblical text is to be judged late on linguistic grounds it must exhibit a clustering of late linguistic items identified using the above three criteria.

The four criteria are carefully thought out in order to guard against common mistakes made by earlier scholars trying to date texts by their language. Hurvitz’s methodology remains a useful way to describe linguistic relationships. Nevertheless, we argue that the nature of the evidence means that even these criteria cannot be used successfully to demonstrate the lateness of any text.16

The second criterion, linguistic opposition, is extremely useful, since there is indeed a variety of linguistic forms in the Hebrew Bible which appear to fulfil identical linguistic roles.17 This criterion is misunderstood, however, if it is taken to mean that LBH books completely replace EBH linguistic forms with LBH ones. This can happen, as in Esther’s total preference for מלכות‎ as its word for “kingdom”.18 However, the more common situation is where both EBH and LBH forms are used in the same book. Thus Chronicles uses מלכות‎ 28 times, and ממלכה‎ 22 times. LBH is generally not a replacement for EBH, but rather a supplementation of it. LBH is EBH plus extra linguistic choices.19

The usefulness of the third criterion, external attestation, for dating is virtually negated by the nature of the external evidence. Since the vast majority of our evidence for Hebrew and Aramaic dates to the post-exilic period or later, it is almost inevitable that BH forms, whether “early” or “late”, will be attested somewhere in a “late” non-biblical source. To draw chronological conclusions from this data would furthermore overlook the fact that these late sources are usually considered to represent linguistic forms available long before their attestation in writing. Thus, it is a widely held consensus that MH is a Hebrew dialect, independent of BH, which existed long before its full literary attestation in the Mishnah.20 We note, for example, linguistic forms attested in Hebrew inscriptions from the monarchic period and in MH, but never in the intervening literature.21 In regard to Aramaic, it is an especially weak argument from silence to claim that if a form is unattested in our extremely limited Old Aramaic sources, it therefore did not exist in that period. Thus Hurvitz has pointed out “we must always bear in mind that although the Elephantine papyri were written down in the fifth century B.C.E., the language employed in these texts was not created suddenly in the Persian period.… It is, therefore, perfectly clear that Elephantine Aramaic on the one hand and Biblical Hebrew on the other, even when exhibiting similar (or identical) linguistic usages, could have drawn, independently and at different times, on a common third source, earlier than both”.22 It is easy enough to find “late” parallels for peculiarities in EBH as well, which demonstrates the limited applicability of this criterion to chronology. Thus, the verb root חדד‎ “to be sharp, quick” is found only in the EBH Hab. 1:8, and the LBH-related Ezek. 21:21. The root is attested in Tannaitic Hebrew and Aramaic, as well as other “late” Aramaic.23 Yet, because it appears in Habakkuk, the external evidence is unlikely to be taken as helping to demonstrate that the form is “late” within Hebrew.

The criterion of external attestation can be especially misleading if attention is paid to only part of the external evidence in a chronological argument. Thus, within BH the word שש‎ generally is the word in EBH sources for “fine linen”, with בוץ‎ functioning in the same capacity for LBH. However, the fact that the word בוץ‎ is attested in the ninth century Phoenician Kilamuwa inscription cannot simply be dismissed as irrelevant for the question of the lateness of בוץ‎ within Hebrew if the criterion of external attestation is to be granted any value at all, which is doubtful.24 As another example, note how the late Aramaic evidence for אגרת‎ “letter” is cited as evidence for the lateness of the word within Hebrew, yet no mention is made of the early, Neo-Assyrian period attestations of the same word in Aramaic.25 So too, the Old Aramaic occurrences of our example word מלכות‎ “kingdom” show that its occurrences in later Aramaic indicate nothing about the chronology of the word in Hebrew.26

Serious problems are also encountered in applying Hurvitz’s first and fourth criteria to dating. The fourth criterion, accumulation, is only in fact necessary due to the problems in applying the first criterion, distribution. The first criterion, we remind ourselves, states that a suspected LBH feature must occur exclusively or predominantly in core LBH books. The phrase “or predominantly” is necessary since in the majority of cases, LBH forms are also attested in EBH texts.27 Thus, while the majority of cases of מלכות‎ “kingdom” occur in core-LBH or LBH-related texts, there are still seven occurrences of the word in EBH texts.28 This phenomenon alone is enough to undermine or seriously modify the chronological approach to BH. Is מלכות‎ actually a “late” linguistic item after all? If so, its appearance in a text should indicate that therefore the text is to be dated late. And if EBH texts which use מלכות‎ were dated late, this means late texts need not exhibit an “accumulation” of LBH features. If against this is it argued that the LBH linguistic feature found in the EBH text is not actually “late” but was also available in an early period, then its value for dating texts “late” is negated. Despite the claims of the criterion of accumulation, there is no reason to assume that an early author could not produce a text with a clustering of “LBH” elements if they were available to him. Or perhaps the LBH elements in EBH texts are evidence of later textual alteration of the language of the BH books?29 Proponents of the chronological model have been very reticent about invoking this explanation. Once it is admitted that the language of the biblical texts has been changed in scribal transmission, the claim that the language of the current texts is evidence of the date of the original author is thrown into doubt.30

Another serious problem emerges in the common situation where a linguistic feature is claimed to be characteristic of LBH, yet is not in fact characteristic of all LBH texts. A good example is the claim that the appearance of unassimilated מן‎ “from” before a noun without definite article is a feature of LBH.31 In this case not only are there numerous examples in EBH texts, but we also have the problem that significant numbers of examples of unassimilated מן‎ are only found in Chronicles and, to a lesser extent, Daniel, among core LBH texts. Another core LBH text, Esther, never fails to assimilate מן‎.32 Is preference for unassimilated מן‎ a symptom of lateness or simply a stylistic choice only brought to prominence by scholars because it happens to appear in some “late” texts? Many similar patterns of distribution and preference for linguistic forms are overlooked by scholars because they have no obvious relationship with chronology. Thus, scholars have long noted that Deuteronomy has a strong preference for לבב‎ as “heart”, while Jeremiah strongly prefers לב‎.33 Jeremiah’s preference is shared by, among others, Genesis, Exodus, Numbers, Judges and Samuel, and hence is not obviously a sign of “lateness”. Among the LBH books, Chronicles and Daniel align with Deuteronomy in preferring לבב‎, whereas the other LBH and LBH-related books prefer לב‎. Because this data cannot be used for the chronological argument, it seldom features in discussions of linguistic variety in BH.34

Since, therefore, most LBH linguistic features are also found in EBH texts and/or are not characteristic even of all LBH texts, we can argue that the large majority, if not all, LBH features are not reliable indicators of “lateness” in a chronological sense.

Does this mean that we have reached the end of our quest for LBH in Pesher Habakkuk before it even began? We believe not, as long as we formulate the aims and methods of our investigation more modestly. The core LBH books do exhibit a preference for some linguistic forms against other segments of BH. It is still a worthwhile task to ask whether a text outside the core shares a significant number of these LBH features with the core LBH books. Thus, we believe that Hurvitz and Rooker are correct to point out that Ezekiel shares more LBH peculiarities than any other prophetic book.35 According to the predictions of the chronological approach, PHab should inevitably exhibit a clustering of LBH features. In regard to designating what linguistic features are LBH, we must follow a relaxed methodology since, as we have seen, a stringent methodology collapses. Any form that is genuinely characteristic of one or more LBH books, exhibiting a “distribution” in core LBH sources and a “linguistic opposition” to a form found in EBH texts, may be considered. Thus, on the strength of the significant number of examples in Chronicles and Daniel, unassimilated מן‎ is classed as LBH. Some LBH features are only apparent over a large number of forms. Thus, preference for זעק‎ over צעק‎ for “cry out” is considered a LBH feature.36 However, “preference” can really only be established with a significant number of forms. Thus the two cases of זעק‎ in the Book of Habakkuk (Hab. 1:2; 2:11) do not clearly indicate a LBH feature in this EBH book. So too the one occurrence of אני‎ “I” (Hab. 3:18) does not clearly indicate a LBH-like preference for אני‎ over אנכי‎.37 We return to the question of “preference” below, 3.2.7.

3.2. LBH Forms in Pesher Habakkuk

As stated above, the aim of this section is not to find chronologically late features of the language of PHab.38 Rather, we are specifically seeking characteristic features of the core LBH books. “LBH” here thus means “linking with the core LBH books” while “EBH” means “linking with the core EBH books”, without any necessary chronological significance.

This said, however, the chronological approach to BH works with the clear presupposition that after a certain point in time, the only sort of BH which writers could produce was LBH. Therefore, since it is by universal agreement a “late” work chronologically, PHab should be in LBH, or else the standard theory is challenged.

The object of our investigation is the pesher sections of PHab. The biblical text of Habakkuk quoted as the lemma is not included in this discussion, since it is presumed here that this was a given for the author and does not reflect his own language. The question of whether, or to what extent, the authors of the Pesharim rewrote the biblical texts they used remains unresolved.39 Nevertheless, we shall discuss the issue of whether the language of the EBH Book of Habakkuk may have influenced certain aspects of the language of the pesher.

3.2.1. פשרו אשר‎ “its interpretation is that …” 1:11, etc.

It is debatable whether this and the other formulaic introductions to the actual commentary are due to the free linguistic choice of the author. After all, one cannot write a pesher without using the word pesher! Nevertheless, we include them here as possible LBH forms used by the author of PHab.

The word פשר‎ is not attested in core LBH, but is found in the LBH-related Qoh. 8:1,40 as well as in the Aramaic sections of the core LBH book of Daniel. It is often considered cognate with the root פתר‎ in the EBH Genesis 40–41. פשר‎ is commonly considered derived from the Akkadian root pašāru, typically via Aramaic, although this is not a necessary assumption. Horgan suggested that a proto-Semitic root *pṯr would account for both the Akkadian form and the form פתר‎, which would then be an Aramaic version of the root with the Aramaic shift *ṯ>t.41 If so, it is interesting that it is the EBH text which displays the more obviously Aramaic form. Furthermore the representation of proto-Semitic ṯ with taw is only a regular feature in Persian period Aramaic i.e. in the post-exilic period.42 The criterion of external attestation might lead us to see פתר‎ in Genesis as the post-exilic form, and פשר‎ in Qoheleth, on the basis of its Akkadian attestation, as potentially pre-exilic.43 Horgan suggests further that the author of Genesis consciously avoided using פשר‎ because of its magical connotations.44 Finally, also in regard to external sources, note that it is the root פתר‎ which is favoured in Rabbinic sources.45 At Qumran, while פשר‎ is much more common (due to its use in the Pesharim), the root פתר‎ is also attested.46 In summary, פשר‎ is at best a weak link to LBH. It exhibits no distribution, being unattested in core LBH, and only once in LBH-related; is without a clear linguistic opposition to פתר‎ in Genesis; and is unlikely in any case to represent the free linguistic choice of the author. These factors together mean that we cannot designate this form as LBH, even under our loose definition of it.

The use of אשר‎ for “that” as in פשרו אשר‎ “its interpretation (is) that …”47 is considered a feature of LBH.48 In this the author of PHab can be said to have had more choice than in his use of פשר‎ since other formulations were available, such as פשרו על‎ “its interpretation concerns …” as in PHab 2:12.49 As is typical with LBH features, אשר‎ for “that” is also well attested in EBH sources. According to the work of R. Holmstedt, of 49 certain cases of what he designates as “complement clause introduced by ʾasher”, 19 appear in the core LBH books, and 11 in the LBH-related pre-exilic books of Ezekiel and Qoheleth, whereas 16 appear in core EBH sources, and three in EBH-related psalms.50 Thus, while we have a form that is at home in EBH, it may be considered particularly characteristic of some LBH sources, especially Esther (6 cases) and Nehemiah (8 cases).51 Hence, using the loose definition of LBH outlined above, we may count this as a LBH feature of PHab.

3.2.2. Preference for Hiphil Over Qal Stem

Although attested in all strata of BH, it is argued that LBH has a particular tendency to use the hiphil stem of certain roots with an equivalent sense to the qal.52

According to most readings of PHab 4:2 we find the hiphil form ילעיגו‎ “they mock”. Some scholars, however, have suggested reading the qal ילעוגו‎ here.53 BDB calls the use of the hiphil stem of לעג‎ “late”.54 In the MT it occurs five times, three in core LBH texts (Neh. 2:19; 3:33; and 2 Chron. 30:10), once in an EBH psalm (Ps. 22:8) and once in the Archaic Biblical Hebrew (ABH) poetry of Job (21:3).55 Of these it is only the vocalisation that distinguishes the hiphil from the apparently identical meaning qal in three cases (Neh. 2:19; 3:33; Ps. 22:8). Of the 12 occurrences in the qal it is only the vocalisation that distinguishes eight of them from the hiphil. This is not a very secure link with LBH. Nevertheless, we accept this as a LBH feature of PHab.

The same phenomenon of preference for hiphil over qal may occur also in PHab 9:11 where scholars have generally understood the phrase הרשיע על בחירו‎ (9:11–12) as “he [the Wicked Priest] acted wickedly against his [God’s] chosen”.56 In BH, the qal verb רשע‎ means “to be wicked, to act wickedly” in both EBH and LBH. In the hiphil, however, the meaning “to condemn as guilty” is typical of EBH texts, never appearing in core LBH or LBH-related texts. Other texts, predominantly LBH, use the hiphil in the sense “to act wickedly”.57

Before declaring this to be another case of the LBH tendency to prefer hiphil to qal, however, we should eliminate the possibility that the scroll is to be translated according to the other, EBH, meaning of the hiphil. This is actually not easy to demonstrate. The translation that the Wicked Priest was punished because “he condemned as guilty [God’s] chosen” fits the context well. The lemma on which this text is commenting deals with חמס‎, which often has connotations of “injustice”.58 Recent scholarship has emphasised that “there is a high level of inter-dependence between the lemma and pesher in the pesharim”.59 Thus the idea of the perversion of justice due to the false condemnation of the innocent is highly appropriate to this context in PHab.

The tendency to translate the phrase in PHab as “act wickedly against” is probably influenced by the fact that in this case the verb coordinates with the preposition על‎. The verb הרשיע‎ “condemn as guilty” in BH often appears with an object, and this is always indicated by the direct object, not any preposition. However, this said, the verb in the sense “act wickedly” never appears in BH with an object. The use of הרשיע‎ with על‎ is thus unique in BH. As we shall discuss below (3.2.4), the appearance of על‎ with this verb is a symptom of PHab’s strong predilection for the preposition על‎. Its appearance here, as opposed to the use of the normal direct object with the verb, cannot be taken as decisive evidence. PHab could have used הרשיע‎ “condemn as guilty” with על‎. The purpose may have been to strengthen the adversative sense of the verb. Evidence against the translation “condemn as guilty” here can be found, however, in column 10, line 5, where the verb הרשיע‎ is indisputably used in the sense “condemn”60 and the object of the condemnation is expressed not with על‎, but with the direct object in the form of a pronominal suffix (ירשיענו‎ “he will condemn him”). Thus, although the author of PHab might not have been consistent with his language use and, as said, may have chosen על‎ in 9:11–12 for extra emphasis, it is probably better to retain the translation “act wickedly” in 9:11 and thus see this as a second case of the LBH tendency to prefer hiphil over qal stems in some verbs.

3.2.3. Verb Suffixes

The radically reduced use of the object marker את‎ with pronominal suffixes is considered a mark of LBH.61 Thus, Daniel never uses את‎ plus suffix, or Polzin claims that non-synoptic Chronicles prefers verbal suffixes over את‎ plus suffix at a ratio of 10:1. This he contrasts with EBH sections from the Pentateuch and Samuel where he claims the ratio is 12:7, still in favour of verbal suffixes.62

PHab uses 21 verbal suffixes, with not a single case of את‎ plus suffix.63 This seems, therefore, to be a strong LBH feature in PHab. It is interesting to note, however, that the EBH Book of Habakkuk as it is fully preserved in the MT, exhibits 18 verbal suffixes, and as in PHab has no examples of את‎ plus suffix.64 The Pesher in this case shares this supposedly LBH feature with the EBH text upon which it is commenting, and is thus possibly influenced by the style of the lemma text.65

We may thus include this as a feature of LBH found in PHab, but we find that this LBH feature is also found in EBH texts. Habakkuk is in fact not the only EBH text with a radically reduced use of את‎ plus pronominal suffix. Nahum, likewise an EBH prophetic book with a pre-exilic setting, has 10 verb suffixes without any occurrences of את‎ plus suffix.66 It might be argued that PHab is itself close to the prophetic genre of these two works. Nevertheless, we may also point to EBH narratives sharing the same aversion to את‎ plus suffix. Thus the core EBH text, 1 Kings 2, has 12 verb suffixes with no use of את‎ plus suffix. The next chapter, 1 Kings 3, has another 4 suffixes with no את‎ plus suffix.67 In this long stretch of EBH text, longer than the whole of PHab, there are thus 16 verb suffixes and no cases of את‎ plus suffix. Polzin’s statistics therefore do not reflect the variegated reality of EBH. Also relevant to note is that the generally EBH book of Ruth has 15 examples of verbal object suffixes and no examples of את‎ plus suffix.68 Note finally that the ninth century Mesha inscription from Moab, which is cited in the literature as evidence for EBH,69 contains 11 or 12 verbal object suffixes and no case of את‎ plus suffix.70 Thus, as is typical, this LBH feature is well attested in EBH texts also.

3.2.4. Preference for על

It is argued that LBH shows a growing preference for the preposition עַל‎, in particular at the expense of אֶל‎.71 PHab certainly displays a strong preference for על‎, using it 40 times, whereas אל‎ only occurs twice.72 The figures for על‎ are inflated by 20 cases of formulae such as פשרו על‎ “its interpretation concerns”.73 Disregarding these we still have 20 uses of על‎ against just two of אל‎.

More than just the sheer volume of PHab’s usage of על‎ is the fact that על‎ is used a number of times in coordination with verbs which normally in BH are used in different collocations.74 Sometimes the use of a particular verb with על‎ is paralleled in core LBH texts. Thus, the verb לעג‎ (4:2) “to deride” is in the MT only used with על‎ in Neh. 3:33, whereas elsewhere it is used with the preposition ל‎ or sometimes ב‎. So too we have בזה‎ “despise” plus על‎ in 4:2, 5, only paralleled in Neh. 2:19. שחק‎ “laugh” plus על‎ in 4:6 represents a different case where, although על‎ is rare (ל‎ being usual), the MT parallels are not strongly LBH: the EBH Ps. 52:8; the ABH Job 30:1; and the LBH-related Lam. 1:7.75 As we have seen, of course, LBH features are not confined to LBH texts. עזר‎ “help” plus על‎ in 5:11 is paralleled in 1 Chron. 5:20 (niphal); 12:22; 2 Chron. 26:7, 13; although I could not find the sense “help against” in any EBH text using a verbal form of עזר‎, but with nouns from the same root note Deut. 33:7 (with מִ-‎) and (with בּ‎) Jdg. 5:23. רחם‎“pity” plus על‎ in 6:11–12 is only paralleled in the LBH-related Ps. 103:13 (twice).76 Other uses of על‎ are not possible to compare with BH usage, but one might suspect they are symptoms of the general preference for על‎. In this category we have יתר‎ “exceed” in 7:7,77 משך‎ (niphal) “extend” in 7:12,78 כפל‎ “double” in 7:15, and הרשיע‎ “act wickedly” in 9:11–12 (cf. above, 3.2.2). On firmer ground, finally, we have גמל‎ “recompense” plus על‎ in 12:3, which is paralleled in a number of BH texts including the LBH text 2 Chron. 20:11 and the LBH-related texts Ps. 103:10 and Ps. 119:17, but also including the EBH Ps. 13:6. From this evidence it can be seen that PHab has a strong preference for על‎, prominently so in cases where other prepositions are more usual in BH, including several cases where the use of על‎ is characteristic of LBH texts.

As with the preference for verbal suffixes, however, here too it is interesting to check the language of the EBH lemma text, the biblical book of Habakkuk. Here too, just like PHab, Habakkuk itself exhibits a strong preference for על‎. It uses על‎ 19 times, as against just 4 cases of אל‎.79 Furthermore, just as with PHab, the biblical book of Habakkuk displays a series of cases where על‎ is coordinated with verbs which in BH normally coordinate with other prepositions or the direct object. Note כסה‎ “cover” plus על‎ in Hab. 2:14 which is found in the core LBH texts Neh. 3:37 and 2 Chron. 5:8; the LBH-related Ezek. 24:7; 31:15 and the post-exilic Mal. 2:16, as well as the core EBH text Deut. 13:9 among its 13 occurrences. הביט‎ “look” plus על‎ in Hab. 2:15 is only found here in the MT Bible, and contrasts with the use of אל‎ and the direct object in Hab. 1:3, 13. סבב‎ “go around” plus על‎ in Hab. 2:16 is only found 5 times in the MT, אל‎, את‎, and ל‎ being much more common. Finally, בטח‎ “trust” plus על‎ in Hab. 2:18 is less common than the use of the preposition ב‎.

The LBH feature of preference for the preposition על‎ is thus clearly attested not only in the Habakkuk pesher, but is present also in the biblical book of Habakkuk.80 This again raises the possibility that the author of PHab was influenced to use this linguistic feature by its prominence in the text he was commenting on.81 A further motivation for avoiding the preposition אל‎ may be suggested. This is that the author has a strong preference for using a word for “God” also spelled אל‎. Perhaps he chose to use על‎ as frequently as he did in order to avoid graphical (and phonetic?) confusion with the divine name.82

3.2.5. Pluralisation

It is argued that it is a feature of LBH to prefer plural forms of words and phrases which normally appear in the singular in EBH.83

The expression כלי מלחמותם‎ “their weapons of war” in 6:4 has both elements pluralised, whereas the normal BH expression, attested some eleven times is כלי (ה) מלחמה‎, with the second element in the singular.84 Similarly the expression דרכי תועבות‎ “abominable ways” in 8:12–13 represents a double pluralisation of a construct chain, although I have not found a MT parallel. So too, finally, there is the unparalleled expression מעשי תועבות‎ “abominable acts” in 12:8, which represents a double pluralisation. Note, however, that double pluralisations such as מעשי ידים‎ are attested in EBH.85 It has long been realised that the doubly plural construct chain is hardly confined to LBH.86

It is striking to discover that, as with the other cases we discussed above, the suggested LBH feature of pluralisation of words normally singular, for which we found some possible evidence in PHab, is also a feature of the EBH book of Habakkuk. In Hab. 2:7 we have the plural form משסות‎ “booty”.87 The other five times this noun appears in BH it is singular. So too in Hab. 2:8, 17, we have the expression דמי אדם‎ “men’s blood”. “Blood” in the plural is less common than the singular, 72 times as opposed to 288.88 Moreover, in this specific idiom we have a linguistic contrast with דם האדם‎ in Gen. 9:6. Finally, also in Hab. 2:1 we have the word בהמות‎ “beasts” in the plural rather than say in a singular collective. The plural of beasts is again the minority form, occurring 14 times against 176 times singular.

We thus find that the LBH feature of pluralisation of words and expressions normally singular in BH is present not only in PHab but also in the EBH biblical book of Habakkuk. It is less easy in this case to argue direct influence from the language of Habakkuk to the language of the Pesher. This is especially the case since the PHab examples relate specifically to cases of double pluralisation in construct chains, whereas those in Habakkuk relate to other types of pluralisation.89

3.2.6. רזי‎ “the secrets of …”

Three times in column 7 of the Habakkuk Pesher we find the plural of the word רָז‎ “secret”. We hear of “the secrets of the words of his servants the prophets” (7:5); “the secrets of God” (7:8); and “the secrets of his [God’s] wisdom” (7:14). The word רז‎ is generally considered to have entered Hebrew (from Persian) via Aramaic. It is well attested in Biblical Aramaic, occurring nine times in the Aramaic sections of the LBH book of Daniel. Within BH, the word רז‎ occurs twice in Isa. 24:16 in the form רָזִי‎ “my secret”. That this is how the word was understood by the Masoretes is made even more likely by the unanimous testimony of the ancient versions. Symmachus, Theodotion, the Peshitta, the Vulgate, and the Targum all understand the word in this way.90 A number of modern scholars agree in seeing רז‎ “secret” here.91 Other scholars have argued that the word should be emended to read רְזִי‎ “my leanness” (AV) > “I pine away” (NRSV), on the basis of a perceived parallelism with the following אוֹי לִי‎ “woe to me”.92 The root רזה‎ in the niphal seems to mean “to dwindle, disappear” in Isa. 17:4, and there is a feminine adjective רָזָה‎ “thin, gaunt” in Num. 13:20 and Ezek. 34:20.93 However, the sense required here, “leanness”, is expressed by the form רָזוֹן‎ in Isa. 10:16; Mic. 6:10; Ps. 106:15.94

Whether רז‎ “secret” occurs in the MT because it was in the “original” text of Isa. 24:16 is disputed. Not only is there the possibility that רזי‎ represents an otherwise unknown word meaning “leanness”; there is also the problem of the absence of this phrase from the Septuagint.95 This absence may be interpreted as evidence that רזי‎ is a later addition to the text of Isaiah.96 Whatever we may decide on this question, the significance of this discussion for PHab is that while רז‎ is found in the current MT, it is found only once in BH, and while Isaiah 24–27, “the Isaiah Apocalypse” is often considered a “later” section in the Book of Isaiah97 it is not considered to represent LBH.98 The word רז‎ in PHab is thus not strictly a link with LBH. However, רז‎ is considered a Persian loanword,99 and Persian loanwords are considered a feature of LBH.100

We have noted that LBH features typically are also found in EBH texts, just not with the same frequency. The discussion of the appearance of רז‎ in the MT of Isa. 24:16 alerts us to the fact that this is the situation with Persian loanwords, which can also be suggested in a number of EBH texts such as Deuteronomy, Kings and Nahum.101 Nevertheless, in line with our loose definition of LBH features we can accept this word as LBH in PHab, albeit in itself representing only a weak and indirect link with LBH.

3.2.7. Accumulation?

We have thus identified a number of LBH features of the language of PHab. Several of them provide only weak and general links with LBH. The three strongest examples, verbal suffixes, preference for על‎, and pluralisation, are all paralleled in the EBH text of Habakkuk.

How do we proceed from here? Do we simply state that since some LBH features are found in PHab that therefore its language fits the late period of its composition? This would clearly be a wrong move, as the mere presence of LBH features cannot be a marker of LBH, since core EBH texts exhibit LBH linguistic features. This has been brought home to us forcefully by the fact that the most prominent LBH features of PHab are also shared by the EBH book of Habakkuk.

This dilemma is the reason scholars of LBH were forced to have recourse to the criterion of accumulation to attempt to use LBH features for dating texts. As mentioned above, since LBH features occur throughout the Bible, this criterion states that a text can only be LBH if it exhibits an “accumulation” of LBH features. However, nowhere to my knowledge has an attempt been made to specify how much of an accumulation is necessary for a text to be LBH, nor how such an accumulation should be measured.

In response to this problem I developed a simple test of accumulation. Plainly put, this counts how many different LBH features occur in a given stretch of text. Normally, this stretch of text will be of 500 words length,102 so that samples are comparable. Within this sample we count how many different LBH features there are. We do not count repetitions of the same feature. Once an author has demonstrated the possibility of using a particular LBH form, there is no reason it cannot be repeated as many times as opportunity presents itself. Thus, for example, the LBH order of substantive before numeral occurs seven times in Ezra 1:9–11 simply because it is a list.

In this exercise we follow the loose definition of LBH linguistic features outlined above. In regard to “preference for” categories other than על‎, and hiphil over qal, we decided to score this as a LBH feature if the feature in question occurs five times or more in the 500 word section with no examples of the EBH form or a ratio of 10–1 if the data so permitted. Thus both PHab and biblical Habakkuk show a preference for verb suffixes and hence register this as a LBH feature. On the contrary, the two examples of LBH זעק‎ in biblical Habakkuk do not qualify as a LBH feature.

Table 1: LBH Features in BH Texts (Descending Order)103


of LBH Features104

Ezra 1:1–11; 9:1–10:29


Daniel 1:1–20; 11:44–12:13


2 Chronicles 30:1–31:3


Nehemiah 1:1–2:17


Esther 5:1–6:13a


Arad Ostraca


1 Kings 22:6–34


1 Samuel 13:1–14:9


2 Samuel 6:1–20a; 7:1–12


2 Samuel 22:1–51

6114 (7.9115)

1 Kings 2:1–29


Psalm 18:1–51

6117 (7.6118)

Pesher Habakkuk 5:3–12:13


Habakkuk 1:1–3:4


Genesis 24:1–36 (J121)


Zechariah 1:1–3:1a


Exodus 6:2–12; 7:1–13; 9:8–12; 12:1–7b (P124)


Table 1 is very clear. While all the samples contain LBH linguistic forms, the core EBH and core LBH books are at different ends of the scale in terms of the amount of accumulation of these LBH features. Thus, while the highest EBH sample, 1 Kings 22, has 8 different LBH features, the lowest LBH sample, Esther 5–6, has 17, more than twice as many as 1 Kings 22, while the other LBH samples have yet higher numbers of LBH features.

Amidst the core EBH samples, we find our text, PHab, as well as the post-exilic book of Zechariah. The notable lack of LBH features in Zechariah 1–8 has been emphasised by Martin Ehrensvärd.126 Ehrensvärd mentions other examples of post-exilic EBH, and below we discuss other Second Temple period texts which demonstrate that the EBH style was fully at home in the post-exilic period.127 PHab also, despite its first century BCE date, clearly fits an EBH profile of low accumulation of LBH features, in contrast to the much higher accumulations of the core LBH books. We recall, further, that of the six LBH forms in the PHab sample above, one (אשר‎ for כי‎) is taken over under the influence of the Pesher genre, while another three (verb suffixes, preference for על‎, and pluralisation) might have been picked up under the influence of the EBH lemma text of Habakkuk. There is doubt, furthermore, about the LBH status of the form הרשיע‎ in our sample (3.2.2). This would leave but one LBH feature, the use of Persian words. However, we note that this specific Persian word (רז‎) is not attested in LBH. One could, therefore, make a case that, unlike other EBH texts, PHab contains no LBH features used due to the free choice of the author. Every one of PHab’s LBH linguistic features is also attested in EBH sources. However, even accepting all the LBH features discussed above, the linguistic profile of PHab still aligns with EBH against LBH. The predictions of the chronological approach have thus been shown to be seriously off the mark. As discussed above, the defining characteristic of LBH is a concentration or accumulation of LBH linguistic features. PHab does not exhibit this defining characteristic and hence is not in LBH. We thus find it was indeed possible128 to write a biblical style Hebrew in the post-exilic period which was not LBH. This finding is another severe blow to the chronological approach to BH and its attempt to date biblical texts on the basis of their language.

4. Ebh Features In Pesher Habakkuk129

The link between PHab and EBH is strengthened even further by numerous cases where the language of PHab exhibits links with EBH against LBH.

4.1. EBH Lexical Features

4.1.1. מאס‎ “reject” 1:11, 5:11130

The verb מאס‎ “reject” is found 76 times in the Hebrew Bible, yet it is never found in core LBH books.131 The LBH book of Chronicles instead uses the hiphil of זנח‎ for “reject” (1 Chron. 28:9; 2 Chron. 11:14; 29:19).132 Rejecting the Torah is found in EBH books using מאס‎ (Isa. 5:24; Jer. 6:19; Amos 4:2), yet similar LBH contexts of disobedience (e.g. Nehemiah 9) avoid using the word, preferring instead other expressions, such as “they cast (וישלכו‎) your Torah behind their backs” (Neh. 9:26).

4.1.2. בגד‎ “act treacherously” 2:1; 3:5; 8:10

None of the 49 verbal or five nominal or adjectival usages of the root בגד‎ are found in core LBH texts, and of LBH-related we find only Ps. 119:158 and Lam. 1:2. LBH instead prefers the root מעל‎ for “act treacherously”, also found in EBH (and PHab 1:6), which occurs in core LBH as a verb 16 times and as a noun 10 times.

4.1.3. עריץ‎ “violent”133 2:6

The word עריץ‎ “violent” is found 20 times in BH, never in core LBH, only in the pre-exilic/exilic LBH-related Ezekiel. Similarly, the cognate134 verb ערץ‎ “be terrified/ terrify” is not attested in LBH or LBH-related. Due to the variability in scholarly understandings of the semantic range of עריץ‎ it is hard to suggest a certain LBH equivalent, but for the meaning “to be terrified” we note the specifically LBH use of the niphal of בעת‎, found in BH only in Chronicles, Daniel and Esther.

4.1.4. ממשלה‎ “kingdom, dominion” 2:13

Although the word ממשלה‎ is found in both EBH and LBH sources, it is noteworthy that PHab did not choose a specifically LBH form like מלכות‎ or a derivative of שלט‎.135

4.1.5. ממרחק‎ “from afar” 3:10

The noun מרחק‎ is found 17 times in BH, never in core LBH, and in LBH-related only once in the pre-exilic/exilic LBH-related Ezekiel (23:40). In contrast to ממרחק‎ “from afar”, LBH prefers other expressions using the root: מרחוק‎ (Neh. 12:43), למרחוק‎ (Ezra 3:13; 1 Chron. 17:17; 2 Chron. 26:15) and מארץ רחוקה‎ (2 Chron. 6:32, 36).

4.1.6. משל‎ “rule” 4:5, 10, 12; 8:9

Although common to both EBH and LBH it is noteworthy that PHab avoids specifically LBH alternatives such as the root שלט‎.136

4.1.7. תפשׂ‎ “capture” 4:7

The verb תפשׂ‎ is used of capturing cities only in EBH (Deut. 20:19; Josh. 8:8; 2 Kgs 14:7; 16:9; 18:13//Isa. 36:1; cf. Jer. 40:10; 49:16). For LBH note the parallel verses 2 Kgs 18:13//Isa. 36:1//2 Chron. 32:1 where the EBH books of Kings and Isaiah say of the fortified cities of Judah that Sennacherib captured them (ויתפשם‎). In contrast, Chronicles says that Sennacherib thought to break into them (לבקעם‎).137

4.1.8. הרס‎ “demolish” 4:8

Only one (1 Chron. 20:1138) of 43 occurrences of this verb is in a core LBH text. Rooker notes the EBH status of הרס‎ and gives an example where, in expressing the idea “to tear down an altar” the EBH 1 Kgs 19:10 uses הרס‎, whereas the LBH 2 Chron. 34:7 uses the piel of the root נתץ‎.139

4.1.9. עוון‎ “wickedness” 4:8

Core LBH prefers to pluralise this word. It occurs six times in the plural (masc: Dan. 9:13; Ezra 9:13; fem: Dan. 9:16; Ezra 9:6, 7; Neh. 9:2), but only three times in the singular (Dan. 9:24; Neh. 3:37; 1 Chron. 21:8). In contrast to this 2:1 ratio in favour of plural in core LBH, overall BH prefers the singular at a ratio of 4:1. PHab does not follow the LBH trend.

4.1.10. נדמו‎ “they kept silent” 5:10

The verb דמם‎ “be silent” occurs 30 times in BH, never in core LBH and in LBH-related only in pre-exilic/exilic Ezekiel (24:17) and Lamentations (2:10, 18; 3:28). An alternative is to analyse the root as דמה‎, which is similarly absent from core LBH.140 LBH uses other words for “be silent” such as אלם‎ (Dan. 10:15; cf. Ezek. 3:26; 24:27; 33:22).

4.1.11. תוכחת‎ “rebuke” 5:4, 10

The noun תוכחת‎ “rebuke” is never found in core LBH. LBH uses other roots with the meaning “admonish, rebuke”, such as זהר‎ (2 Chron. 19:10; cf. Qoh. 4:13; 12:12; 15 times in Ezekiel) and יסר‎ (2 Chron. 10:11, 14// 1 Kgs 12:11, 14).

4.1.12. מוראם‎ “their fear” 6:5

The noun מורא‎ “fear, object of fear” appears 12 times in BH, never in core LBH or in LBH-related works, although it does appear in the post-exilic EBH of Malachi141 (1:6; 2:5). LBH prefers other words for “fear”, including substantives like פחד‎.

4.1.13. לחריב‎ “to devastate” 6:8

The root חרב‎ in the hiphil “devastate” is found 13 times in BH, never in core LBH. LBH prefers other verbs for devastation and destruction such as אבד‎.

4.1.14. פריבטן‎ “fruit of the womb” 6:11–12

This expression is found 11 times in BH, never in LBH or LBH-related. Comparable lists in LBH simply leave this element out, e.g. Esth. 3:13: “young and old, women and children” cf. 2 Chron. 20:13; 31:18.

4.1.15. חקק‎ “decree” 7:13

The verb חקק‎ is used 19 times in BH, never in core LBH, and in LBH-related only in the pre-exilic/exilic LBH-related Ezekiel (4:1; 23:14). For the sense “to decree” see e.g. Isa. 10:1. LBH uses other words for enacting a decree such as הקים‎ and העמיד‎.

4.1.16. ערמתו‎ “his wisdom” 7:14

The noun ערמה‎ “craftiness, prudence” is only found in biblical Wisdom literature and in core EBH (Exod. 21:14; Josh. 9:4).142 LBH shares other, common words for “wisdom” such as חכמה‎.

4.1.17. קבץ‎ “gather” 8:11; 9:5

Although the verb קבץ‎ “to gather” is attested in both EBH and LBH, it is noteworthy that PHab avoids using the LBH synonym כנס‎.143

4.1.18. פעל‎ “do” 8:13; 12:8

The verb פעל‎ “to do” is found 56 times in BH, never in core LBH, and only once each in LBH-related psalms Ps. 119:3 and Ps. 125:5. Related nouns are used twice in Chronicles (1 Chron. 11:22//2 Sam. 23:20; 2 Chron. 15:7). Thus out of 111 occurrences of the root פעל‎, only four at most relate to LBH contexts. LBH instead just utilises the more common BH root עשה‎.

4.1.19 גויה‎ “body” 9:2

Although the word גויה‎ for “body” is found in both EBH and LBH, we note that PHab does not use the LBH synonym גופה‎.144

4.1.20. שלל‎ “plunder” 9:5, 6

PHab chooses the common BH noun שלל‎ for “plunder” rather than the LBH synonym בִּזָּה‎.

4.1.21. לענותו‎ “to humble him” 9:10

The use of the root ענה‎ in piel meaning “to humble” is attested 69 times in the MT Bible. However, it never appears in core LBH books and in LBH-related texts only twice each in Ezekiel and Psalm 119. LBH uses other terms for “to humble” such as the hiphil stem of כנע‎.

4.1.22. הרשיע‎ “condemn” 10:5

As discussed above in section 3.2.2, LBH uses the hiphil of רשע‎ only in the sense “to act wickedly”, never in the EBH sense “to condemn as guilty”. Note the parallel texts 1 Kgs 8:32// 2 Chron. 6:23. EBH Kings says that God will judge his servants by condemning the wicked (להרשיע רשע‎). The parallel in LBH Chronicles, however, says that God will judge his servants by repaying the wicked (להשיב לרשע‎). Also note LBH Dan. 1:10 which uses the piel of the Aramaic root חוב‎ for the sense “to make guilty”.145

4.1.23. עדה‎ “congregation” 10:10146

Hurvitz argues that within BH the use of the word עדה‎ rather than קהל‎ for “congregation” is a characteristic of EBH texts as opposed to LBH ones.147

4.1.24. גדפו‎ “they reviled” 10:13

The rare verb גדף‎ “to revile” is found six times in EBH texts and only once in a LBH-related text, the pre-exilic/exilic Ezekiel. Cognate nouns are also found in Ezekiel, as well as EBH Zephaniah and the exilic EBH Isaiah 40–55. LBH uses other words for “to revile, insult” such as חרף‎, found also in EBH, which PHab uses alongside גדף‎. Compare PHab “and they reviled (גדפו‎) and insulted (ויחרפו‎) the chosen ones of God” with “and he insulted (ויחרף‎) Israel” (1 Chron. 20:7 [//2 Sam. 21:21]; cf. Neh. 6:13; 2 Chron. 32:17).

4.1.25. זמם‎ “he planned” 12:6

The verb זמם‎ “to plan” occurs 13 times in the MT Bible, and the related nouns זמה‎ and מזמה‎ appear 29 and 19 times respectively, yet never in core LBH. The verb is found once in exilic LBH-related Lamentations, and the noun זמה‎ is common in the pre-exilic/exilic LBH-related Ezekiel. LBH uses the common BH word חשב‎ for “to plan”. In a negative context comparable to PHab see, for example, Esth. 9:24: “he plotted (חשב‎) against the Jews to destroy them”.

4.1.26. עובדי‎ “servants/worshippers of” 13:3

The use of the plural participle of עבד‎, rather than the cognate noun for “servants, worshippers” is restricted to EBH texts.148

4.2. EBH Morphological and Grammatical Features

4.2.1. מפיא‎ “from the mouth” 2:2

PHab always assimilates the nun of the preposition מן‎ “from” to a following word without a definite article (2:2; 7:11; 10:4; 11:12; cf. מן הארץ‎ in 13:4), against the LBH tendency to leave מן‎ separate before an anarthrous noun.149

4.2.2. בשמעם‎ “when they hear” 2:7

Polzin argues that there was a sharp decline in the use of the infinitive construct with beth or kaph in LBH, leading to its complete absence in Mishnaic Hebrew.150 He provides no guidance as to how to judge this decline overall, but it is worth noting the occurrences of infinitive construct plus beth in PHab 2:7; 7:12; and 10:16.

4.2.3. וקלסו‎ “and they mock” 4:3; וידבר‎ “and he spoke” 7:1

Some scholars have claimed a breakdown of the classical Hebrew verbal system in LBH, including the breakdown of the use of converted tenses. PHab, on the contrary, consistently uses converted verbs, in accordance with EBH practice.151

4.2.4. עם רב‎ “a large people/army” 4:3, 7

The core LBH books Ezra, Nehemiah and Chronicles, and the LBH-related Ezekiel construe עם‎ very commonly as plural.152 Even though adjectives like רב‎ are normally singular when referring to עם‎, plurals can occur.153 PHab does not exhibit the LBH tendency to construe collectives as plurals.

4.2.5. למו‎ “to them” 5:6

Although attested 55 times in the MT Bible, the preposition lamed with the archaic third person masculine plural suffix is never attested in core LBH, and in LBH-related only in the poetry of Psalm 119 and Lamentations. In contrast, the standard BH להם‎ occurs 100 times in core LBH books.

4.2.6. לאותותם‎ “to their standards” 6:4

Against the LBH tendency to place the long third person masculine plural suffix on the feminine plural ותיהם‎-, PHab follows EBH practice in using the shorter form ותם‎-. See also in this line of PHab מלחמותם‎ “their wars”.154

4.2.7. נשים וטף‎ “women and children” 6:11

PHab’s word order “women and children” is found 14 times in EBH sources,155 and never in LBH. In contrast, the reverse “children and women” is found eight times in BH, four times in core LBH, once in LBH-related, and three times in EBH.156 This is thus another case where PHab follows EBH practice against LBH.

4.2.8. מרדו ב‎ “they rebelled against” 8:11

Against PHab’s LBH-like preference for the preposition על‎ (above, 3.2.4), note that here it follows the common BH use of the preposition beth with the verb מרד‎ “to rebel” against the use of על‎ in Neh. 2:19 and 2 Chron. 13:6.

4.3. Preliminary Conclusions

In the previous section we showed that, contrary to expectations, PHab has no higher a concentration of LBH features than many EBH texts, and decidedly fewer than appear in core LBH texts, and hence cannot be classified as LBH. In this section we have seen that additionally, PHab has many cases where its language exhibits close links to EBH as opposed to LBH. Whereas thirty-four lexical and grammatical features of PHab align with EBH (section 4), there are only six links with LBH (section 3). Thus, given the choice of classifying PHab as either EBH or LBH, we must clearly classify the language of PHab as EBH.

4.4. Imitation?

PHab’s language thus aligns much more closely with EBH than LBH. We have already raised the issue of whether such language use is due to imitation of biblical works in EBH or due to a continuation of the EBH style (section 2, above). These two possibilities are in fact not mutually exclusive, since education in the ancient world focussed on mastering a standard curriculum of ancient texts.157 It is widely acknowledged that well before the time of the composition of PHab in the first century BCE, the Jewish educational curriculum was based on biblical texts158 and that the core texts were EBH texts such as the Pentateuch, Isaiah, the Twelve Prophets, and Psalms, with the Wisdom works Job and Proverbs.159 Education thus involved mastery and memorisation of core EBH books, with a corresponding mastery of their language.160 Thus to say that PHab represents a continuation of the EBH style is to acknowledge that the author, like his predecessors, mastered EBH style by mastering the language of earlier works written in EBH.

That PHab’s “imitation” of earlier linguistic models represents a broad mastery of the EBH style is evident on consideration of the distinctively EBH linguistic features we have just described. It is noteworthy that the great majority of them are not found in the biblical book of Habakkuk. Even those few that are found in biblical Habakkuk, commonly do not occur in the lemma of the section where PHab uses the same linguistic form. Thus, note that משל‎ (4.1.6; cf. Hab. 1:14), תוכחת‎ (4.1.11; cf. Hab. 2:1),פעל‎ (4.1.18; cf. Hab. 1:5 [and 3:2]) and למו‎ (4.2.5; cf. Hab. 2:7) occur in both PHab and biblical Habakkuk, but in different sections.161 In addition, תפש‎ (4.1.7; cf. Hab. 2:19) not only occurs in a different section, but also in a different sense.

Apart from the grammatical features of assimilation of מן‎ (4.2.1) and use of the waw-consecutive (4.2.3), which are found throughout both PHab and biblical Habakkuk, there are only two or three cases where the language of the pesher directly echoes an EBH feature of the lemma. The clearest case is קבץ‎ (4.1.17) which in PHab 8:11 is found in the pesher to Hab. 2:5 which uses the same root, albeit in a different conjugation (qal vs. niphal). Note, however, that קבץ‎ is also used in PHab 9:5 with no correlation in the lemma. PHab 8:3 cites the lemma with a form of the verb בגד‎ (4.1.2; יבגיד‎ or יבגוד‎ vs. MT בגד‎ participle) and the pesher in PHab 8:10 uses the same verb. The participle of בגד‎ is found in PHab 2:1, 3, 5, commenting on Hab. 1:5, which is not preserved in PHab. Some scholars reconstruct the lemma to include the word בגדים‎ also.162 Finally, the use of the noun שלל‎ in PHab 9:5 (4.1.20) is related to the lemma since the cognate verb is used twice there (Hab. 2:8).

The majority of the EBH lexical (18 out of 25; 4.1.1, 3–5, 8–10, 12–16, 19, 21–26) and grammatical (5 out of 8; 4.2.2, 4, 6–8) features of PHab are not found in the biblical book of Habakkuk. This would seem to indicate that if the EBH language of PHab is produced by “imitation”, it is due to general knowledge of the EBH style, not the direct influence of the lemma text. If the author of PHab was struggling to master an alien style of language use, we might have expected him to rely more on the language of the text upon which he was commenting. Especially instructive in this regard is PHab 3:6–14. In PHab 3:6–9 the biblical lemma is quoted (Hab. 1:8–9a) which in its description of the Chaldeans uses the common BH expression, used in LBH as well as EBH, מרחוק‎ “from afar” (PHab 3:7).163 In contrast to this, PHab (3:10) uses the exclusively EBH form ממרחק‎ (PHab 3:10; 4.1.5). In other words, even though provided with a common and perfectly legitimate BH linguistic form in the lemma, PHab chose instead to use a more specifically EBH form.

The evidence suggests, therefore, that PHab’s EBH language was produced due to general mastery of the EBH style. This was emphasised long ago by Elliger.164 We find no evidence that the author was struggling to write EBH. On the contrary, the language of the pesher sections of PHab shows a certain independence of the language of the lemma.165 PHab could write EBH successfully because the author was trained to do so by mastery of earlier, classical texts in EBH. In this regard, he was probably no different to earlier EBH authors. EBH was a style that continued to be learned and used throughout the Second Temple period. PHab’s language is thus produced by “imitation” only in the broadest possible sense of that term.

5. Non-MT Language Features in Pesher Habakkuk

Although it is beyond the strict scope of this paper, it is worth briefly pointing out some of the linguistic features of PHab which are not normal in either EBH or LBH, some of which, in fact, are not attested in BH in its MT form.166 Many of these are orthographic, such as the use of the digraph יא‎- in פיא‎ “mouth” (2:2; contrast פי‎ in 2:7) and כיא‎ “because” (2:3 etc; contrast כי‎ in 3:2), or the digraph וא‎- in יאמינוא‎ “they will (not) believe” (2:6; contrast e.g. 2:14)167 and gentilics of the pattern הכתיאים‎ “the Kittim” (2:11 etc). Other cases involve vocalisations of words contrary to the Tiberian tradition in the current MT such as ישופטנו‎ “he will judge him” (12:5). These peculiarities can be paralleled in some Qumran copies of EBH books, especially those in the so-called “Qumran practice”.168

PHab has, however, some non-MT linguistic forms not involving orthography and pronunciation. Several of these are known as “Qumran” forms, such as הואה‎ “he” (1:9; contrast הוא‎ in e.g. 1:13),169 the predominant use of אל‎ for “God” (1:11 etc), lexical peculiarities such as קץ‎ as “age, period of time” (5:7; 7:2, 7), גמר‎ “consummation” (7:2), תכונה‎ as “fixed or right time” (7:13) and יחד‎ “community” (12:4). In addition, note grammatical peculiarities such as the use of the preposition lamed, not beth, in the expression לאחרית הימים‎ “in the last days” (2:5–6).

In addition, there are various peculiarities that turn up as rare forms in BH, such as the ubiquitous dropping of the he of the hiphil infinitive (3:1;170 4:13; 6:8; 8:12; 10:10, 11; 11:8, 15; contrast 3:5; 7:8; cf. niphal in 7:12),171 the śin/samekh interchange in מאשו‎ “they rejected” (1:11; contrast 5:11)172 and the “Qumran” form להמה‎ “for them” (12:14) also found in Jer. 14:16 (cf. בהמה‎ in the MT of Hab. 1:16!). This listing is not exhaustive, but gives the main features of the evidence.

6. Conclusions

PHab does not exhibit a concentration of LBH linguistic features comparable to or exceeding the core LBH books of the MT Bible. In fact, the number of LBH features is no higher than in core EBH texts. In addition, PHab exhibits a high number of linguistic links with EBH in opposition to LBH. On this basis, we can say that PHab’s language aligns more closely with EBH than LBH. This result is contrary to the explicit expectations of the chronological theory of BH. According to that model the amount of LBH should increase over time, from virtually none in the pre-exilic period, through an exilic transition, to a post-exilic period characterised by LBH, which should be completely dominant at the time of the composition of PHab in the first century BCE. This model does not fit the evidence.

The primary characteristic of EBH books that marks them apart from the core LBH books is a relatively low accumulation of LBH linguistic features. Quite apart from our case of PHab, other post-exilic works, such as Zechariah 1–8 also exhibit low, EBH accumulations of LBH features. The second century BCE book of Ben Sira, like PHab, also has a typically EBH low accumulation of LBH features.173 In fact, in Young, Rezetko and Ehrensvärd’s investigations, no Qumran document yet studied exhibits an accumulation of LBH forms comparable to the core LBH works.174 In addition to PHab and Ben Sira, the Community Rule and the War Scroll175 all have less than or equal the number of LBH features found in the Arad Ostraca (see Table 1), extra-biblical sources from the pre-exilic period. In other words, some sources from the end of the Second Temple period have less LBH elements than the Arad Ostraca from the end of the First Temple period. Chronology is not the explanation for these accumulations of LBH features, but rather that some authors have a stylistic preference for them.

Instead of a model whereby LBH is considered a linear development of EBH, which is incompatible with the evidence, a better model sees LBH merely as one style of Hebrew in the Second Temple period, alongside EBH.176 The post-exilic authors and scribes who composed and transmitted works in EBH exhibit a tendency to conservatism in their linguistic choices, only rarely using forms outside a narrow core of what they considered literary forms. At the other extreme, the LBH authors and scribes exhibited a much less conservative attitude, freely adopting a variety of linguistic forms in addition to (not generally instead of) those favoured by the EBH scribes. Between extreme conservatism (e.g. Zechariah 1–8) and extreme openness to variety (e.g. Ezra), there was probably a continuum into which other writings may be placed (e.g. the Temple Scroll177). That we need to include not only authors but scribes in this picture is clear from those cases where we have the same biblical book in two linguistic forms, the classic example being the Book of Isaiah in its EBH MT form and the more LBH 1QIsaa.178

These two general styles of BH continued throughout the Second Temple period. We have seen here that PHab represents a continuation of the more conservative EBH approach which tended to avoid those linguistic forms favoured by LBH. Given the linguistic peculiarities mentioned in section 5, above, it is probably lacking nuance to simply label PHab’s language “EBH”. Perhaps PHab, including at least some of its non-MT linguistic features, thus represents “late EBH”179 or “Qumran EBH”. In any case, the discovery of the relationship of PHab with EBH rather than LBH is yet another sign that the chronological approach to BH has to be abandoned.

[1] I dedicate this article to my teacher and colleague, Noel Weeks to mark his retirement from the department of Ancient History at the University of Sydney. Thanks are due to Shani Berrin Tzoref, Greg Doudna, Martin Ehrensvärd, Robert Holmstedt, Robert Rezetko and Ziony Zevit, who read this article and by their comments greatly improved it; to Matthew Goff and Alan Lenzi for advice on רז‎ in Isa. 24:16; and to David Carr for discussion of section 4.4. Obviously, any remaining errors are my own responsibility.

[2] Major studies representing this view include A. Hurvitz, The Transition Period in Biblical Hebrew: A Study of Post-Exilic Hebrew and its Implications for the Dating of Psalms (Hebrew; Jerusalem: Mosad Bialik, 1972); A. Hurvitz, A Linguistic Study of the Relationship between the Priestly Source and the Book of Ezekiel: A New Approach to an Old Problem (CahRB, 20; Paris: Gabalda, 1982); M. F. Rooker, Biblical Hebrew in Transition: The Language of the Book of Ezekiel (JSOTSup, 90; Sheffield: Sheffield Academic Press, 1990); R. M. Wright, Linguistic Evidence for the Pre-exilic Date of the Yahwistic Source (Library of Hebrew Bible/ Old Testament Studies, 419; London/ New York: T&T Clark, 2005); and the articles comprising the first half of I. Young, ed., Biblical Hebrew Studies in Chronology and Typology (JSOTSup, 369; London/ New York: T&T Clark, 2003).

[3] See e.g. S. R. Driver, An Introduction to the Literature of the Old Testament (Oxford: Clarendon, 19139), 505; D. Talshir, “The Habitat and History of Hebrew during the Second Temple Period,” Biblical Hebrew Studies in Chronology and Typology, 251–75; W. M. Schniedewind, “Steps and Missteps in the Linguistic Dating of Biblical Hebrew,” HS 46 (2005), 382.

[4] On Ezekiel: Hurvitz, P and Ezekiel; Rooker, Ezekiel. On late psalms: Hurvitz, Transition Period. On the prose tale of Job: A. Hurvitz, “The Date of the Prose Tale of Job Linguistically Reconsidered,” HTR 67 (1974), 17–34. On Lamentations: F. W. Dobbs-Allsopp, “Linguistic Evidence for the Date of Lamentations,” JANESCU 26 (1998), 1–36. On Song of Songs: F. W. Dobbs-Allsopp, “Late Linguistic Features in the Song of Songs,” Perpectives on the Song of Songs (ed A. C. Hagedorn; BZAW, 346; Berlin/ New York: de Gruyter, 2005), 27–77. On Qoheleth: C. L. Seow, “Linguistic Evidence and the Dating of Qoheleth,” JBL 115 (1996), 643–66. I suspect that some, many, or all of these texts are mis-categorised as LBH-related, see for example I. Young, “Is the Prose Tale of Job in Late Biblical Hebrew?” VT (forthcoming), but for the present I will work with this categorisation for them.

[5] See I. Young, R. Rezetko, and M. Ehrensvärd, Linguistic Dating of Biblical Texts Volume 1: An Introduction to Approaches and Problems; Volume 2: A Survey of Scholarship, a New Synthesis and a Comprehensive Bibliography (Bible World; London: Equinox, 2008), 1.250–79, and the works cited in the previous for examples.

[6] P. R. Davies, “Biblical Hebrew and the History of Ancient Judah: Typology, Chronology and Common Sense”; M. Ehrensvärd, “Linguistic Dating of Biblical Texts”; R. Rezetko, “Dating Biblical Hebrew: Evidence from Samuel-Kings”; I. Young, “Late Biblical Hebrew and Hebrew Inscriptions,” Biblical Hebrew Studies in Chronology and Typology, 150–63, 164–88, 215–50, and 276–311 respectively; and J. Naudé, “A Perspective on the Chronological Framework of Biblical Hebrew,” JNSL 30 (2004), 87–102 represent early steps in this direction. I. Young, “Biblical Texts Cannot be Dated Linguistically,” HS 46 (2005), 341–51; M. Ehrensvärd, “Why Biblical Texts Cannot be Dated Linguistically,” HS 47 (2006), 177–89; R. Rezetko, “ ‘Late’ Common Nouns in the Book of Chronicles,” Reflection and Refraction: Studies in Biblical Historiography in Honour of A. Graeme Auld (eds R. Rezetko, T. H. Lim and W. B. Aucker; VTSup, 113; Leiden: Brill, 2007), 379–417; and especially Young, Rezetko and Ehrensvärd, Linguistic Dating, represent a more developed approach.

[7] I have consulted a variety of texts of PHab, in particular W. H. Brownlee, The Midrash Pesher of Habakkuk (SBLMS, 24; Missoula: Scholars, 1979); B. Nitzan, Pesher Habakkuk A Scroll from the Wilderness of Judaea (1QpHab) (Hebrew; Jerusalem: Bialik, 1986); F. García Martínez and E. B. C. Tigchelaar, The Dead Sea Scrolls Study Edition (2 vols.; Leiden: Brill; Grand Rapids: Eerdmans, 1997–98); M. P. Horgan, “Habakkuk Pesher (1QpHab),” The Dead Sea Scrolls Hebrew, Aramaic, and Greek Texts with English Translations Volume 6B Pesharim, Other Commentaries, and Related Documents (ed J. H. Charlesworth; Tübingen: Mohr Siebeck; Louisville: Westminster John Knox, 2002), 157–85; M. G. Abegg, “1QpHab,” Dead Sea Scrolls Reader Part 2 Exegetical Texts (ed D. W. Parry and E. Tov; Leiden/Boston: Brill, 2004), 78–92. I also found helpful the linguistic commentary embedded in the vocalisation of the early editions of A. M. Habermann, Megilloth Midbar Yehuda The Scrolls from the Judean Desert (Jerusalem: Machbaroth Lesifruth, 1959) and E. Lohse, Die Texte aus Qumran (München: Kösel, 1964).

[8] Schniedewind, “Steps,” 383–84; cf. F. H. Polak, “Style is More than the Person: Sociolinguistics, Literary Culture, and the Distinction between Written and Oral Narrative”, Biblical Hebrew Studies in Chronology and Typology, 98–103, and the general introduction to the chronological model in Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.10–44.

[9] For more discussion of the question of “imitation”, see section 4.4, below.

[10] M. J. Bernstein, “Pesher Habakkuk,” Encyclopaedia of the Dead Sea Scrolls (eds L. H. Schiffman and J. C. VanderKam; Oxford: Oxford University Press, 2000), 647, 649; T. H. Lim, Pesharim (Companion to the Qumran Scrolls, 3; London/ New York: Sheffield Academic, 2002), 21, 72. Lim (p.72) refers to “the presumed date of the composition of pesher Habakkuk (c.60–50 BCE)”. An alternative view proposed is the late second century BCE.

[11] S. Berrin, “Qumran Pesharim,” Biblical Interpretation at Qumran (ed. M. Henze; Grand Rapids/ Cambridge: Eerdmans, 2005), 110.

[12] E.g. R. D. Haak, Habakkuk (VTSup, 44; Leiden: Brill, 1992), 107–08, 114, 130–33; M. A. Sweeney, “Habakkuk, Book of,” ABD, 3.2; M. A. Sweeney, The Twelve Prophets Volume Two (Berit Olam; Collegeville: Liturgical Press, 2000), 454–55; O. Dangl, “Habakkuk in Recent Research,” CurBS 9 (2001), 139–44.

[13] For the criteria discussed here, see the references to Hurvitz’s work in note , and add A. Hurvitz, “Linguistic Criteria for Dating Problematic Biblical Texts,” Hebrew Abstracts 14 (1973), 74–79. A detailed introduction to Hurvitz’s methodology can be found in Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.12–23.

[14] For the term “LBH-related” see above with note .

[15] On מלכות‎ see Hurvitz, Transition Period, 79–82.

[16] For a full discussion see Young, Rezetko and Ehrensvärd, Linguistic Dating, especially 1.45–142, and on external attestation especially 1.143–72, 201–79. On the continued utility of Hurvitz’s methodology in the non-chronological model, see especially 2.102, 105.

[17] Although sometimes false oppositions can be created. For instance, is אגרת‎ used in Esther and Chronicles to designate specifically a festal letter, and hence is not in opposition to the usual word ספר‎ also used in these two books? On this, see Rezetko, “ ‘Late’ Common Nouns,” 399–400.

[18] For other examples, see section 4, below.

[19] See the discussion of Chronicles’ vocabulary in Rezetko, “ ‘Late’ Common Nouns”.

[20] See, for instance, G. A. Rendsburg, Diglossia in Ancient Hebrew (AOS, 72; New Haven: American Oriental Society, 1990); I. Young, Diversity in Pre-Exilic Hebrew. (FAT, 5; Tübingen: Mohr-Siebeck, 1993), 76–81, 87–93; M. Bar Asher, “Mishnaic Hebrew: An Introductory Survey,” HS 40 (1999), 118–119; Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.242–43. E. Qimron, “Observations on the History of Early Hebrew (1000 B.C.E.-200 C.E.) in the Light of the Dead Sea Documents,” The Dead Sea Scrolls Forty Years of Research (eds D. Dimant and U. Rappaport; Leiden/ Jerusalem: Brill/ Magnes/ Yad Izhak Ben-Zvi, 1992), 361 n.49 states: “It stands to reason that the dialects underlying both D[ead] S[ea] S[crolls] Hebrew and MH already existed in the First Temple period.”

[21] See e.g. Young, “LBH and Inscriptions,” 301–02, on the appearance of the participle of יכל‎ in the eighth century B. C. E. Arad ostracon 40:13–14.

[22] A. Hurvitz, “The Language of the Priestly Source and its Historical Setting—the Case for an Early Date,” Proceedings of the 8th World Congress of Jewish Studies, Jerusalem, August 16–21, 1981, Panel Sessions Bible and Hebrew Language (Jerusalem: World Union of Jewish Studies, 1983), 92 (his italics). On Aramaic, see further Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.201–22.

[23] KBL, 1.291.

[24] On בוץ‎ see Young, “LBH and Inscriptions,” 283. For other examples where the criterion of external attestation is ignored or misused see Young, “LBH and Inscriptions,” 277–80, nn. 3–5.

[25] Compare A. Hurvitz, “The Historical Quest for ‘Ancient Israel’ and the Linguistic Evidence of the Hebrew Bible: Some Methodological Observations,” VT 47 (1997), 312 with the Assyrian evidence in F. M. Fales, Aramaic Epigraphs on Clay Tablets of the Neo-Assyrian Period (Studi Semitici Nuova Serie, 2; Roma: Università Degli Studi “La Sapienza”, 1986), 185 etc; M. L. Folmer, The Aramaic Language in the Achaemenid Period A Study in Linguistic Variation (OLA, 68; Leuven: Peeters, 1995), 629–32; cf. Rezetko, “ ‘Late’ Common Nouns,” 399–400.

[26] DNWSI, 2.644.

[27] For detailed substantiation of this important fact see Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.83–87, 111–19.

[29] Or more extensive textual reworking such as the introduction of later sections either short or long.

[30] For the argument that the language of the biblical texts has been changed in transmission, see Young “Biblical Texts,” 349–51; Young, Rezetko and Ehrensvärd, Linguistic Dating, especially 1.341–60; 2.100–01, with references to earlier studies.

[31] E.g. R. Polzin, Late Biblical Hebrew: Toward an Historical Typology of Biblical Hebrew Prose (HSM, 12; Missoula: Scholars Press, 1976), 66.

[32] Rezetko, “Dating Biblical Hebrew,” 230–31.

[33] H. Cazelles, “Jeremiah and Deuteronomy,” A Prophet to the Nations Essays in Jeremiah Studies (eds L. G. Perdue and B. W. Kovacs; Winona Lake: Eisenbrauns, 1984; trs. L. Perdue from “Jérémie et le Deutéronome,” Recherches de Science Religieuse 38 [1951] 5–36), 93.

[34] A number of other such examples are presented in Young, Rezetko and Ehrensvärd, Linguistic Dating, 2.103–04, 106–59.

[35] Hurvitz, P and Ezekiel; Rooker, Ezekiel.

[36] E. Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll (1QIsaa) (STDJ, 6; Leiden: E. J. Brill, 1974), 34; Rooker, Ezekiel, 134–38. Of course it is also the preferred form in many EBH texts such as Samuel.

[37] See e.g. Rooker, Ezekiel, 72–74.

[38] I here present those forms which, even if somewhat dubious, can be accepted as LBH according to my loose definition of it. In addition I discuss פשר‎ since this is routinely introduced as a LBH word. There is obviously no space to discuss in detail my reasons for not accepting other forms. It will be clear from those forms I accept below that I have used a very broad definition of LBH, and that any rejected forms must therefore be very far from compelling. Thus, note Qimron’s list “Words Mainly Attested in the DSS and in the Late Biblical Books” (E. Qimron, The Hebrew of the Dead Sea Scrolls [HSS, 29; Atlanta: Scholars, 1986] 88–97; in fact it also discusses syntactical issues such as collocations of verbs with prepositions). This list contains remarkably few entries related to PHab. Of those that are, several are acceptable here. I understand Qimron’s list to be maximal, and hence it includes several items that are in the category of possible and/or tangential links with LBH. Thus, for example, he connects אמנה‎ “faith” in PHab 8:2 with LBH. BDB find this meaning for this form only in Neh. 10:1, which would hardly qualify it as characteristic of LBH (BDB, 53b). However, the latest Koehler-Baumgartner dictionary glosses this word in Neh. 8:1 as “agreement” and connects it with the word ברית‎ “covenant” (KBL, 1.64). אמנה‎ in PHab 8:2 is rather to be seen as related to the common BH אֱמוּנָה‎ (Brownlee, Midrash Pesher, 125; Nitzan, Pesher Habakkuk, 176).

[39] G. J. Brooke, “The Biblical Texts in the Qumran Commentaries: Scribal Errors or Exegetical Variants?” Early Jewish and Christian Exegesis: Studies in Memory of William Hugh Brownlee (eds C. A. Evans and W. F. Stinespring; Atlanta: Scholars, 1987), 85–100; G. L. Doudna, 4Q Pesher Nahum A Critical Edition (JSPSup, 35; London: Sheffield Academic, 2001), 67–70; Lim, Pesharim, 18, 54–63; J.-H. Kim, “Intentionale Varianten der Habakukzitate im Pesher Habakuk Rezeptionsästhetisch Untersucht,” Bib 88 (2007), 23–37.

[40] Its precise meaning in Qoheleth is not clear from the context. For a recent study see S. C. Jones, “Qoheleth’s Courtly Wisdom: *Ecclesiastes 8:1–9,” CBQ 68 (2006), 211–28.

[41] M. P. Horgan, Pesharim: Qumran Interpretations of Biblical Books (CBQMS 8; Washington: Catholic Biblical Association of America, 1979), 236; H. J. Fabry and U. Dahmen, “פֵּשֶׁרpešer; פָּתַרpāṯar; פִּתָּרוֹן‎/פִּתְרוֹןpittārôn/piṯrôn,” TDOT 12 (2003), 152. It is important to recall, however, that פתר‎ is not attested in any Aramaic text.

[42] R. Degen, Altaramische Grammatik der Inschriften des 10.-8. Jh. V. Chr. (Wiesbaden: Franz Steiner, 1969), 32–36, 43; S. Segert, Altaramäische Grammatik (Leipzig: VEB Verlag Enzyklopädie, 1990), 92; Folmer, Aramaic Language, 70–74; T. Muraoka, and B. Porten, A Grammar of Egyptian Aramaic (Leiden: Brill, 20032), 7–8.

[43] In any case, I consider Qoheleth a pre-exilic book, see Young, Diversity, 140–57; Young “Biblical Texts,” 347–48; M. A. Shields, The End of Wisdom A Reappraisal of the Historical and Canonical Function of Ecclesiastes (Winona Lake, IN: Eisenbrauns, 2006), 22–27; cf. C. Rabin, “The Song of Songs and Tamil Poetry,” SR 3 (1973–74), 216. Of course, like any biblical text, Qoheleth was the subject of constant scribal reworking, cf. the many variant details of the Qumran manuscripts 4QQoha and 4QQohb (I. Young, “The Biblical Scrolls from Qumran and the Masoretic Text: A Statistical Approach,” Feasts and Fasts A Festschrift in Honour of Alan David Crown [eds M. Dacy, J. Dowling and S. Faigan; Mandelbaum Studies in Judaica, 11; University of Sydney: Mandelbaum, 2005], 102). There is thus no reason to believe that a pre-exilic origin of a book necessitates that every detail of the MT form of the book is pre-exilic. This especially applies to language, which is second only to orthography in its exposure to updating (Young “Biblical Texts,” 349–51).

[44] Horgan, Pesharim, 235.

[45] Berrin, “Qumran Pesharim,” 113.

[46] M. G. Abegg, The Dead Sea Scrolls Concordance Volume One The Non-Biblical Texts from Qumran (Leiden/Boston: Brill, 2003), 629 lists occurrences in two manuscripts of the Damascus Document, as well as one in 4Q298, the “Words of the Maskil to All Sons of Dawn”.

[47] Brownlee, Midrash Pesher, 173; Horgan, “Habakkuk Pesher,” 179; M. Wise, M. Abegg and E. Cook with N. Gordon, “1QpHab, translation,” Dead Sea Scrolls Reader Part 2 Exegetical Texts (eds D. W. Parry and E. Tov; Leiden/Boston: Brill, 2004), 89, among others, suggest that אשר‎ in 10:13 should be rendered “since” or “because”. However, Brownlee, and Horgan, Pesharim, 46 point out that a translation like “who” is also possible, thus e.g. G. Vermes, The Complete Dead Sea Scrolls in English (London: Penguin, 1997), 483. So too in 12:5 Brownlee, Midrash Pesher, 196 translates אשר‎ as “for”, but, for example, Horgan, “Habakkuk Pesher,” 183 translates it as “whom”, and cf. Horgan, Pesharim, 51–52.

[48] E.g. Rooker, Ezekiel, 111–12.

[49] On the introductory formulas in the Pesharim see Horgan, Pesharim, 239–44. For a listing of the use of על‎ in such formulae in PHab, see note , below.

[50] R. D. Holmstedt, “The Story of Ancient Hebrew ʾasher,” ANES 43 (2006), 10 n.10; cf. R.D.Holmstedt, The Relative Clause in Biblical Hebrew: A Linguistic Analysis (Ph.D. diss., University of Wisconsin—Madison, 2002), 294 n.25.

[51] Cf. also the LBH-related Qoheleth with nine. However, also note four cases in EBH Samuel.

[52] M. Moreshet, “Hiph‛il le-lo’ hevdel min ha-Qal bi-lshon HaZaL (be-hashva‛ah li-lshon ha-Mikra,” Sefer Bar-Ilan 13 (1976), 249–81; Polzin, Late Biblical Hebrew, 133–34; Qimron, Hebrew, 49.

[53] Segert in Brownlee, Midrash Pesher, 75. This is possible because of the great similarity of waw and yod in the script of PHab. For “pausal” forms in non-pausal positions in Qumran Hebrew see Qimron, Hebrew, 50–53; cf. e.g. יקבוצו‎ in 9:5.

[54] BDB, 541b.

[55] For Job as ABH see D. A. Robertson, Linguistic Evidence in Dating Early Hebrew Poetry (SBLDS, 3; Missoula, MT: Scholars, 1972), 149, 155. For the problems with defining ABH see Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.312–40.

[56] See the translations cited in this study in notes 7 and 133. Brownlee, Midrash Pesher, 59–61 restores [יעו‎]ירש‎ in the sense “act wickedly” also at 2:14, but most read the traces and restore differently.

[57] BDB, 957 notes this latter usage as “late”; cf. Qimron, Hebrew, 95. The LBH texts cited are Dan. 9:5; 11:32; 12:10; Neh. 9:33; 2 Chron. 20:35; 23:3. There are three non-LBH texts. The poetry of Job regularly uses the hiphil in the sense “condemn as guilty”, but in 34:12 it has the sense “act wickedly”. Psalm 106 has “act wickedly” in verse 6; it is not one of Hurvitz’s LBH psalms since it has no accumulation of LBH features (Hurvitz, Transition Period, 173). Finally, although it is commonly emended, the MT of the EBH 1 Sam. 14:47 seems to read “wherever [Saul] turned, he acted wickedly” (P. K. McCarter, 1 Samuel [AB, 8; New York: Doubleday, 1980], 254).

[58] Cf. e.g. Wise et al., “1QpHab,” 89.

[59] S. Berrin, “Lemma/ Pesher Correspondence in Pesher Nahum,” The Dead Sea Scrolls Fifty Years After their Discovery Proceedings of the Jerusalem Congress, July 20–25, 1997 (eds L. H. Schiffman, E. Tov and J. C. VanderKam; Jerusalem: Israel Exploration Society, 2000), 341–50; S. Berrin, The Pesher Nahum Scroll from Qumran An Exegetical Study of 4Q169 (STDJ, 53; Leiden/Boston: Brill, 2004), 18–19, quote from p.18.

[60] See the translations.

[61] Polzin, Late Biblical Hebrew, 28–31; Rooker, Ezekiel, 86–87; Wright, Linguistic Evidence, 37–41.

[62] Polzin, Late Biblical Hebrew, 28–31.

[63] 4:7, 7, 8; 5:11; 7:2, 4; 8:2; 9:10, 10; 10:4, 5, 5, 11; 11:5, 7, 8, 15; 12:5, 13, 13, 14.

[64] Hab. 1:3, 10, 12, 12, 15, 15; 2:2, 7, 8, 11, 17, 17, 18; 3:2, 10, 14, 16, 19.

[65] But see section 4.4 on the relationship between the language of PHab and that of biblical Habakkuk.

[66] Nah. 1:4, 12, 12; 2:3, 4; 3:6, 6, 15, 15, 15.

[67] 1 Kgs 2:5, 8, 8, 9, 24, 24, 26, 30, 31, 32, 34, 42; 3:1, 20, 26, 27.

[68] Ruth 1:16, 21; 2:4, 9, 10, 13, 15, 19; 3:6, 13, 13, 13; 4:15, 15, 16.

[69] E.g. Rooker, Ezekiel, 115 n.167; F. H. Polak, “The Oral and the Written: Syntax, Stylistics and the Development of Biblical Prose Narrative,” JANESCU 26 (1998), 104–05; A. F. Rainey, “Mesha and Syntax,” The Land That I Will Show You: Essays on the History and Archaeology of the Ancient Near East in Honour of J. Maxwell Miller (eds J. A. Dearman and M. P. Graham; JSOTSup, 343; Sheffield: Sheffield Academic, 2001), 287–307.

[70] Mesha 4, 4, 6, 8–9, 11, 12–13, 15–16, 17, 18?, 19, 20, 20.

[71] Hurvitz, Transition Period, 22; Rooker, Ezekiel, 127–31.

[72] על‎: 1:3, 4; 2:3, 10, 12; 3:4, 4, 9; 4:2, 2, 5, 5, 6; 5:9, 11; 6:7, 10, 11; 7:1, 4, 7, 10, 12, 15; 8:1, 8, 9, 12, 16; 9:4, 9, 12, 16; 10:9, 11:4, 12; 12:2, 3, 12; 13:1. Note in addition that [ע]ל‎ is restored by various scholars in 2:7; 4:10 and 11:15, and is repeated due to dittography in 7:2. אל‎: 7:1(אל‎); 11:7 (אליהם‎). Some scholars have interpreted the unusual form אבית‎ in PHab 11:6 as a contraction of אל בית‎ (e.g. Horgan, Pesharim, 49). Others take it as equivalent to בבית‎ (e.g. Brownlee, Midrash Pesher, 182–84; Nitzan, Pesher Habakkuk, 190). Qimron interprets it as having a prosthetic aleph prefixed to בבית‎ to avoid an initial consonant cluster (Qimron, Hebrew, 39). The first beth would then have assimilated to the second, leaving just אבית‎ (S. Morag, “Language and Style in Miqsat Ma‛ase Ha-Torah: Did Moreh ha-Sedeq Write This Document?” Tarbiz 65 (1996), 213 [in Hebrew]).

[73] 2:12; 3:4, 9; 4:5; 5:9; 6:10; 7:4, 10; 8:1, 8, 16; 9:4, 9, 16; 10:9; 11:4, 12; 12:2, 12; 13:1.

[74] Already partially noted by K. Elliger, Studien zum Habakuk-Kommentar vom Toten Meer (BHT, 15; Tübingen: J. C. B. Mohr [Paul Siebeck], 1953), 82; Nitzan, Pesher Habakkuk, 121; Qimron, Hebrew, 88–97.

[75] For LBH, however, note the unique use of the hiphil of שׂחק‎ in 2 Chron. 30:10, which coordinates with על‎.

[76] Hurvitz, Transition Period, 107–09.

[77] Brownlee, Midrash Pesher, 116 discusses the various options for understanding this word, as a verb or as a noun.

[78] Although cf. Neh. 9:30 where the sense of the qal is “you were patient” (NRSV). Brownlee, Midrash Pesher, 120 sees a specific “subjective connotation” suggested by the use of על‎ with this verb in PHab: “when to them the last time seems to be delayed”.

[79] על‎: Hab. 1:4, 4, 15, 16, 17; 2:1, 1, 1, 2, 6, 6, 14, 15, 16, 16, 18; 3:1, 8, 19. Note that this includes five cases of על כן‎ in chapter 1. אל‎: Hab. 1:2, 13; 2:5, 5.

[80] Note that the lemma text of Hab. 2:15 in PHab 11:3 has the more common biblical expression הבט אל‎. All other preserved sections have the על‎ in common with the MT.

[81] But see below, section 4.4 on the relationship between the language of PHab and that of biblical Habakkuk.

[82] Thus, PHab 9:11–12 could have been read “God condemned his chosen” if אל‎ had been used rather than על‎. Note also the scribal correction in 7:1 where the original scribe wrote וידבר אל חבקוק‎ which was corrected by a second scribe (Nitzan, Pesher Habakkuk, 171) to read “And God (אל‎) spoke to (אל‎) Habakkuk”. Whether this is a clarification or correction of a scribal error is not clear due to the broken context. If a scribal error it illustrates the chances for possible confusion between the divine name and the preposition.

[83] See e.g. Hurvitz, Transition Period, 37–38; Polzin, Late Biblical Hebrew, 42–43; Rooker, Ezekiel, 75–77.

[84] Cf. Nitzan, Pesher Habakkuk, 169.

[85] See Jer. 1:16; 44:8; Ps. 8:7; 92:5; 111:7; 138:8; as well as the core LBH 2 Chron. 34:25.

[86] S. Gevirtz, “Of Syntax and Style in the ‘Late Biblical Hebrew’—‘Old Canaanite’ Connection,” JANESCU 18 (1986), 28–29; G. A. Rendsburg, “Hurvitz Redux: On the Continued Scholarly Inattention to a Simple Principle of Hebrew Philology,” Biblical Hebrew Studies in Chronology and Typology, 113–15; Rezetko, “Dating Biblical Hebrew,” 231–33.

[87] Hence the fabulous Authorized Version translation: “and thou shalt be for booties unto them”!

[88] Note the use of דמים‎ in PHab 10:10. However, the use of the plural of this word is too common for it, on its own, to be considered a clear case of pluralisation.

[89] And see below, section 4.4, on the relationship between the language of PHab and that of biblical Habakkuk.

[90] The evidence of the versions is cited in H. Wildberger, Jesaja 2.Teilband Jesaja 13–27 (BKAT; Neukirchen-Vluyn: Neukirchener Verlag, 1978), 932; J. Niehaus, “Rāz-Pešar in Isaiah XXIV,” VT 31 (1981), 376, 378; J. D. W. Watts, Isaiah 1–33 (WBC, 24; Waco: Word, 1985), 324. 1QIsaa has the same consonants as the MT. Brownlee, Midrash Pesher, 111 sees the specific influence of Isa. 24:16 in column 7 of PHab.

[91] See e.g. G. B. Gray, A Critical and Exegetical Commentary on The Book of Isaiah I–XXVII (ICC; Edinburgh: T&T Clark, 1912), 419; O. Kaiser, Isaiah 13–39 A Commentary (OTL; London: SCM, 19802), 189–90; Niehaus, “Rāz”; J. Blenkinsopp, Isaiah 1–39 (AB, 19; New York: Doubleday, 2000), 353–54. The latest edition of the Koehler-Baumgartner lexicon thinks that this and the alternative translation to be discussed “appear equally possible”: KBL, 3.1210.

[92] E.g. I. Willi-Plein, “Das Geheimnis der Apocalyptik,” VT 27 (1977), 73; Watts, Isaiah, 323–24. One wonders whether another factor is the reluctance to see Persian words where they are not “supposed” to be, as documented in Young, “LBH and Inscriptions,” 284–85; cf. Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.289–309.

[93] KBL, 3.1209.

[94] As pointed out by Gray, Isaiah, 419; cf. KBL, 3.1209–10.

[95] Wildberger, Jesaja, 932.

[96] Gray, Isaiah, 419 and Blenkinsopp, Isaiah, 355 refer to a “glossator”. Against this, Willi-Plein, “Geheimnis,” 71–72 argues that the Septuagint represents a simplification, and hence is not original.

[97] E.g. Kaiser, Isaiah, 173–79.

[98] Gray, Isaiah, 463–72 argues for the lateness of the language and style of Isaiah 24–27, but his arguments have not been carried on in modern discussions of LBH, probably partly because of the limitations of his methodology. On the contrary, Wright, Linguistic Evidence, 68 n.53 considers a date in the exile, with reference to W. M. Millar, “Isaiah 24–27 (Little Apocalypse),” ABD 3.489, and refers also to the study of S. B. Noegel (“Dialect and Politics in Isaiah 24–27,” AuOr 12 [1994] 177–92) who considers the linguistic peculiarities of Isaiah 24–27 as features of northern, Israelian Hebrew. See also the discussion of Isaiah in Young, Rezetko and Ehrensvärd, Linguistic Dating, 2.33–35.

[99] Or more precisely an Iranian loanword, deriving from Avestan according to KBL, 5.1980–81.

[100] C. L. Seow, “Linguistic Evidence and the Dating of Qoheleth,” JBL 115 (1996), 646–50; M. Eskhult, “The Importance of Loanwords for Dating Biblical Hebrew Texts,” Biblical Hebrew Studies in Chronology and Typology, 12–14.

[101] Eskhult, “Loanwords,” 14 n.10; Young, “LBH and Inscriptions,” 284–85; Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.303–09.

[102] In fact, by necessity sometimes texts will not be 500 words in length. Thus we discuss below 2 Samuel 22//Psalm 18, which texts each have only about 380 words. The biblical book of Habakkuk has 671 words, but if we just wished to examine chapters 1–2, since chapter 3 is not commented on in PHab, we have only 459 words. Nevertheless, where at all possible, the stretch of text analysed is 500 words in each case.

I use the term “words” to refer to Hebrew graphic units. Thus וּבָעִיר‎ “and in the city” counts as but one “word”, rather than four. Hebrew graphic units correspond on average to about 1.5 words in this latter sense, and hence a 500 word (graphic unit) sample is approximately equivalent to a 750 word sample in English. D. Biber, “Methodological Issues Regarding Corpus-based Analyses of Linguistic Variation,” Literary and Linguistic Computing 5 (1990), 258–61 argues that a 1000 word English sample is reliable for analyses of linguistic variation of grammatical features. Cf. C. L. Miller, “Methodological Issues in Reconstructing Language Systems from Epigraphic Fragments,” The Future of Biblical Archaeology Reassessing Methodologies and Assumptions (eds J. K. Hoffmeier and A. Millard; Grand Rapids/ Cambridge: Eerdmans, 2004), 285 for the application of this principle to ancient Semitic linguistics. Note that Biber is not arguing that 1000 words is a minimum, only that 1000 words is adequate. The argument being made here is rather less linguistically sophisticated than the studies for which Biber found 1000 words adequate. 500 graphic units represents a compromise between having a large enough sample, and the problem that too large a sample size will render the method unable to be used on texts of the size of biblical Habakkuk or PHab.

[103] Data from Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.129–36. Out of the data presented there, here we concentrate on samples from the core LBH books and each of the core elements in EBH, i.e. the Pentateuch, the Deuteronomistic History, pre-exilic prophets, and EBH poetry. I also include the pre-exilic Arad inscriptions, the post-exilic prophet Zechariah (and of course PHab) since I discuss them later.

[104] There is no space in this context to justify each feature judged to be LBH; below we merely list them. Although Rezetko, Ehrensvärd and I have thoroughly checked the samples, it is still possible that we have missed some forms in some of the samples, but the results are so clear that a slight adjustment here or there will not affect the picture that emerges.

[105] יָה‎- names (1:1; 10:2); מלכות‎ and עבדות‎ with וּת‎- afformative (1:1; 9:8, 9); אלהי השמים‎ (1:2); motion verb + לְ‎ (1:3, 11); בית האלהים‎ (1:4, 9:9, 10:1) ותיהם‎- (1:6; 9:1, 2, 11, 12); נדבhithpael (1:6); על יד‎ (1:8); Persian words (1:8, 9); substantive before numeral (1:9 [x3], 10 [x3], 11); כפורים‎ (1:10 [x2]); זהב … כסף‎ order (1:11); (û)b/keqotlô temporal clause (9:1, 3, 5; 10:1 [x2]); נשׂא‎ as ‘to marry’ (9:2, 12); weqatálti (9:2, 13); double plurals (9:1, 2, 11, 14); waʾeqtlah (9:3 [x2], 5 [x2], 6); שמםpoel participle; עד ל‎- (9:4, 6); בִּזָּה‎ (9:3, 4); עמדhiphil (9:9); אחרי זאת‎ (9:10); לאין‎ (9:14); ידהhithpael (10:1); preference for verb suffixes 8–0 (1:4, 7, 8 [x2]; 9:8, 9, 11 [x2]).

[106] מלכות‎ with וּת‎- afformative (1:1, 20); מקצת‎ (1:2); בית האלהים‎ (1:2); infinitive for direct speech (1:3, 4, 18); Persian words (1:3, 5, 8, 13, 15, 16; 11:45); מדע‎ (1:4, 17); מנהpiel (1:5, 10, 11); משתה‎ as ‘drinking’ (1:5, 8, 10, 16); substantive before numeral (1:5, 12, 14, 15; 12:11, 12); יָה‎- names (1:6 [x2], 7 [x2], 11 [x2], 19 [x2]); גאלhithpael (1:8); אשר‎ for כי‎ (1:8 [x2]); pluralisation (1:15; 12:2); nun of מן‎ unassimilated (1:15); היה‎ + participle (1:16); בין ב-hiphil (1:17); בין..ל…‎ (11:45); עמד‎ for קום‎ (12:1, 13); weqatálti (12:5); (û)b/keqotlô temporal clause (12:7; cf. 1:15, 18); waʾeqtlah (12:8); weyiqtol instead of weqataltí (12:10 [x2], 13 [x2]); רשעhiphil for qal (12:10); preference for verb suffixes 8–0 (1:2, 4, 5, 14, 18 [x2], 20; 11:44).

[107] על‎ instead of another preposition (30:1 [x2], 9, 18, 22); אגרת‎ (30:1, 6); infinitive for direct speech (30:1, 5); עמדhiphil (30:5; 31:2);ותיהם‎- (30:7, 22); חנון … רחום‎ order (30:9); היה‎ + participle (30:10 [x2]); לעגhiphil for qal (30:10; cf. שׂחקhiphil); motion verb + ל‎ (30:11); postpositive לרב‎ in the sense ‘a lot of’ (30:13, 24); day-month word order (30:15); pluralisation (30:17); מצאniphal as ‘to be present’ (30:21; 31:1); ידהhithpael (30:22); רוםhiphil as ‘to contribute for sacrifice’ (30:24 [x2]); substantive before numeral (30:24 [x2]); (û)b/keqotlô temporal clause (31:1); עד ל‎- (31:1); אחזה‎ (31:1); מחלקות‎ as ‘divisions’ of people (31:2 [x2]); nun of מן‎ unassimilated (31:3); preference for collectives with plural verbs 6–0 (30:3, 13, 17, 18, 23; 31:1; cf. 30:25).

[108] יָה‎- names (1:1 [x2]; 2:10); בירה‎ (1:1, 2:8); מדינה‎ (1:3); בכהwayyiqtol + long III-He (1:4); waʾeqtlah (1:4; 2:1, 6, 9, 13); היה‎ + participle (1:4; 2:13, 15 [x2]); אלהי השמים‎ (1:4, 5; 2:4); ידהhithpael (1:6); המלך‎ ZYX (2:1); חיה‎ jussive + long III-He (2:3); אשר‎ for כי‎ (2:3, 5, 7, 8, 10, 17; Holmstedt [e-mail 21.05.06] considers most of these examples uncertain; only Neh. 2:10 is cited in Holmstedt, Relative Clause, 294 n. 25; “Story,” 10 n. 10); אם על המלך טוב‎ (2:5, 7); יטב לפני‎ (2:5, 6); מהלך‎ (2:6); זמן‎ (2:6); אגרת‎ (2:7, 8, 9); על‎ instead of another preposition (2:7; cf. 2:4); Persian word (2:8); substantive before numeral (2:11); preference for verb suffixes 8–0 (1:2, 9 [x2], 11; 2:5 [x2], 6, 7).

[109] מלכות‎ with וּת‎- afformative (5:1 [x3], 3, 6; 6:8 [x2]); בַּקָּשָׁה‎ (5:3, 6, 7, 8); אם על המלך טוב‎ (5:4, 8); (û)b/keqotlô temporal clause (5:9); זועqal (5:9); substantive before numeral (5:14); יטב לפני‎ (5:14); infinitive for direct speech (6:1, 4); היה‎ + participle (6:1); אשר‎ for כי‎ (6:2); גְּדוּלָּה‎ (6:3); לאמר‎ as embedded infinitive expressing purpose/result (6:4); יותר מן‎ (6:6); על יד‎ (6:9); Persian word (6:9); דחף‎ (6:12); preference for verb suffixes 5–0 (5:11 [x2]; 6:9, 11, 13).

[110] Substantive before numeral (1:3, 7; 16:5; and many other cases); weqatálti (3:2–3; 16:4); על‎ instead of another preposition (3:3); (û)b/keqotlô temporal clause (16:3); על יד‎ (24:15); nun of מן‎ unassimilated (26:2); רצה‎ as ‘to want’ (40:6–7); יָה‎- names (107:2; 110:1, 2); לקחniphal for qal passive (111:4).

[111] אל‎/על‎ interchange/על‎ instead of another preposition (22:6 [cf. 22:15], 17, 32); יָה‎- name (22:11); אשר‎ for כי‎ (22:16); masculine plural suffix for feminine plural (22:17); נכה‎ and עלהwayyiqtols + long III-He (22:24, 34, 35); substantive before numeral (22:31); היה‎ + participle (22:35; cf. 2 Chron. 18:34); preference for verb suffixes 6–0 (22:8, 16, 21, 26, 27, 34; note that את‎ + suffix in 22:14 is forced).

[112] פוץhiphil for qal (13:8); עלהwayyiqtol + long III-He (13:12); אל‎/על‎ interchange (13:13; 14:4; cf. 13:12); מצאniphal as ‘to be present’ (13:15, 16); ה‎ definite article non-syncope (13:21); יָה‎- name (14:3).

[113] אל‎/על‎ interchange/על‎ instead of another preposition (6:3, 10; cf. 6:6); weqatálti (6:16); היה‎ + participle (6:16; 7:6); ל‎ for את‎ (6:16); היהwayyiqtol + long III-He (7:6, 9); waeqtlah (7:9).

[114] Nun of מן‎ unassimilated (22:14); pluralisation (22:22, 48, 49; cf. 22:12); היהwayyiqtol + long III-He (22:24); waʾeqtlah (22:24); absence of cohortative (22:50; cf. Ps. 18:50); preference for verb suffixes 31–2 (22:3, 5 [x2], 6 [x2], 15, 15 [Kethib], 17 [x2], 18, 19, 20, 21, 34, 36, 38, 39 [x2], 40, 41, 42, 43 [x3], 44 [x3], 49 [x3], 50 vs. 22:1, 20).

[115] Since 2 Samuel 22 contains only 382 words, the figure in parentheses gives the projected number of LBH features in a 500 word sample.

[116] Absence of locative he (2:3, 6, 8, 9; cf. 2:26); מלכות‎ with וּת‎- afformative (2:12); המלך‎ ZYX (2:17); על‎ instead of another preposition (2:26); יָה‎- names (2:28; cf. 2:5, 22 with צרויה‎, but the etymology is disputed); preference for verb suffixes 7–0 (2:5, 8 [x2], 9, 24 [x2], 26).

[117] Nun of מן‎ unassimilated (18:4, 49); pluralisation (18:22, 48); absence of cohortative (18:38; cf. 2 Sam. 22:38); על‎ instead of another preposition (18:42); ותיהם‎- (18:46); preference for verb suffixes 31–1 (18:2, 5 [x2], 6 [x2], 15 [x2], 17 [x2], 18, 19, 20 [x2], 21, 33, 34, 36 [x2], 38, 39, 40, 41, 42, 43 [x2], 44 [x3], 49 [x2], 50 vs. 18:1).

[118] Since Psalm 18 contains only 394 words, the figure in parentheses gives the projected number of LBH features in a 500 word sample.

[119] Biblical quotes are excluded from the sample. אשר‎ for כי‎ (5:3, 7; 6:3, 6; 7:7, 15); רשעhiphil for qal (9:11); אל‎/על‎ interchange/על‎ instead of another preposition (5:11; 6:11; 7:7, 12, 15; 9:12; 12:3); pluralisation (6:4; cf. 8:12–13; 12:8); Persian word (7:5, 8, 14); preference for verb suffixes 17–0 (5:11; 7:2, 4; 8:2; 9:10 [x2]; 10:4, 5 [x2], 11; 11:5, 7, 8, 15; 12:5, 13 [x2]).

[120] עשׂהwayyiqtol + long III-He (1:14); pluralisation (2:7, 8, 17); על‎ instead of another preposition (2:14, 15, 18; cf. 2:16); זהב … כסף‎ order (2:19); preference for verb suffixes 15–0 (1:3, 10, 12 [x2], 15 [x2]; 2:2, 7, 8, 11, 17 [x2], 18; 3:2, 4).

[121] M. Noth, A History of Pentateuchal Traditions (Scholars Press reprint series, 5; trs. B. W. Anderson; Chico: Scholars, 1981), 29, 264.

[122] אלהי השמים‎ (24:3, 7); אשר‎ for כי‎ (24:3; Holmstedt, Relative Clause, 294 n. 25 does not cite this example but Holmstedt, “Story,” 10 n. 10 does); אל‎/על‎ interchange (24:11); preference for verb suffixes 7–0 (24:3, 7, 16, 17, 18, 19, 27; note that את‎ + suffix in 24:14 is forced).

[123] יָה‎- names (1:1 [x2], 7); day-month word order (1:7); motion verb + לְ‎ (1:16).

[124] E.g. Noth, Pentateuchal Traditions, 18, 268.

[125] Preference for אני‎ over אנכי‎ 8–0 (6: 2, 5, 6, 7, 8, 12; 7:3, 5).

[126] Ehrensvärd, “Linguistic Dating”; Ehrensvärd, “Why Biblical”.

[127] See further Young, Rezetko and Ehrensvärd, Linguistic Dating, vol. 1, especially 1.56, 106–09, 119–29, 137–41, 250–79.

[128] As stated above, whether this is due to imitation or due to natural continuation of the EBH style is irrelevant to the point being made, but see section 4.4 below.

[129] The aim in this section is to describe PHab’s linguistic links with EBH as opposed to LBH. Thus, while interesting, and relevant in broader discussions about PHab’s language, we do not deal with the question of which of these linguistic usages are common and which are unusual in other Qumran texts.

[130] מאש‎ in 1:11; cf. below section 5.

[131] In LBH-related: Ezekiel six times; twice in Lamentations.

[132] BDB notes this usage as “late” (BDB, 276a). The qal of זנח‎, with the same sense as the hiphil, is also not used in core LBH.

[133] The older BDB dictionary glosses עריץ‎ as “awe inspiring, terror inspiring” (BDB, 791–92). The recent edition of the Koehler-Baumgartner dictionary gives the glosses “violent, powerful, acting violently, potentate, tyrant” (KBL, 2.884). Translations of PHab vary: “ruthless ones” (Horgan, Pesharim, 13; Horgan, “Habakkuk Pesher,” 163); “cruel” (M. Wise, M. Abegg, Jr. and E. Cook, The Dead Sea Scrolls A New Translation [Rydalmere: Hodder & Stoughton, 1996], 116); “men of violence and breakers (of the covenant)” (Vermes, Complete Scrolls, 479); “violators (of the covenant)” (Brownlee, Midrash Pesher, 53; García Martínez and Tigchelaar, Dead Sea Scrolls, 1.13); “enemies” (Wise et al., “1QpHab,” 81).

[134] BDB, 791–92; KBL, 2.888.

[135] Note the regular practice of the Targum to render ממשלה‎ with a form of שלט‎.

[136] As with ממשלה‎ above, משל‎ is commonly rendered in the Targum with words from the root שלט‎.

[137] This is not specifically a LBH term, see e.g. 2 Kgs 25:4. The point is that Chronicles did not use the exclusively EBH term תפש‎ here.

[138] Although not directly paralleled in Samuel, the verb הרס‎ is used of the Ammonite capital in David’s instructions in 2 Sam. 11:25, and hence this EBH passage may have influenced Chronicles’ linguistic choice in describing the fall of the Ammonite capital.

[139] Rooker, Ezekiel, 142.

[140] דמה‎ is favoured by, for example, KBL, 1. 225. The two roots are discussed by H. G. M. Williamson, “The Translation of 1 Q p Hab. V, 10,” RevQ 9 (1977–78), 263–65 who argues that a translation “they were reduced to silence” fits the meaning of the roots better. However, recent translations that I consulted have not followed Williamson’s suggestion.

[141] On Malachi as post-exilic EBH, see Ehrensvärd, “Linguistic Dating,” especially 175–86.

[142] Although these two texts exhibit only the negative sense “craftiness”.

[143] On כנס‎ as LBH see e.g. Hurvitz, P and Ezekiel, 123–25; Rooker, Ezekiel, 156–58.

[144] On גופה‎ as LBH see Polzin, Late Biblical Hebrew, 132.

[145] J. J. Collins, Daniel A Commentary on the Book of Daniel (Hermeneia; Minneapolis: Fortress, 1993), 128 n.31; KBL, 1.295. For חוב‎ as an Aramaism see M. Wagner, Die Lexicalischen und Grammatikalischen Aramaismen im Alttestamentlichen Hebräisch (Berlin: Alfred Töpelmann, 1966), 52.

[146] Brownlee, Midrash Pesher, 91; Nitzan, Pesher Habakkuk, 166, among others, restore ע[דת]ם‎ in PHab 5:12 rather than ע[צת]ם‎ as in e.g. Horgan, “Habakkuk Pesher,” 168.

[147] A. Hurvitz, “Linguistic Observations on the Biblical Usage of the Priestly Term ‛Edah’,” Tarbiz 40 (1970–71), 261–67 (in Hebrew); Hurvitz, P and Ezekiel, 65–67.

[148] See 2 Kgs 10:19, 21, 22, 23; Ps. 97:7.

[149] On the “loose” LBH status of this non-assimilation, see above in section 3.1.

[150] Polzin, Late Biblical Hebrew, 45. The relevance of Mishnaic Hebrew to BH chronology is seriously questionable, since Mishnaic Hebrew is widely believed to represent a parallel, co-existing dialect, not a genetic descendent of BH; see above, section 3.1 and the discussion in Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.223–49; 2.72–77.

[151] Note the following quotes: from M. S. Smith, The Origins and Development of the Waw-Consecutive: Northwest Semitic Evidence from Ugarit to Qumran (HSS, 39; Atlanta: Scholars, 1991), 39: “The Pesharim contain no clear cases of unconverted imperfect with waw, but exhibit at least ten cases of converted imperfects”; p.40: “The Pesharim have at least eleven converted perfect forms and no cases of unconverted perfect forms”. It is thus beside the point for Nitzan, Pesher Habakkuk, 114 to remark that only a low proportion of verbs in PHab are converted forms. Given the brief nature of most sections of the pesher, the opportunities for consecution are limited, and the pesher uses the converted forms each time it is appropriate. A similar cause—hardly related to chronology!—helps explain the rare occurrence of converted forms in Hebrew inscriptions from the monarchic period, see Young, “LBH and Inscriptions,” 294–95. Furthermore, as Nitzan notes, the proportion of converted verbs in PHab is very similar to that in biblical Habakkuk. For the lack of continuity between the Qumran and LBH verbal systems see Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.277–78.

[152] I. Young, “‛Am Construed as Singular and Plural in Hebrew Biblical Texts: Diachronic and Textual Perspectives,” ZAH 12 (1999), 48–82.

[153] Young, “‛Am,” 58 n.41.

[154] Horgan, Pesharim, 28; Horgan, “Habakkuk Pesher,” 165 n.29 reads מחשבתם‎ in PHab 3:5 as a defectively written plural “their plans”. Similarly Horgan, Pesharim, 29 takes ובבהמתם‎ in 3:10 as a defectively written plural. If correct, this may belong under “pluralisation”, see 3.2.5.

[157] D. M. Carr, Writing on the Tablet of the Heart Origins of Scripture and Literature (Oxford: Oxford University, 2005).

[158] Carr, Writing, 168, 253–254.

[159] J. Trebolle, “A ‘Canon Within a Canon’: Two Series of Old Testament Books Differently Transmitted, Interpreted and Authorized,” RevQ 19 (2000), 383–99. Carr, Writing, 155 points out the peripheral role of the core LBH books of Chronicles, Esther, Ezra and Nehemiah.

[160] Carr, Writing, 16, 230 emphasises the ability of Second Temple period Jewish authors to produce various registers of BH. Cf. Qimron, “Observations,” 353–54: “The [Qumran] sectarians studied the Bible day and night so that its phraseology became a living component of their own language.”

[161] Note, however, that משל‎ in 8:9 is based on a play on words with משל‎ “proverb, taunt song”. See Brownlee, Midrash Pesher, 133, 143–44.

[162] See e.g. Brownlee, Midrash Pesher, 54; Nitzan, Pesher Habakkuk, 152; Horgan, “Habakkuk Pesher,” 160 n.20, who reconstruct the lemma here as “Look, O traitors (בגדים‎)” with the LXX, rather than the MT “Look among the nations (בגוים‎)”.

[163] Brownlee, Midrash Pesher, 68 understands this to go with the previous clause, hence “from afar they will swoop as an eagle”. Horgan, “Habakkuk Pesher,” 165 takes it with the preceding, hence “their riders spread out from a distance”.

[164] Elliger, Habakuk-Kommentar, 80–86, and more generally pp.78–117.

[165] This may qualify the suggestions made above in sections 3.2.3, 3.2.4 and 3.2.5 that prominent LBH features of PHab were picked up under the influence of the style of biblical Habakkuk. Nevertheless, since these are not common features of EBH and are shared specifically by a particular text and the commentary on that text, the hypothesis is still worth considering. It is still obvious that in general there is a relationship between the language of the lemma and its pesher.

[166] See Nitzan, Pesher Habakkuk, 103–22; and on Qumran in general, Qimron, Hebrew.

[167] Horgan, “Habakkuk Pesher,” 162 n.31 says “It is unclear whether the א‎ at the end of this line is part of an anomalous 3rd pl. form, or whether it is another sign at the end of the line”.

[168] On the “Qumran practice” see E. Tov, “The Orthography and Language of the Hebrew Scrolls Found at Qumran and the Origin of these Scrolls,” Textus 13 (1986), 31–57; E. Tov, “Further Evidence for the Existence of a Qumran Scribal School,” The Dead Sea Scrolls Fifty Years After their Discovery, 199–216; E.Tov, Textual Criticism of the Hebrew Bible (Minneapolis: Van Gorcum, 20012), 107–11; E. Tov, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (Leiden/ Boston: Brill, 2004), 261–73, 277–88.

[169] Brownlee, Midrash Pesher, 42 suggests that since the pronoun elsewhere in PHab is spelled short, this form should interpreted as a noun equivalent to biblical הֹוָה‎ “ruin”. Most scholars have continued to read the pronoun here. Horgan, Pesharim, 23 points out that nowhere else in Qumran is the noun which Brownlee suggests spelled with aleph. Horgan, Pesharim, 23; Horgan, “Habakkuk Pesher,” 160 in fact read היאה‎ “she” here. It is common for Qumran documents to exhibit a mixture of long and short forms of pronouns in the same text. This may indicate that this form too should be classified under “orthography and pronunciation”.

[170] Brownlee, Midrash Pesher, 64–65; and Nitzan, Pesher Habakkuk, 157 mention an alternative derivation of לכות‎ from כתת‎ not נכה‎, which might remove this example from this category. However, this alternative reading is generally rejected.

[171] See Rendsburg, Diglossia, 95–102. All of the examples he discusses are in EBH texts.

[172] For śin/samekh in EBH see Young, Diversity, 190–91.

[173] Four LBH features, see Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.266–75.

[174] Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.132–36, 39, 271–75.

[175] Both the Community Rule and the War Scroll samples in Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.134, 273 have an accumulation of nine LBH features, the same as the Arad Ostraca.

[176] For the possibility of pre-exilic LBH, see Young, Rezetko and Ehrensvärd, Linguistic Dating, 2.89–91. Our focus here, however, is on the Second Temple period.

[177] The Temple Scroll sample presented in Young, Rezetko and Ehrensvärd, Linguistic Dating, 1.133, 273 has the highest accumulation of LBH features of any Qumran text yet studied. However, its accumulation of 13 LBH features is still significantly lower than the lowest core LBH sample presented in Table 1, above.

[178] Kutscher, Isaiah Scroll.

[179] It will be clear to the reader that in this approach the labels “EBH” and “LBH” have been emptied of their original chronological significance. EBH is a style, which may have developed over time. However, any chronological development of EBH should be seen as parallel to (or at best slightly influenced by) the separate LBH style.