. . .

Summary

The linguistic evidence discussed here accords with the conclusions of Dershowitz’s literary-critical analysis of the Shapira Deuteronomy manuscripts in this volume and corroborates a monarchic date for V.

According to the principle of consilience, as recently laid out by Hendel and Joosten, a given claim regarding the date of a text is validated by the convergence of diverse lines of evidence. They write, “Consilience in our scholarly models is the best we can achieve, and it is enough,” commenting on their own argument that “the linguistic and historical inferences are consilient, indicating the correctness of the theory.” Similarly, the linguistic and literary data in the case of V are convergent, attesting to the likely correctness of the hypothesis that it antedates the biblical Deuteronomy.

Furthermore, nothing in the language of the Valediction of Moses is suggestive of either forgery or Hellenistic composition. On the contrary, the language of V is consistent with pre-exilic Hebrew, especially as attested directly in the epigraphic corpus. Moreover, the text includes no obvious late features or Aramaisms, which is especially notable, considering how difficult it would have been for anyone educated in Hebrew in the Hellenistic period (or the nineteenth century) to do so. Since the 1960s, research on Late Biblical Hebrew has exposed a large number of lexical, orthographic, and morphosyntactic features that first appear in Persian period texts, and our understanding of the grammar of LBH and post-biblical Hebrew has likewise expanded and changed. These post-exilic features and their relevance for dating biblical texts were largely unknown to scholars in the nineteenth century, yet V contains none of them. This weighs strongly against the possibility of a forgery.

The orthography of V is also significant. Almost no Hellenistic period manuscripts are orthographically conservative, and post-biblical texts consistently present fuller and more liberal spelling practices than their MT parallels. V, on the other hand, presents an orthography that is considerably more conservative and defective than MT, and it is similar, with only minor variations, to the monarchic epigraphic material. This too constitutes a strong counterargument to claims of a Hellenistic composition.

Despite many similarities outlined above, the Hebrew of the Valediction of Moses nevertheless deviates from that of the Masoretic Text in various ways and appears to reflect a dialect other than standard CBH. This is to be expected, especially if the Shapira manuscripts are pre-exilic artifacts, which would leave little opportunity for V to have undergone the sort of linguistic updating that is so prevalent in the texts of later Hellenistic, let alone Masoretic, biblical manuscripts. When the apparent linguistic anomalies in V correspond to attested ancient usage– particularly when this ancient usage was not known to nineteenth-century scholars – it militates against forgery. Furthermore, we should be careful before concluding that a feature is anachronistic just because it is otherwise attested only in later texts. To illustrate, Arad 1:4 and 5:2, both of which date to the First Temple period, contain the noun עוד “surplus.” This noun is never found in the Hebrew Bible, even in LBH texts, but it is attested later in Mishnaic Hebrew (m. Ter. 4:7). Surely our conclusion should not be – and indeed is not – that the Arad ostraca are modern forgeries. Rather, these ancient inscriptions add a new piece of information to the unfolding story of Hebrew.

A similar cautionary lesson may be learned from previously unverified texts that have stood the test of time, such as the Dead Sea Scrolls. In 1953, Solomon Zeitlin argued on linguistic grounds that a Bar Kokhba letter could not be ancient:

The letter begins with the word משמעון “from Simon.” This opening word of address proves beyond any shadow of doubt that this letter was neither written by Simon the leader of the revolt against the Romans, nor by any one of that period. We have a considerable number of letters which have come down to us from antiquity [...] None of them has the prefixal mem to indicate “from.” […] The letter mem prefixed to the author’s name came into use in the Middle Ages. Hence we may say with certainty that the word משמעון “from Simon” shows that this letter was written in the Middle Ages.

Zeitlin may well have been correct that prefixed mems were unattested in the relevant period, but the conclusion he drew from this fact was dramatically wrong, as we now know. Given the severe paucity of data regarding early Hebrew, countless features that were alive and well at the time – many of which are attested in later Hebrew chronolects – are unknown to us due to accidents of history. Occasionally, we are lucky enough to make discoveries that, if not incautiously disregarded, fill in some of the gaps in our knowledge.

The Valediction of Moses – being neither a late forgery nor a Hellenistic composition, but rather a pre-biblical book – is of immense value for establishing the early history of the Hebrew language. The work we have done here on V’s linguistic character is preliminary; we expect that future linguistic studies will shed much light on the both the Valediction of Moses and Classical Biblical Hebrew.