In this revised dissertation (T. Lim and P. Hayman, supervisors), D. Shepherd explores the relationship of the so-called “targum” of Job from Qumran (henceforth 11Q10) with its counterparts in the Rabbinic (RtgJob) and Syriac (Peshitta or P-Job) traditions through a “systematic, synoptic three-way comparison” (p. 21). He also assesses the recent suggestion that 11Q10 should be simply identified as a translation of its Hebrew source rather than a “targum” (p. 22). Taking methodological insights from the previous work of H. M. Szpek on the Peshitta to Job, he uses the categories of syntax and style to test the formal representation of the Hebrew text (as preserved in the Masoretic tradition) in the sections of Job which have survived in all three witnesses. The research focuses on three criteria: omissions, transpositions, and the treatment of the waw conjunction. In spite of the limitation of both the sample and the criteria to study it, Shepherd is confident he can achieve reliable conclusions.

The book is divided in three parts, one for each criteria. The omitted elements (chap. 1–5) are usually small functional words or morphemes, rather than large portions of text. Only one omission is shared by all three versions (37:13), and a second one by 11Q10 and P-Job (36:7), both being related to a “linguistic-stylistic adaptation of the source text” (p. 40). Shepherd studies 28 instances where 11Q10 offers indications for a unique omissions. Sometimes (as in 19:14) the space in a lacuna is too short for “a word-to-word rendering of the source text”; the omission is not certain, however, since in several cases (as in 21:24) the word-order is unwarranted and could have been altered. When the omission is certain, it may have occurred for several reasons: the translator may have struggled to understand a difficult text (29:7); he may have done “preceding modification” that resulted in a minus a little further on in the text (34:13); he may have sacrificed an element “for the sake of idiomatic fluency” in the target language (33:24), or he may have omitted an element which he perceived “as either not required [...] or not permitted” (38:26–p. 73). The better preserved Syriac text displays 30 clear instances of omission with similar causes, the most frequent being a perceived redundancy in the source text (36:9–p. 110). Only two instances of minuses occur in the section of RtgJob wich overlaps 11Q10: the first (31:28) is apparently motivated by “the concern to avoid ambiguity with respect to the deity” (p. 115), whereas the second (42:5) seems to have been required as the result of a “prior and theologically constrained modification in the context” (p. 116).

In the second part (chap. 6–10), Shepherd studies 21 cases in the preserved portions of 11Q10 where a transposition may have occurred; he finds 29 in P-Job and only two in RtgJob for this section of the book. In 11 additional instances, 11Q10 and P-Job deviate from the Hebrew word order, three times in the same way. There is no case of a transposition shared by all three witnesses. In a small number of instances of unique (11Q10 on 40:5) or shared instances (22:17), the transposition may have been found by the translator in his Vorlage. Most of them, however, seem to be the result of the translator’s intervention stimulated by a previous modification (11Q10 on 36:28), a difficult or potentially ambiguous text (11Q10 and P-Job on 22:7 and 36:7), a willingness to harmonize the word order of parallel segments of texts (11Q10 and P-Job on 31:15), a linguistic or stylistic constraint or preference (11Q10 on 21:6; P-Job on 32:15).

The third part (chap. 11) is devoted to the treatment of the waw conjunction in a series of selected examples. Unique omissions, additions or substitutions of the waw conjunction are far more frequent in 11Q10 and P-Job than in RtgJob (p. 245). Unsurprisingly, 11Q10 and P-Job also share “a significant number” of modifications, but almost none with RtgJob (p. 256). 11Q10 and P-Job are apparently motivated by their willingness “to produce a more idiomatic rendering of their Hebrew source”, whereas RtgJob “is particularly scrupulous in its representation” of it (p. 257).

From his investigation, Shepherd infers that the translators’ adaptation of a Masoretic type of Vorlage better accounts for most modifications found in Qumran and Syriac versions of Job. RtgJob is much more literal in its translation, but interpolates far more “midrashic material”. This leads Shepherd to conclude that “in terms of translation approach” Qumran and Syriac versions “are clearly and unequivocally independent of the targumic tradition” (284); this argument would support the view that 11Q10 “is no more deserving the title ‘targum’ than its counterpart in the Syriac translation tradition” (p. 286). A list of 188 works cited and five indices provide useful tools to retrieve specific information from the book.

Shepherd’s “reconsideration” of 11Q10 and his attempt to locate it in the larger context of Aramaic translations of Job is a significant contribution to the field. This study is well informed and the general argument is clearly exposed, in spite of a sparse use of subtitles in several chapters. Shepherd’s precise examination of each case and his balanced judgment usually lead to convincing conclusions. Having restricted his definition of a targum to the characteristics found in RtgJob, Shepherd seems correct in suggesting that 11Q10 should not be labelled as such. But the question of what exactly defines a targum certainly deserves much more debate.