Karen H. Jobes. The Alpha-Text of Esther: Its Character and Relationship to the Masoretic Text. Society of Biblical Literature Dissertation Series, no. 153. Atlanta: Scholars Press, 1996. ISBN: 0-7885-0202-6 (cloth); 0-7885-0203-4 (paper). Pp. xv+256+appendices and indices (unpaginated, approx. 250 pp.). US $44.95/$29.95.

1. This volume represents Jobes' dissertation completed under the supervision of Moisés Silva at Westminister Theological Seminary in Philadelphia. It is a thorough investigation of the Alpha-Text (AT) of Esther as it compares to both the Masoretic Text (MT) and the so-called Septuagint (LXX) version.

2. In the first chapter Jobes employs the syntactical criteria developed by Raymond Martin to analyze both the AT and LXX (Martin 1974). Jobes provides a brief definition of the seventeen criteria and includes the results of her investigation for each chapter and the six additions in an appendix. For AT (minus the additions): 9 criteria indicate translation; 2 composition; 6 are inconclusive. For LXX (minus the additions): 9 criteria favor translation; 5 composition; 3 are inconclusive. The results indicate that both are translations. Jobes also argues that it is unlikely that AT would retain so many features indicating its character as a translation if it also were a recension of the LXX. However, since Jobes later argues that Martin's criteria for distinguishing translation from composition Greek must be employed with caution, there would seem to be little basis to make such a judgment. Jobes also concludes that the six additions are not long enough to provide enough data to assess their origin, though addition E appears to be composed in Greek.

3. In addition to her analysis of the syntactical criteria in the AT and LXX, Jobes offers valuable criticism of Martin's syntactical criteria. The criteria themselves are based on the different frequencies of occurrence of particular syntactic features (e.g., the frequency of <grc>de/</grc>) in books that are known to be translations as opposed to those that are known to be original compositions in Greek. However, the frequencies of occurrence provided by Martin derive solely from the control texts that he has selected. Since the syntactical features selected by Martin appear with different frequencies among all texts, whether translations or compositions, the values he has chosen to represent the range of usage of a syntactical feature in a translation versus a composition are arbitrary. It is questionable whether the criteria can distinguish between composition and translation Greek or just different styles of Greek. Jobes offers two additional criticisms of Martin's methodology. First, she demonstrates that a large number of occurrences are required for some features because the difference between composition Greek and translation Greek is very small. Second, Jobes notes that it is difficult to employ Martin's methodology because there are no means to assess which of the criteria are the most important for distinguishing between composition and translation Greek.

4. Jobes is not content merely to point out the weaknesses of Martin's work. For each of the last two criticisms she offers constructive solutions. For the former she provides an equation to determine the minimum number of occurrences of a criterion needed to assure that it can be applied to the text in question. Second, in order to compare the results of each criterion directly, Jobes employs descriptive statistics to normalize the ratios. However, even with these welcome improvements Jobes maintains that Martin's criteria should be used only with caution.

5. The second chapter is devoted to determining the similarities between AT and MT. Jobes appropriately notes that one must distinguish between formal agreement and semantic agreement, so she proposes assigning values on a scale of 1-5 to describe how closely the texts agree: 1 indicates that the Greek is basically the same as the Hebrew and 5 indicates it is completely different. The depth of Jobes's analysis is assured by the fact that she applies this scale of values to five separate criteria. She borrows her categories from Tov's criteria of formal equivalence (Tov 1981: 54-60) and examines lexical consistency, equivalence of morphemes, word order, correspondence and linguistic adequacy. In order to facilitate this process Jobes divides the texts of AT, MT, and LXX into small syntactic units (2814), and she includes the parallel texts in an appendix. As an example of her rating system Jobes states on p. 60:

For instance, in unit 1:11e [i.e., Est 1:11], the Hebrew reads <heb>b.:keter</heb> and is translated by the AT with <grc>e)n tw=| diadh/mati</grc>. The Greek AT unit includes the definite article <grc>tw=|</grc> where there is no article in the Hebrew unit, but otherwise the Greek unit is equivalent. . . . This comparison was assigned 1's in the categories of consistency, equivalence, word order and linguistic adequacy. The category of correspondence . . . was assigned a value of 2 because the Greek has a definite article where none is found in the Hebrew.
The use of a 2 to distinguish differences on the formal level is reasonable; however, Jobes never offers examples to illustrate what distinguishes ratings 3 through 5.

6. Jobes concludes that the overall agreement between AT and MT is low (28% semantic agreement; 18% formal agreement), but this is due primarily to the pluses and minuses in the AT. When one considers only the units that have corresponding Hebrew and Greek (590 of 1803 or 33%), then the extent of agreement is much higher (84% semantic agreement; 54% formal agreement). These statistics compare favorably with the LXX (88% semantic agreement; 64% formal agreement). Jobes includes tables and graphs that provide a chapter-by-chapter breakdown of the agreement between AT and MT for each of the five criteria as well as the overall totals.

7. The statistics provided in this chapter represent an enormous amount of work and are extremely valuable for research into the texts of AT, LXX, and MT. However, there are some details that detract from her presentation. First, the appendix in which she provides the alignment of the texts is welcome indeed, but she does not give the values she assigned to each of her criteria for the syntactic units. This would have been very helpful in understanding how she computed her statistics and for evaluating her argument that in only 90 of the 590 cases where the Hebrew and Greek texts correspond could one argue for a variant Vorlage (p. 73). Second, the statistics for lexical consistency were based on an analysis of twenty verbs and sixteen nouns that occur in the Hebrew. Though the LXX does exhibit a trend to be more lexically diverse than AT for these words, the sample is small and limited for the most part to common words. An analysis of rarer words might have offered different results. Furthermore, she does not provide references even for the words that she does analyze (pp. 86-94).

8. The third chapter examines the differences between AT and MT. Jobes catalogues the various types of minuses, pluses, substitutions, and textual differences that occur in AT, and she also provides a listing of references for each type of difference. She then proceeds to tackle the question of the redactional history of the book. Her discussion focuses on chapters 8-10. Clines (1984) and Fox (1991) have argued that AT has no connection with MT for chapters 8-10, but Jobes argues on the basis of similarities of content and textual agreements that AT "translated a Hebrew text very similar to the MT in chapters 8-10 but was subsequently revised almost beyond recognition" (p. 134). The material omitted from AT exhibits a translation technique, found also in ch. 2, in which AT minimizes Esther's role and shows less interest in the origin and celebration of Purim (see pp. 131-137).

9. Chapter four compares AT to LXX. Jobes again provides tables and graphs to depict the correspondence between AT and LXX and agrees with other researchers that AT is not a recension of LXX. Her detailed analysis (pp. 162-193) of the six additions shows a considerable degree of difference between AT and LXX (semantic agreement ranges from 47-76%; formal agreement from 33-51%). Jobes concludes that AT preserves the older form of additions A, B, C, E and F, but there was extensive secondary redaction to both AT and the later LXX.

10. In chapter five Jobes interacts more directly with the research of Tov (1982), Clines (1984), and Fox (1991). Contra Tov, who argues that AT is a recension of LXX that was corrected toward a Hebrew midrashic rewriting of Esther, Jobes concludes that the agreements between AT and LXX are relatively few. She argues that there are many more places where one would expect to find agreement where none exists. Therefore, it is more likely that the agreements represent later corruption during textual transmission. Tov also argues--based partially on Martin's conclusion that additions A, C, D, and F are translations of Semitic sources--that the text of the canonical sections and the additions of AT are one unit. However, Jobes' more thorough analysis of the syntax in chapter 1 concludes that the composition of these additions cannot be determined, while B and E were probably written in Greek.

11. Clines argues that AT was a translation of a Vorlage different from MT that ended at 8:17, while Fox contends that the original ending of the Vorlage was Est 7:38. Jobes' research cited in paragraph 8 refutes these conclusions.

12. Following a chapter that summarizes her conclusions is an excursus in which Jobes compares AT to the Old Greek of Daniel. The reasons for this comparison range from the frivolous (the Chigi manuscript happens to preserve the minority texts of Esther and Daniel; both texts have additions) to superficial similarities (Daniel 4-6 omits material from MT). I was also disappointed by the fact that Jobes would venture to make this comparison without acknowledging the fact that Walde has already argued that the translators of 1 Esdras and Daniel OG were the same (Walde 1913). Regardless of the reasons, Jobes attempts to sketch out some shared features of the two versions. She begins with the conclusion of Wenthe that OG Daniel exhibits greater lexical diversity than Th (Wenthe 1991). Of the twenty-four words examined by Wenthe, nine are found in Esther, and both AT and LXX Esther are more lexically diverse than OG Daniel. In fact, "A comparison of the translation equivalents in OG Daniel to those in the Greek versions of Esther shows no apparent pattern in common" (p. 239). In other words, this analysis produces no evidence of a relationship between AT and OG Daniel. Jobes continues this line of investigation by accepting Wenthe's assumption that OG Daniel was produced by one translator (I do not concur; cf. McLay 1996: 214-216) and his analysis of the translation pattern of <heb>mal:k.Ut</heb> and <heb>melek:</heb>. Since these terms are often translated differently in chs. 4-6, Wenthe argues that the Vorlage was not the same as MT. Jobes finds similar results for chapters 8-10 of AT, but she does recognize that this is not evidence of any connection between AT and OG Daniel. So, this line of argument essentially leads nowhere (pp. 239-242). She then investigates fifteen other possible connections between AT and OG Daniel. There are two that are noteworthy: <grc>doch\n mega/lhn</grc> appears in both the proem of chapter 5 in OG Daniel and in Est 1:9 AT; both employ <grc>po/lis</grc> to translate <heb>b.iyrfh</heb>.

13. In this same excursus Jobes also notes 5 differences between AT and OG Daniel. In one case she cites information provided by S. Jeansonne (Jeansonne 1988:118) about the argument of F. F. Bruce, who states that OG Daniel indulges in a theological rendering in 9:6 (p. 247). This is curious because Jeansonne actually argues against the point that Bruce attempted to make! On the whole, I found this excursus lacking the depth of analysis that characterizes the remainder of the book. It could well have been omitted.

14. One further criticism may be noted: I would have preferred detailed interaction with, as well as references to, the secondary literature in footnotes. Notwithstanding the relatively minor criticisms, I hope that this review has brought to light the importance of this volume for anyone working in the book of Esther, as well as its interest for those working in the Septuagint.

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


Clines, David J. A. 1984. The Esther Scroll: the Story of the Story. Sheffield: JSOT.

Fox, Michael V. 1991. The Redaction of the Books of Esther. Society of Biblical Literature Monograph Series, no. 40. Atlanta: Scholars Press.

S. Pace Jeansonne. The Old Greek Translation of Daniel 7-12. Catholic Biblical Quarterly Monograph Series, no. 19. Washington: Catholic Biblical Association.

Martin, Raymond A. 1974. Syntactical Evidences of Semitic Sources in Greek Documents. Septuagint and Cognate Studies, no. 3. Missoula, MT: Scholars Press.

McLay, Tim 1996. The OG and Th Versions of Daniel. Society of Biblical Literature Septuagint and Cognate Studies, no. 43. Atlanta: Scholars Press.

Tov, Emanuel 1981. The Text-Critical Use of the Septuagint in Biblical Research. Jerusalem Biblical Studies, no. 3. Jerusalem: Simor.

Tov, Emanuel 1982. "The 'Lucianic' Text of the Canonical and the Apocryphal Sections of Esther: A Rewritten Biblical Book." Textus 10: 1-25.

Walde, Bernhard 1913. Die Esdrasbücher der Septuaginta: Ihr gegenseitiges Verhältnis untersucht. Biblische Studien, vol. 18, pt. 4. Freiburg im Breisgau: Herder.

Wenthe, D. O. 1991. "The Old Greek Translation of Daniel 1-6." Ph.D. diss., University of Notre Dame.

Tim McLay
Halifax, NS, Canada