Gentry, Peter John. Review of The Asterisked Materials in the Greek Job. SBLSCS 38. Atlanta: Scholars Press, 1995. ISBN: 0-7885-0093-7. Pp. xxxviii+559. US $49.95 (cloth), $33.95 (paper).
1. One of the most problematic areas in Septuagintal research is the remarkably comprehensive text-critical work executed by Origen, especially the six-columned description found in his Hexapla. This massive compilation of textual variants was of course never transmitted in its entirety; however, the fifth column, which contained the Old Greek as constructed by Origen, had a major impact on later textual traditions.
2. A major problem with this detailed work, as is well known, is Origen's use of appropriate symbols to describe the differences which he encountered in the texts he used. The intentions of Origen were naturally clear: to distinguish between those passages that were extant in the Hebrew but not in the Greek, which he marked with an asterisk, and the passages that appeared in the Greek without any counterparts in the Hebrew, which he marked by means of an obelus. The problem, however, remains; during the later historical vicissitudes of the Hexapla, these sigla unfortunately became intermingled, with the result that it became impossible to trust any diacritical mark which referred to the monumental work by Origen. His work has a bearing on the whole of the transmission history of the LXX, as it contaminated many later manuscripts. Subsequent scribes inaccurately rendered these sigla or in many instances simply omitted them, and the result was textual confusion. Two positive developments are the preservation of parts of Origen's text-critical work by Eusebius of Caesarea and the translation of the fifth column into Syriac. This Syro-hexapla can be used as basic comparative material, even though it has problems of its own. Fortunately these problems are now being dealt with systematically by the newly formed Syro-hexaplaric project, an undertaking of the Peshitta Institute in Leiden.
3. The various textual problems I mentioned above are nowhere more evident than in the Greek book of Job. And as to be expected of a student of the "Toronto school", Peter Gentry has formulated a novel research proposal, focussing upon the asterisked materials in the textual tradition of the Greek Job with the greatest detail. The questions he intends to address are: why are specific lines asterisked? what is the origin of this material? what is its nature and its textual affiliation? Another issue he deals with is the question of the sources utilised by Origen in respect to Job.
4. Gentry has done his ground work excellently. In true text-critical fashion he approaches the question of the exact limits of the asterisked materials from the perspective of both external evidence--i.e., the manuscript evidence--and internal evidence--concentrating on the work of the translator(s)/revisor(s). One way in which he examines the internal evidence is by contrasting the translation technique of the asterisked materials with that found in the OG.
5. The asterisked materials are assembled from two primary sources: first, lines described by Origen which found their way into the so-called ecclesiastical text, and second, other materials attributed to Theodotion in the church fathers and the Catena manuscripts. The main source used by Gentry is of course Ziegler's critical edition of Job in the Göttingen series. Gentry has a precisely demarcated methodological approach towards the asterisked materials which entails first the identification of these materials.
6. Although the author uses the work by Ziegler (1982) as a point of departure, he nevertheless rightly approaches this publication critically. Gentry deals with the materials in a meticulous way. He works through each example of asterisked data quoted by Ziegler. In some instances he discovers discrepancies between Ziegler's introduction and his text, whereupon he offers appropriate corrections. One prominent feature that he discusses is the difference of stichometry between Rahlfs and Ziegler. This applies also to the lines that were asterisked by Ziegler but considered OG by Rahlfs (1935). One case in point is Job 9:3b, <grc>ou) mh\ u(pakou/sh| au)tw|=</grc>, which is also taken as OG by Pietersma contrary to Ziegler. The main argument used by Pietersma (1985:305-311) and followed by Gentry (p. 19) is that the manuscript evidence (from the C tradition) is extremely weak. Gentry is moreover critical of Ziegler's statement that the Old Latin omits 3b. Gentry sees the Old Latin version at this point as a "kind of conflation of 3b and 3c" (p. 21).
7. What impressed me is the nuanced combination of arguments utilised by Gentry when making a decision on whether a certain line should be deemed to be asterisked or Septuagintal. He combines the external and internal criteria which I mentioned above. For example, he regards the textual support for taking Job 25:6b (<grc>kai\ ui(o\s a)nqrw/pou skw/lhc</grc>) as asterisked to be weak. However, when looking at the issue from the perspective of translation technique, he comes to a different conclusion. At this point the meticulous manner in which he approaches the subject matter becomes evident yet again. He studies the way each Hebrew lexical item is rendered by the Greek translator and in the revisor's text (R). In the final analysis he then comes to the conclusion that Ziegler made a sound decision by taking 25:6b as asterisked.
8. An interesting issue addressed by Gentry is the question of textual material containing features of the hexaplaric material pre-dating the Hexapla. This phenomenon occurs in the Greek version of Proverbs too (cf. the discussion in Cook 1997), so it may be helpful to consider the evidence related to Proverbs alongside that of Job. The occurrence of what appear to be hexaplaric readings in older material is called a "hebraisierende Tendenz" by Zuntz (1956:165). Not much has been published on the question of hexaplaric influence in LXX Proverbs (Cook 1996), at least not before the discovery of the Antinoopolis papyri in 1913-14. One of these mss, 928, contains a number of fragments from Proverbs that were published by C. H. Roberts (1950:18).
9. Another relatively unknown textual witness, the Prophetologion used in the eastern church for liturgical purposes, was studied by Zuntz (1956:125). This corpus, consisting of about 160 mss dating between the 9th and the 16th centuries, had not been utilised sufficiently by scholars before Zuntz. The mss in the corpus are based on a work originally created in the 8th century and since then transmitted by church officials. According to Zuntz, the Prophetologion exhibits a surprisingly stable text which significantly corresponds to the Antinoopolis fragments mentioned above.
10. Zuntz does a painstaking comparison between these textual corpora. He finds that the papyrus fragments of Proverbs contain 21 readings that are practically unique in the LXX. Significantly, they represent a "hebraisierende Tendenz", which I have mentioned already. They also exhibit a definite relationship with a specific group of textual witnesses which Zuntz (1956:165) designates as g. To this group belong the Prophetologion and the late minuscules 336, 443s, and 252, as well as the majuscule V. Apparently the two latter witnesses do not agree to the same extent as the rest of the group (Zuntz 1956:166). This group should be seen as "hexaplaric", for V and 252 are after all prominent Origenic mss.
11. The dating of the Antinoopolis papyri is naturally of crucial importance in this regard. For if it should be placed in the third and not the fourth century, then it is surprising that the same tendency towards the Hebrew that is so typical of Origen is already in evidence so early. Zuntz (1956:174) is conspicuously careful in his treatment of this issue. For example, he mentions the possibility that the papyrus could date from the latter period of Origen's life and that it is possible that it indeed represents a text that was actually reviewed by Origen himself. He nevertheless regards this as improbable. He also mentions the views expressed by J. Ziegler that, as far as the book of Isaiah is concerned, the hexaplaric witnesses Q and Syro-hexapla are not the result of Origen's work, but the work of students of Eusebius of Caesarea (Ziegler 1983:52); another possibility is that Origen could have found additions tending towards Massoretic readings in his Hebrew Vorlage (Ziegler 1983:62). In the final analysis he sees 928 as a real Septuagint text and not a representative of some other translation, for example, Theodotion (Zuntz 1956:175). (Zuntz [1956:127] mentions a letter in which Roberts confirms his view that the third century date is preferable.)
12. The implications of Zuntz's study are important. The notion that adaptation towards the Hebrew took place earlier than previously thought is, of course, not new. The <grc>kai/ge</grc> recension found by Barthélemy (1963) represents earlier evidence of this phenomenon. The discovery of a similar development in the context of the book of Proverbs is nevertheless significant for the unravelling of its complicated textual history. It could have the implication that some of the double translations which occur in LXX Proverbs are in fact pre-hexaplaric phenomena.
13. The hypothesis by Zuntz has not been universally accepted, nor has it been discussed exhaustively. P. Katz (1957:77-84), among others, has expressed some reservations about this theory. He nevertheless does not reject the fact of Hebraising recensional work in the Septuagint before the existence of the Hexapla. He in fact views Zuntz's viewpoint as an independent testimony to this recension. His main criticism is aimed at the endeavours by Zuntz to relativise the Hebraising revision that Origen actually executed. Katz (1957:80) has a very clear view of the unique situation in the books of Proverbs and Job. Talking about the deliberate use of Greek mss by Origen that were the closest to MT, he makes the following statement: "und bei so willkürlich übersetzten Büchern wie Prov und Hiob musste dieser Rückgriff so gut wie den ganzen Umfang der Bücher umfassen".
14. It is therefore evident that the books of Proverbs and Job are unique in this regard. Perhaps a study of this phenomenon in Proverbs and Job can assist us in tracing the textual histories of both these LXX texts more accurately. There are significant correspondences, as well as differences, between these two Greek versions (Cook 1997a).
15. The meticulous work is simply extended by the author when he endeavours to characterise the materials. As is the case with the identification of the materials, his methodology is to the point and applied in a systematic and quantitative manner. He accounts for the structural differences between the source and target languages. The literal approach of R towards the Hebrew fortunately assists the researcher in determining the translation technique that he follows.
16. Gentry compares the asterisked Greek and the Hebrew in detail as far as various parts of speech are concerned. Various categories of nouns are discussed, whereafter follow other parts of speech such as pronouns, adjectives, adverbs, etc. The evidence speaks for itself: and far as the noun goes, in his view it "points to an extremely high level of correspondence formally between the Greek of R and the Hebrew of MT in the rendering of nouns" (p. 99).
17. Much space and time are spent on verbal forms, for again Gentry deals with the total picture, addressing the aspect/tense, number and person, and voice of all verbal forms. He systematically deals with finite verbs, participles, infinitives, pseudo-verbs, verbs rendered by a different syntagmeme, and finally verbs that are used in R for nouns in MT. Moreover, to paint the full picture he also deals with the way R renders Hebrew verbs from a lexical perspective. Again the hand of Albert Pietersma can be observed in the methodology followed by Gentry. The textual evidence is categorised into four levels of variety: (1) a Hebrew lexeme rendered by one Greek lexeme; (2) more than one Hebrew lexeme rendered by one Greek lexeme; (3) a Hebrew lexeme rendered by more than one Greek lexeme; and (4) more than one Hebrew lexeme rendered by more than one Greek lexeme. In order to further systematise categories, 1 and 2 are called stereotype equivalents, the third category regular equivalents, and the fourth non-standard equivalents.
18. This exhaustive analysis of structural and lexical perspectives puts Gentry in a position to draw significant conclusions as to the details of the translator of the asterisked materials in Job. In his own words: "While R is a literal translator, he is attentive to context and differences of social situation and also provides stylistic variations to avoid using the same verb in contiguous contexts" (p. 299).
19. Gentry subsequently discusses the phenomenon of transliterations, offering solutions varying from technical terms to guesswork and ordinary confusion of Hebrew consonants by the translator. After an exhaustive characterisation of nearly 300 pages, the author finally concludes this section with a discussion of the treatment of word order by R. The fact that according to him only four out of a total of 434 stichs in the Hebrew are different from the word order of the Hebrew is a clear indication of the literal manner in which the translator approached his subject matter!
20. After completing the characterisation of R, Gentry moves onto the question of the placing of the asterisked materials in the history of the transmission of the text. In this regard he makes a number of illuminating observations. First, the OG represents a totally different approach towards translation from R's literal, quantitative procedure. This leads him to compare these texts differently, namely "holistically" or qualitatively rather than quantitatively. He draws a third significant conclusion: "The differences between R and OG so outweigh any similarities that one must conclude that R is an independent translation and not a systematic revision of OG based on certain principles" (p. 386).
21. An important part of Gentry's dissertation is his comparison of R and the so-called <grc>kai/ge</grc> group (pp. 389-493). Again the author approaches the theme exhaustively, taking into account practically all research that has been conducted on this issue. He also includes Aquila and the translator of the Greek Psalms in addition to the translator of the Greek Minor Prophets Scroll from Nahal Hever. As before, he differentiates among various parts of speech, such as nouns and verbs. This puts him in a position to reach refined conclusions as to the application of lexemes by R and Aquila. His final conclusion is that the approach of Aquila represents an extreme refinement of the translation technique developing in R (p. 492). He finds that Aquila is more consistent than R in his application of fixed equivalents and less sensitive to contextual factors.
22. Even though there are some striking similarities between R and Aquila, there are also conspicuous differences. In this regard his final statement in Chapter 3 is significant: "So there is a development in attitude to translation from LXX to R to Aquila which is at least typological if not chronological" (p. 493). I would like to add that the translator of Proverbs also fits into this picture, although I am not yet sure as to where he should be placed. On the one hand, he exhibits an extremely conservative approach towards his subject matter, going to great lengths to avoid possible misunderstanding of given passages. I have concluded (Cook 1994:473) that he is actually more conservative than is the author(s) of the Hebrew version of Proverbs! On the other hand, he is prepared to adapt his parent Hebrew text to the extent that in some instances the impression is left that a Targumist is at work! He thus exhibits a diametrically opposed approach towards the subject matter (his translation technique is just the opposite of "literal" or "quantitative"), but with the same intention. These remarks should be read together with Gentry's conclusion that the diversity within the <grc>kai/ge</grc> group, as well as the precise line of demarcation between <grc>kai/ge</grc> and LXX, has yet to be taken seriously by modern scholarship (p. 496).
23. The dissertation by Gentry does not present spectacular results. It is more an exercise in sound methodology. He has demonstrated that a meticulously executed contrastive analysis combined with a comparative analysis is the approach to follow when endeavouring to analyse deviating textual materials. While reading him, one always gets the impression that he has drawn on sound data upon which to base his arguments. His nuanced critique of the results of some scholars' <grc>kai/ge</grc> research is highly significant in this regard. It comes as no surprise that, after studying the broader picture, in the final analysis he concludes that there is no <grc>kai/ge</grc> recension as such, but that the deliberate adaptations rather point to an attitude towards texts or a tradition (p. 497). From this conclusion it is already evident that he has furthered our knowledge of recensional activity in the book of Job substantially.
24. This "holistic" approach (as Gentry calls it) can only be applauded. Younger scholars and students will do well to follow the approach demonstrated by Gentry, which he evidently learned at the feet of John W. Wevers and Albert Pietersma in Toronto. In the final analysis they must also be congratulated on the work of their student.
25. On a more personal note, I am looking forward to subsequent research by Gentry. At the end of his dissertation he has formulated a number of suggestions for further study which should certainly be followed up. There is a direct relationship between his research and my own, the complicated text of LXX Proverbs, where much work still remains to be done.
© TC: A Journal of Biblical Textual Criticism, 1997.
Barthélemy, Dominique 1963. Les devanciers d'Aquila. VTS 10. Leiden: Brill.
Roberts, C. H. 1950. The Antinoopolis Papyri. Part I. London: Egypt Exploration Society.
Johann Cook Department of Ancient Near Eastern Studies University of Stellenbosch