Contrasting Uses of the Greek bible: Hellenistic-Jewish Literature and the New Testament
Tessa Rajak
The connections between Septuagint and New Testament are not, as they once were, a topic that arouses general excitement. The issue of the relation between Hebrew Bible and New Testament (in the Christian terminology) is perpetually alive, but the more specific questions arising out of the bible in Greek have taken a back seat, and for understandable reasons. By coming at them from a different angle, I hope I can persuade you that they are still crucial, both for Jewish and for New Testament history - if indeed those two histories can at that turning point in time be distinguished.
My title may seem to lack something - Qumran! This isn't ingratitude for the wonderful work of the Orion Centre - not least this excellent conference in which I am delighted to be participating. In fact, Qumran does have a role to play in what I shall be saying. I am concerned, as a cultural historian, with looking at models of relationships between the 'Judaisms' of the period. (Talking about 'Judaisms' in isolation must surely lead on now to a new, synthetic stage of scholarship). The object of this analysis is largely literary corpora; perforce, because this is where our basic evidence lies. My underlying interest come down to questions about the relationship between a text and its users, and the answers, unavoidably, will be suggested largely by the internal evidence of other texts themselves. In the case of Qumran and NT, we can more easily make inferences about the broader communities to which they are addressed, because the recipients figure largely in the writings and indeed are directly addressed, than we can in the case of the various of Septuagint translations or of the corpus of Hellenistic-Jewish (including apocryphal) literature. Would that we knew something more of the communities in which they were once embedded.
My own recent work has been related to Hellenistic Judaism and to the role of the Greek bible translations within it. It is in seeking to assess this obviously central feature that I increasingly find myself trying to describe and define what the tradition in its entirety amounts to. For such a purpose, it would surely be a mistake to omit the rich and varied corpus of New Testament books from the roster of the output of Hellenistic Judaism. I shall have more to say about that later. In doing this for the purpose of cultural mapping, I am making no claim about the process of the religious 'parting of the ways'. It is valuable, further, to gain a degree of perspective by assessing our evidence for this tradition in terms of the other major kinds of Jewish output of the second temple period. The riches yielded by Qumran, so far from being merely sectarian, are incontrovertibly the starting point for any modern attempt to grasp the range of preoccupations current in second temple Palestine. I need hardly tell that to this audience. But of course we still have a long way to go in piecing together a comprehensive picture, let alone one that might achieve any consensus.
In addition to a broad brush-historical approach, I shall draw briefly in the second half of the paper on work which uses another technique, dependent upon detailed examination of key terms. This appreciation of the value of a lexicographical approach stems from another facet of my work on the Septuagint, the database of political and legal vocabulary which is part of the British Greek-Bible in the Greco-Roman World' project which I co-direct (though I shall not in fact be addressing terms currently part of that database).
If, then, the study of the Hellenistic-Jewish contribution to the form and content of the NT texts, once a major scholarly preoccupation, has in recent decades been sidelined, that is for understandable reasons. One explanation is precisely that it has been overshadowed by interest in other kinds of Jewish influence: Qumran parallels above all (see especially Lim 1997), but also Targum and Midrashic procedures, have come to the fore (Ellis 1991: 54-74). I am not sure what the feeling is about this in New Testament scholarship, but some Septuagint scholars are expressing the need somewhat to adjust the balance. Thus in his handbook on the Greek versions of the bible (2000: 320), Natalio Fernandez Marcos writes of the 'primacy that Qumran studies have enjoyed in connection with the New Testament.' And Timothy McLay, a younger LXX scholar, laments in the introduction to a new publication on our topic (2003: 1): 'a search of several NT introductions written since 1985 reveals that none has a section on the LXX'.
This is more than a question of the excitement of the (relatively) new. The older agenda has begun to look not only tired but problematic. One important factor is that, in so much earlier Christian scholarship, Hellenistic Judaism was seen as crucial to the praeparatio evangelica, a perfect route to eliding the real Judaism of the early church. A certain atmosphere still hangs around the very term 'Hellenistic Judaism' for that very reason (and I wish we had a better term, because this is not even thematically or chronologically helpful). This treatment of an entire culture is a remarkable phenomenon in the history of scholarship (and of anti-Judaism), which deserves its own study. But I cannot resist quoting here from an extreme instance, the 1850 Apology for the Septuagint, written by Edward Grinfield, author of a big collection of Hellenistic parallels to the New Testament (the Scholia Hellenistica in Novum Testamentum) and founder of Oxford University's Grinfield Lectures on the Septuagint. Admittedly, Grinfield was rather far from a scholar of the front rank: his argumentation was weak, his conclusions about Septuagint as the language of Jesus absurd, his triumphalist determinism tiresome, his zeal uncontrolled, and his claim for giving primacy to Septuagint as against Hebrew bible in the Church of England unsuccessful. Not entirely surprisingly, he has dropped out of the modern handbooks. But many theologians in his day and subsequently, would have shared his understanding when he wrote of the Septuagint as the 'viaduct between the 'Old and New Testament' or even as the 'herald of the better covenant', a tool designed 'to change the Hebrew into the Hellenist by gradually enlarging his conceptions'. The 'Hellenistic language of the LXX' was not for Grinfield alone 'the appropriate diction of the holy spirit' (1850: 88; 76; 124). I should perhaps not conceal from you that that Edward Grinfield believed the Septuagint could serve as the perfect instrument to forward the great purpose of the conversion of the Jews, a purpose to which Grinfield seems to have been much attached, and whose endeavours, as it happens, gained a new lease of life in the 1850's both in England and in British activity in the Holy Land. Grinfield had learnt that, understandably, Jews were rejecting versions of the New Testament in Hebrew; instead, they might be taught to read their own bible in what he regarded as Christian Greek.
So much for the past. And yet the neglect, however understandable, is especially regrettable in a period when the broader study of Hellenistic Judaism has expanded in an extraordinarily fruitful way. Septuagint studies still have to make their due contribution to this renaissance. When it comes to methods of assessing the specific impact of LXX on the NT, naturally enough, the study of the very numerous citations has been in the forefront. In many cases, allusions, imprecise verbal echoes, combinations of phrases and adaptations have been considered alongside them, and distinguishing between the different categories is in itself a challenge. Moreover, each book or group of books in the NT displays its particular preferences and throws up its own distinctive and complex problems. Thus in the synoptic Gospels, 18 of the 46 distinct quotations are peculiar to Matthew (including the 11 special quotations preceded by a formula), but only 3 each to Mark and to Luke. 9 out of 12 quotations in John are unique to him. The 23 quotations in Acts fall almost entirely in the speeches. The Pauline Letters have 78 direct quotations, but of these 71 are in Romans, I-II Corinthians and Galatians. The Pastoral Epistles scarcely quote at all. Revelation does not quote directly, but has more Septuagint phraseology than any other book (Swete 1900: 391-2). There are of course multiple explanations for these observed patterns.
To follow the immense literature on this subject is fascinating and there have been major achievements in the field, which I shall not list now. But it is also frustrating, because so much of the work has been directed to specialized scholarly objectives. As Krister Stendahl wrote (1968:40): 'the study of quotations of the OT standing in the NT can have many functions and the way in which it is handled is in part coloured by the purposes the various students had behind their studies.' Hermeneutic concerns are not absent, but, in fact, the principal purposes seem to have been two, and they are often intertwined - as in Stendahl's own influential School of St. Matthew (1954, 1968). From the NT side, it is a matter of studying the sources and origins of the texts and especially, in the case of the Gospels, the synoptic problem and the original languages. On the LXX side, it is a quest to reconstruct the textual history of the Greek versions, by separating out quotations based on the major MSS manuscripts from those closer to the MT or from those which seem to be unique. The complexity of these issues has perhaps stood in the way of widening the horizons of study, and, as in so much LXX research, textual questions have perhaps obtruded more than they need to. McLay's recent contribution (2003) is a case in point, where a welcome invitation to scholars to return to an old subject area, with an initial broad agenda, is in its implementation restricted. Were we to have hopes of being able to say with any degree of confidence that a particular text form in a NT quotation proves its derivation from a Hebrew text rather than any Greek one, or else that a derivation from a particular type of Greek text different from that used elsewhere in the same NT book can be demonstrated beyond a shadow of doubt, the gains could indeed be enormous. But the flux in biblical text forms which the Qumran material has revealed to us, along with the fundamental problems in the Septuagint text, that have arisen from the successive ancient revisions, retranslations and collations which the Old Greek underwent, together contrive to ensure that little progress can be made. Add to that the possibility that quotation could perfectly well be done inaccurately, perhaps from memory, and, then again, the lack of clarity in the distinction between intended exact citations on the one hand and allusions or paraphrases, on the other hand, and the elusive goal is rendered virtually unattainable.
The fact remains, however, that of the NT books, each and every one is suffused with LXX material. According to Swete's still invaluable Septuagint introduction (1900: 26), there are direct citations (under his definition) of every biblical book in the NT, except Ezra, Nehemiah, Esther, Ecclesiastes, Song of Songs and a few Minor Prophets. At the same time, over half of these citations are of Isaiah and Psalms. And some texts, like the Epistle to the Hebrews, are virtually a fabric of quotations. Scholarship must seek to take due account of this phenomenon without getting bogged down. The influence is such that we can justifiably describe the NT writers and their core readers as members of text communities, in the strong sense of these terms. 'The Bible was the touchstone not only of the NT writers' religious teachings but also of their total life and culture' (Ellis 1988: 692).
In just such a sense, the producers of the literature written at Qumran, who were also the keepers of its library, offer another example of this same type of grouping, a 'house of Torah' as the community called itself (CD 7:10). Not only was the biblical corpus for them a 'cherished inheritance' but their overriding preoccupation, as Fishbane well expresses it (1988: 340), was 'with a vast labour of learning and elaboration…a living commitment to the truth and significance of Miqra'.
But how do the worlds of Jewish writing in Greek outside the translated Hebrew bible itself stand in this regard? That they manifested constant respect for biblical literature, through their choice of subject matter and the frequent display of intimate knowledge, is plain. This is the means by which the authors publicly defined themselves as Jewish writers. But plainly too the relationship is a distinctive one in this case. How many of these authors would we speak of as inhabiting 'a house of Torah'?
First, it is necessary to clarify the object of our questioning; as for the difficulties, they need hardly be spelled out. We are talking now about a long line of Jewish writing in Greek, conventionally studied under various rubrics. The new Sch?rer (1986) provides the fullest and best overview and John Collins' standard study (1983), now in a second edition (2000), offers an interpretation. Still, it is not easy to establish a sound footing on which to base a general assessment. Works have been lost, such as the full history of Jason of Cyrene; no doubt there are some lost works we do not even know about. Important early contributions exist in a few pitiful fragments, such as those of the historical, dramatic, and philosophers which were transmitted selectively via earlier excerptors and then through Christian writers. Apocryphal and pseudepigraphic works present a special difficulty in their lack of named or identifiable authors. Geographical contexts are more often unknown than known and they are certainly varied, even if, generally speaking, the Sitz im Leben is the Greek-speaking poleis around the Mediterranean. A few texts might have Jerusalemite authors, for example the first surviving fragmentary historian Eupolemus, who is often identified with the ambassador of Judas Maccabaeus; and, again, the translator of I Maccabees from Hebrew into Greek. The culmination of the line is Philo and Josephus, writers who are each in some respects sui generis, which makes the task of definition even more difficult.
There is a unity, however, which makes it appropriate to speak of a tradition precisely in the visible and constant dependence on the Greek versions of the Hebrew bible. That the language of the text is Greek does make a big difference, as we shall see. We should remind ourselves once again that we can safely talk only about writings, or sometimes about writers, rather than about their audiences. And there may well have been an educational gap between the two. But the writers, at any rate, apparently know some, if not all, of the Greek bible well, and they use it inventively and creatively. They may even use similar techniques as Qumran pesher does when they cite bible in Hebrew. This is nicely illustrated by Dimant (1986: 3-6) in relation to the theological application of phrases from the Song of Moses in Deuteronomy to the martyred brothers' prayer in II Maccabees 7.6. Anther example she adduces are the semi-quotations of Psalm 2, v.10 (part of a sapiental address to the kings of the earth) in the opening invocation of Wisdom of Solomon to oi$ kri/nontej th_n gh_n, together with Wisdom 6.1, which establish a kind of 'exegetical dialogue' between Wisdom and its sources. Dimant generalizes her case in her general study (1988) of formal characteristics in the Apocrypha and Pseudepigrapha, which concerns itself with works that I have classed under the Hellenistic-Jewish description. Dimant's claim is that biblical allusions are used by authors writing in Greek in just the same way as Hebrew authors use them, though this is to some extent qualified by her own demonstration of a quite widespread tendency to avoid direct biblical citation, except in prayer (distinctly reminiscent of the practice of Acts).
There do exist, in any case, instructive parallels and comparable practices. Overall, however, one remains with a clear sense that the creativity of the work produced in Greek is differently focused; and as a group, these books are distinct in the range they cover, not only of formal genres - many of which are, quite simply, accepted genres in Greek literature - but also of things they do with the bible. There is also a marked difference in proportion, if we are to use what survives as a sample. Most of the works in question indubitably engage in exegesis, one way or another. But elucidation of the meaning of the bible is rarely their primary task. In terms of method, even the most Hellenized of the writers are fond of typology: one thinks of the Exodus motif threading its way through the Letter of Aristeas, of the Deuteronomic models in I Maccabees combined with those from Kings, or of the Esther paradigm in III Maccabees. They use direct citation sparingly, but they favour rewritten bible and the imaginative exploitation of biblical themes. In writing of Miqra as a model for language at Qumran, Fishbane (1988: 356) aptly speaks of the creation of a 'thick archaic texture' through the interweaving of Hebrew passages, and the special resonance created by the richness of the intertextual associations. Such density cannot be found even in the most 'biblical' of compositions in Greek, let us say Wisdom of Solomon, nor in the most linguistically biblical of translations into Greek, I Maccabees.
These differences may be summarized, and in some measure explained, by saying that the main thrust in Hellenistic-Jewish literature is outwards, towards the fusion of two literary heritages, by people who wished to demonstrate themselves at home in both. How far they achieved their objectives is another question. Beyond this, there were in the nature of things other demands placed upon these ancient scholars. They had a translation to guard and to tend. Thus in relation to their bible, much of the main work they were doing is invisible to us. This is the ongoing work of making new translations, given that some books, such as Ruth and Lamentations, may not have been put into Greek for the first time until the firs century CE. Coupled with this was the labour of revision, from at least the first century BCE. As is well known, the Minor Prophets scroll from Nahal Hever reflects a first century BCE revision towards the Hebrew text (Barth?lemy 1963; Tov DJD 8). What is implied by such activity is of course continuing knowledge of Hebrew within Greek-Jewish circles. That may be obvious, but it is rarely stated. There is in fact no reason whatsoever to suppose that knowledge of Hebrew was ever entirely lost among the entire educational and religious leaders of Jewry in Greek milieux, whatever can or cannot be said about Philo on the basis of his false etymologies. Moreover, I would assert that this holds good from the very moment of the presentation of the Torah translation at Alexandria, allegedly under Ptolemy's aegis, though it has become a truism of the history books that the translation was necessitated by the Jewish community's loss of Hebrew. Rare notes of implicit dissent have recently come from Baumgarten (2002), who treats Alexandrian Jewry boldly and unquestioningly as a bilingual society.
So we are looking at bridge builders, with a firm footing in each camp. Translation epitomizes their activities. Their labours constitute no less important or impressive an enterprise than that which comes from the heart of one or other of the two traditions they lived with. In Jewish memory, they have been greatly underestimated. I would query the implicit disparagement by Chaim Rabin (1968: 21) of the 'essentially alien character of Hellenistic Judaism, receiving isolated principles, but without the attendant intellectual atmosphere.' Bridges suggest traffic in both directions. Hellenistic Jewish writers and teachers were probably a significant conduit of such Greek philosophical concepts and scholarly methods as were appropriated, consciously or unconsciously, by Jews who thought and worked in Hebrew or Aramaic. And yet, at least in relation to the Jewish textual heartland, there is justification in viewing the Jewish output in Greek as secondary, its writers as the led rather than the leaders. I think it is possible to assert this without falling into a circular argument about what is centre and what is periphery.
Another crucial point follows from the argument so far. The translators and writers in Greek were by no means disconnected from Eretz Israel. Why were the fragments of the Minor Prophets placed in the caves of Nahal Hever? And what brought the Greek fragments of Exodus, Leviticus, Numbers, Deuteronomy and the Letter of Jeremiah to Caves 4 and 7 at Qumran? (Baillet ed. DJD 3; Skehan ed. DJD 9). Unless the explanation is simply that the documents accidentally turned up in these quarters and, being biblical, were treated with respect and stored for safe-keeping (which is not impossible), we must posit some connection between the two worlds. Maybe the fragments belonged to, or were written by, students from the diaspora, people of Paul's ilk. Extensive speculation is fruitless, but an awareness of the possibilities can be useful. As for the Minor Prophets scroll, its dating is regarded as palaeographically secure, but, given its Nahal Hever provenance, it was conceivably in use at a later date, i.e. by the Bar Kokhva generation. Other Greek literary connections between Eretz Israel and the diaspora are perhaps provided by the historian Eupolemus (as mentioned); by Jason of Cyrene whose work was epitomized in II Maccabees and, who, wherever he actually lived and worked, remained associated with the diaspora by his name, yet was had access to the Maccabaean history in extenso; and by the various synagogues and the Hellenists who figure in Acts 6.
It still remains for us to enquire what else is distinctive about Jewish culture in Greek. Some have posited a tradition of allegorical interpretation, of which we seem to have a starting point in the exegete Aristobulus and we certainly have a climax in Philo. But it makes sense to look in the first place in the direction of language. Linking the LXX and the NT books, and present in almost all the other texts, is a fund of great creative usage of the Greek language. This is a distinctive cultural practice of the first importance. I shall not speak about syntax here, but confine myself to the lexical inventiveness, with a few of whose whose fruits everyone is familiar, for it is the inventiveness which has given us terms still alive in many languages today - diaspora, proselyte, holocaust, idol - to name but four. New terminology arises from the courage of the successive LXX translators in creating a translation vocabulary. I think it relatively unlikely that the terminology predates in a significant way the written translation of Torah, largely because I envisage the very first generations of Jews in Alexandria is primarily Aramaic-speaking. I shall offer you straightaway some heavily summarized observations on a small number of examples:
ei1dwlon, generally in Greek a shadow or phantom, that is to say something false, comes to be the principal and most distinctive word used for the image of a deity or even for the deity itself. Both Philo and Josephus use the word in this sense. Daniel (both Greek versions) occasionally has the form ei0dwlei=on, in which he is followed by I Maccabees. Terms corresponding to the standard 'pagan' Greek vocabulary do figure sometimes in LXX: a2galma, ei)kw/n (denoting the likeness of the image and taking over in Daniel as the term for the various images involved in the story), o(moi/wma (used along with ei)kw/n at Deuteronomy 4 in Moses' exhortation on the subject of graven images, gluptoi/). The TWNT dictionary describes the function of the special word ei1dwlon as polemical, part of an intensifying campaign against idolatry.
u(yo/w : in Septuagintal Greek, the verb u(yo/w, to lift up, exalt, is commonly used in a metaphorical sense, translating several Hebrew verbs, but in the vast majority of cases related to the root rom. The verb makes a rare appearance in a similar sense in the historian Polybius (5.56.12). In Josephus, we find it employed in the Jewish War in such a sense five times, which is telling even if the application is to puffed up mortals rather than divinities. The metaphorical use of a rare noun, u3ywoseij, figures at Psalm 149.6, rendering romemot, the elevation of God through the singing of his praises. The affinity with the divine name theos hypsistos, common to both pagan Greeks and Jews, added to the force of this group of words. In New Testament texts, there is an important theological twist and the eschatological element comes to the fore: John especially exploits a double allusion to the crucifixion and the final ascent. Interesting speculations are prompted by the use of the ordinary noun u3yoj as a metaphor in literary criticism, sublimity, by a philosopher of the early Roman imperial period, Longinus (or Pseudo-Longinus) who famously stands out among pagan writers for his quotation from Genesis.
a3gioj with its long-lived cognates, -i/asma/// -ia/zw/-asmo/j , represents the Hebrew root qadosh and its derivatives, but also sometimes translates at least a dozen other roots, as explored by Gehmann (1954). These cognates are not forms known in ordinary Greek, though /-i/zw is occasionally found in the Classical language. The root tends to connote, following its most frequent Hebrew counterpart, an object set apart either positively or negatively. The use of the neuter, in both singular and plural, to refer to the Temple, and the literal transference of the Hebrew superlative name for the Holy of Holies to produce the Greek formation a3gioj a(gi&wn, are good instances of the creation of a semi-technical vocabulary. The application to the Temple perhaps ensures the popularity of the terms, which become indispensable to Phil and Josephus. The author of IV Maccabees is particularly attached to the cognates, even coining his own. The standard Greek i3eroj is more rarely used in relation to Judaism, but it is noteworthy that I Esdras makes it a term of preference for the Jewish cult.
What we are discussing is not the phenomenon of linguistic 'semitisms' so often taken as the touchstone of the Septuagint's un-Greekness. Indeed, not all the words in this repertoire represent single Hebrew semantic fields, as is clear from my selection: when they regularly translate a number of different biblical Hebrew terms they create their own field. We have learnt to be wary, under the instruction of James Barr (1961), of the seductions of an approach which locates changes in group mentality in the story of individual words. His brilliant attacks exposed clearly the underlying assumption, that words carry an inner as well as an outer meaning, so that the structure of the Hebrew or of the Greek mind can somehow be discerned in its vocabulary. Apart from the Humboldtian metaphysics Barr spotted behind such operations, there is the major problem that they are prone to become the vehicles of a theological endeavour. But Barr's strictures do not preclude explorations of LXX vocabulary as a tool for understanding the world of LXX. He cannot and does not deny that the ascription of new range and associations to existing Greek vocabulary, through its role in the translation process, is an important feature of the LXX process. We need an analysis which does justice to their innovative quality and to the striking histories of keywords in the Septuagint vocabulary. Their extension into Hellenistic-Jewish writing is a fertile field, and, in relation to Philo, Naomi Cohen (2002) has made an interesting start.
The quite substantial collection of terms to which I am referring - its exact limits are not clearcut and is naturally subject to debate - has often been regarded as crucial to the building of NT literature, where they appear with an unparelled density and by no means exclusively within quotations or semi-quotations. There is naturally disagreement as to exactly how indispensable they were in the creation of a new discourse (Jobes and Silva 2000: 184). But certainly, among the words in this group, there are some which acquired a permanent Christian resonance in Greek and a long-lasting theological significance - of those we have discussed, we think immediately of a3gioi (for saints) and e)kklhsi/a. Among other instances, it is worth mentioning the term diaqh/kh for berith: this is a known Greek legal term, but the expected word for this sense would have been sunqh/kh; Josephus and Aquila, interestingly, make the change to sunqh/kh. Here we a come up once more against the potentially problematic dimension of this area of investigation, problems encapsulated for us in the crude instrumentalism of Edward Grinfield. But again, this should not be taken as a reason to back off completely.
Modern study of this topic has rightly proceeded book by book, because the deployment of Septuagint language in different parts of the NT is interestingly varied in its reflection of the character and the agendas of the different authors, as well perhaps of locality and period. But this too seems to be part of a larger phenomenon, for when we turn again to Hellenistic-Jewish literature, we find that the histories of Septuagintal keywords in no way follow a uniform pattern there either. My few chosen examples have shown this clearly enough, and, given the vagaries of word usage in any changing society, it is what we might expect. Their very origins are varied: some are neologisms, some are words with some slight classical Greek antecedents, some are ordinary Greek words used in new ways. In the LXX itself, we have to reckon with the very marked variation in practice between different translators, and sometimes the patterns are quite intricate: qahal, the people of Israel as an assembled body, is translated exclusively by the word sunagoge in the first four books of Moses and largely so in the prophets, but exclusively by ekklesia in all the historical writings and virtually exclusively so in Psalms. As to the subsequent histories of these words, some do not seem to make the grade within surviving Greek-Jewish literature: they figure scarcely or not at all. The literary register of a text was a material factor in the vocabulary its author accepted or rejected. And there were winners and losers among words. Some terms vanish from sight, and it looks as though some characteristic LXX expressions simply did not sit well in the new-fangled later Greek-Jewish literature (where stylistic aims tended to mattered and literary aspirations could be high). Sometimes variants of original LXX words seem to have been devised at a later stage and fared better than the original term. Indeed, there appear, as far as we can tell, to have been several waves, or generations of vocabulary formation. One contributor to these developments was no doubt changing fashion in the use of the Greek koine dialekte in the wider world - itself a topic which has not yet been systematically studied.
In Jewish circles, the linguistic development whereby the Old Greek might have come to seem in some respects old-fashioned, and which pushed Jewish writers in Greek to move forwards and establish for themselves new stylistic freedoms in their parabiblical writing, might be said to have had an interesting counterpart, a kind of opposing tendency. For, at the same time the work of revision and retranslation brought Greek biblical version back towards correspondence with the Hebrew and the preferred language of translation came to be a less idiomatic, more Hebraized idiolect. The culmination of this tendency was the version of Aquila. We might construct the picture as a growing bifurcation in the treatment of biblical Greek.
These linguistic features visible when we study vocabulary of the Hellenistic-Jewish writers reflect cultural developments in Jewish society. They are a product of continuing Jewish adaptation to and of the Greek language, both within the sphere of translation and within the wider sphere of bible-oriented writing. Language was a major factor. When NT communities came into being whose messianic purposes made them into full-scale bible-based communities, they could not only draw on, but also significantly expand the practice of word-creation which had been forged by earlier generations of Hellenistic Jews. And yet, at the same time, language was not everything, and the biblical culture of those Jews who made and used the Greek verbal building blocks to interpret or rewrite the bible did not evolve in isolation, as we have seen, but in a fertile interaction with the land of Israel. I have argued that Hellenistic Judaism took on board influences and interpretative practices which had originated there - including, in all likelihood, in Qumran. This route for the embedded transmission of techniques has to be taken into account in sizing up the very great contribution made by Hellenistic Judaism to the New Testament.

Bibliography of works to which reference is made
Baillet. M, Milik, J.T, and de Vaux, R. 1962 Les 'petites grottes' de Qumr?n. DJDJ 3 (Oxford).
Barr, J. 1961 Semantics of Biblical Language (Oxford).
Baumgarten, A .I. 2002 'Bilingual Jews and the Greek Bible' in Kugel ed., 13-30.
Barth?lemy, D. 1963 Les devanciers d'Aquila: premi?re publication int?grale du texte des fragments du Dod?caproph?ton. Vetus Testamentum Suppl. 10 (Leiden).
Cohen, N.G. 2002 'Context and Connotation. Greek words for Hebrew Concepts in Philo', in Kugel ed., 31-61.
Collins, J. 2000 Between Athens and Jerusalem : Jewish identity in the Hellenistic diaspora. (2nd ed Grand Rapids, Michigan; 1st ed. New York, 1983).
Dimant, D. 1987 'The Problem of Non-Translated Biblical Greek', in C.E. Cox ed., 6th Congress of the International Organization for the Septuagint and Cognate Studies. Jerusalem 1986 (Atlanta, Georgia), 1-19.
Dimant, D. 1988 Use and Interpretation of Mikra in the Apocrypha and Pseudepigrapha', in Mulder ed., 379-419.
Ellis, E. E. 1988 'Biblical Interpretation in the New Testament Church', in Mulder ed., 691-725.
Ellis, E.E. 1991 The Old Testament in the New: Canon and interpretation in the light of modern research. WUNT 54 (T?bingen).
Fishbane, M. 1988 'Use, Authority and Interpretation of Mikra at Qumran', in Mulder ed., 339-77.
Fernandez Marcos, N. 2000 Septuagint in Context: Introduction to the Greek versions of the Bible (Boston and Leiden).
Gehmann, Henry S. 1954 9Agioj in the Septuagint, and its Relation to the Hebrew Original', VT 4, 337-48.
Grinfield, E.W. 1850 Apology for the Septuagint (London).
Kugel J.L. ed., 2002 Shem in the Tents of Japhet. Studies on the Encounter of Judaism and Hellenism (Leiden, Boston, Cologne).
Jobes, K.H. and Silva, M. 2000 Invitation to the Septuagint (Grand Rapids, Michigan).
Lim, T.H. 1997 Holy Scripture in the Qumran Commentaries and Pauline Letters (Oxford).
McLay, R.T. 2003 The Use of the Septuagint in New Testament Research (Grand Rapids, Michigan and Cambridge).
Mulder, M.J. 1988 Mikra: Text, translation, reading and interpretation of the Hebrew Bible in ancient Judaism and in early Christianity (Assen, Maastricht, Philadelphia).
Rabin, C. 1968 'The Translation Process and the Character of the Septuagint', Textus 6, 1-26.
Sch?rer, E. (rev. and ed. G. Vermes, F. Millar and M. Goodman) 1986 The History of the Jewish People in the Age of Jesus Christ 175 BC to AD 135. A New English Edition, vol3, part 1. (Edinburgh).
Skehan, P.W, Ulrich, E. and Sanderson, J.E. 1992 Qumran Cave 4.IV. Palaeo-Hebrew and Greek Biblical Manuscripts. DJD 9. (Oxford).
Stendahl, K. 1968 The School of St Matthew and its Use of the Old Testament (2nd ed. Philadelphia; 1st ed. Uppsala, 1954).
Swete, H.B. 1900 An Introduction to the Old Testament in Greek (Cambridge).
Tov, E. with Kraft, R.A. and Parsons, P.J. 1990 The Greek Minor Prophets Scroll from Nahal Hever (8HeXIIgr). DJD 8. (Oxford).