Printer Friendly

Language, power and identity in ancient Palestine.



In this age of the nation-state, conventional views about group identity are still profoundly influenced by the assumptions of romantic nationalism. One such assumption concerns the importance of shared language in the construction of national or ethnic identity. In this paper I discuss how the Hebrew language functioned in the group ideology of the ancient Jews, and propose that its function there does not correspond to post-romantic expectations.

I argue that there were three stages in the social history of Hebrew. In the first, which lasted until c. 300 B.C.E., Hebrew was commonly spoken by Israelites/Jews in Palestine,(1) but this fact played a negligible role in their construction of their group identity. In this respect they were probably typical of the small nations living around the eastern Mediterranean, and unlike the Greeks, who began to consider shared language an essential element in their corporate identity no later (and possibly much earlier) than the fifth century B.C.E.

In the second stage, from 300 B.C.E. to 70 C.E., Hebrew was no longer commonly spoken, having been replaced by Aramaic. Hebrew, however, remained important because it was still the language of the Jerusalem temple and of the Pentateuch. In part because of consistent imperial patronage, these institutions gradually became the central symbols of Jewish corporate identity in the three or four centuries after their respective foundation (c. 500 B.C.E.) and compilation (c.400 B.C.E.). By the third century B.C.E., Hebrew began to be used on coins and the like in a way which may have been intended to evoke these symbols, and thus Jewish identity. But the temple and the Torah also became real repositories of power, so that there developed around both of them closely related classes of curators. These men used Hebrew to distinguish themselves from the rest of the population, and since curatorship of the Torah was in theory open to all males, mastery of Hebrew was also a path to prestige, and study of it was widespread in certain circles. In this second stage then, Hebrew, no longer commonly spoken, became a commodity, consciously manipulated by the leaders of the Jews to evoke the Jews' distinctness from their neighbours, and the leaders' own distinctness from their social inferiors.

The third stage began with the destruction of the temple in 70 C.E. and petered out after the Christianization of the Roman Empire in the fourth century. In this period, the temple and the Torah retained their symbolic centrality, but lost their political importance. As a result, Hebrew retained its evocative power but, in the absence of a strong curatorial class, lost much of its practical significance. A similar set of conditions prevailed in the Jewish Diaspora - had in fact prevailed there even before 70 C.E. - with similar results: here Hebrew was of marginal importance, even more so than in post-destruction Palestine, because the Diaspora lacked a strong, temple-based, curatorial tradition.

My argument about Hebrew and its changing function is logically secondary to an argument about the structure of ancient Jewish society and how it changed, in part in response to changing patterns of imperial domination. I thus propose a relatively complex analytic scheme, because I would like to show how only a presentation of Hebrew as a component of larger social and political structures and transformations can begin to make sense of the complexities of the evidence - complexities which conventional empiricist, quasi-ethnographic, accounts of the history of the language have failed to explain.

But such an analytic scheme has a rather different sort of advantage, too. Of all the national, ethnic and tribal groups living in the Mediterranean world and the Near East under Achaemenid (539-332 B.C.E.), Hellenistic (332-31 B.C.E.) and Roman rule (31 B.C.E.-312 C.E., for the purposes of this paper), only the Jews produced a large body of literary writing significant parts of which have survived to the present. (The case of Greek writing is more complicated because though the Greeks, too, were dominated, at least by the Romans, the Greek style was adopted and promoted by the Romans themselves in the East.) But the opportunity to utilize this material to examine the cultural effects of domination by the ancient empires has largely been missed. This is in part because the specialized knowledge required to master these ancient texts has tended to inhibit comparison and generalization. But historians have also been held back by the sense, rarely acknowledged or admitted, that the Jews were incomparable. One purpose of my analytic scheme, then, is to suggest a way in which evidence about the Jews might be used to make more general points about the cultural effects of imperial domination in antiquity.



The observation that different nations speak different languages, attested in a few scattered near eastern texts of the Bronze Age, becomes commonplace in texts of the Iron Age. A familiar example is the catalogue of Trojan allies in the Iliad (in which, however, only one group, the Carians, is distinguished by language, as barbarophonos);(2) collections have been made of comparable Egyptian and Akkadian texts,(3) in which language, side by side with physical appearance and behaviour, appears as a feature of national differentiation much more prominently than in Homer.

In Mesopotamian and Egyptian texts (and iconography), these "ethnographic" catalogues tend to serve an imperialist function - emphasizing the diversity, strangeness and ferocity of enemies overcome and reduced to tribute (in Akkadian the same word, nakru, means both enemy and foreigner).(4) But the tribute-bearing nations too, among them the Israelites, had observed that language could be a feature of national difference: this is at most implicit in the Tower of Babel story in Genesis 11. This story is not, of course, an empire-glorifying list of subject nations but, at least in its original form, a myth about how the creator god imposed linguistic diversity on humanity. No similar myths are preserved from elsewhere in the Near East, but they may be assumed to have existed in Egypt, where Iron Age and later liturgical texts commonly praise various creator gods for diversifying human speech.

But recognition of linguistic diversity and the occasional mocking of foreign, especially Egyptian, speech(5) do not imply any particular self-consciousness about a national language. Indeed, the Pentateuch (in its present form c. 400 B.C.E.), which is concerned to promote the separateness of the Israelites/Jews, has (apart from the Babel story) little to say about language at all. Certainly it does not, as some later texts do, praise Hebrew as God's language, or prescribe its use in public cult or private prayer: Hebrew is never even mentioned. There is in the entire Hebrew Bible a single passage in which Hebrew - now for the first time identified as a language separate from its neighbours ("Judahite") - is definitely associated with Israelite identity (Neh. 13:23-30):(6)

In those days too I [Nehemiah] saw that the Judanites had taken Ashdodite, Ammonite and Moabite wives. And their sons spoke half Ashdodite [or, "their sons, half spoke Ashdodite"] . . . and could not speak Judahite. And I fought with them and cursed them and beat some of them and pulled out their hair, and I made them swear by God: "You shall not give your daughters to their sons nor marry your sons or yourselves to their daughters. Did not Solomon king of Israel sin on account of such (women)? And among the many nations there was no king like him, and he was beloved of his God, and God made him king of all Israel, yet even him did foreign wives cause to sin. Shall we then listen to you and do all this great evil,(7) to commit a trespass against our God by marrying foreign wives?" And one of the sons of Yoyada' son of 'Elyashib the high priest was son-in-law of Sanballat the Horonite,(8) and I put him to flight. Remember them, O God, for having defiled the priesthood and the covenant of the priests and Levites. For I purified them [the priesthood] of every foreign thing.

Even here, the function of the statement that the children of mixed marriages could not speak Judahite is not clear; it does not figure in Nehemiah's actual argument with the perpetrators, where he denounces mixed marriage as a violation of the Law, a trespass against God and a source of impurity, but not as a linguistic crime.(9)

Why was Hebrew apparently not central to the self-understanding of the Israelites? I would suggest, in brief, and as a partial explanation only, that the extreme political fragmentation of the eastern Mediterranean - a consequence of topography and of the political practice of most of the imperial powers who ruled the area - was not accompanied by a corresponding linguistic diversity. We should imagine the entire region from the Amanus to Pelusium and from the desert to the sea as a linguistic continuum in which each group could understand its neighbours' languages - languages which were hardly more diverse than the dialects of Greek. In the book of Ruth, for example, it is assumed that Judahites and their Moabite neighbours spoke mutually comprehensible languages.(10) Yet, despite the occasional "international" alliances between kings or aristocratic families, there was a powerful tendency towards particularism confirmed and exploited by the policies of Assyrians, Chaldeans and Achaemenids, who ruled the area from the ninth to the fourth centuries. So, you loathed your neighbours - but spoke their language.

The case of the Israelites and their neighbours may then serve as a warning against the common assumption, usually tacit and untheorized, derived from modern nationalist ideology and confirmed for classicists and ancient historians by the example of the Greeks (meaning especially the classical Athenians), that shared language is necessarily a significant component of national identity - from the actors' perspective. The Israelites of the Iron Age shared a language but apparently attached little importance to this fact, for they shared it also with their neighbours; later, the Jews did not share a language but still often thought of themselves as a nation.

There are two substantial surviving corpora of writing from the Mediterranean world of the Iron Age (say 700 to 300 B.C.E.) which we would recognize as "literary", the Greek and the Israelite. Comparison of these corpora is the best way of determining what is peculiar about Israelite writing; without attempting to present a rigorous argument, I would like to suggest that some of these peculiarities have some relation to the non-coincidence among the Israelites of language and national self-definition. I give two examples.

First, the absence of an elaborate metalanguage to describe different types of written (and spoken?) discourse. There are words for different genres, "psalm", "song", "dirge" and "prophetic speech", but no words for the stylistic components of these genres (as in the Greek iambs, dactyls, cola and so on) nor, for that matter, for "poetry" or "prose", though we would say that the genres mentioned are generally speaking written in poetry. Even within genres, there is a remarkable stylistic lability: poetry gives way to prose, and prose is studded with poetry - often not marked, but simply slipped into. The absence of terminology, and the lability, imply (but only imply) an absence of linguistic self-consciousness which in turn has important social-historical implications: literate (or articulate) Iron Age Israelites did not think about language in the same way as Greeks of the same period did; to the extent that Israelites were conscious of "style", it was not for them disembedded from its social function. Hence there was no use for a word for "poetry", which describes so many different types of expression as to be socially meaningless. Nor was there any use for words describing units smaller than the generic: a prophet apparently learned how to speak "prophecy", and a psalmist to sing "psalms".(11)

Greek (especially, but not only, Athenian) practice was obviously utterly different. Use of an elaborate prosodic terminology was accompanied by the sense that certain forms were appropriate for certain occasions, but also, at least by the end of the Archaic period, that the conventions were only that: they could be stretched, manipulated, selectively violated.(12) Hence, to take just one example, Greeks, starting with Aristophanes's Acharnians at the latest, wrote parody, the most self-conscious of literary styles, while Israelites, who were perfectly capable of expressing themselves satirically, did not.(13) I would suggest that these opposite characteristics of Iron Age Hebrew and Greek writing reflect utterly different attitudes to the desirability of sophistication and elaborateness in the manipulation of language. This is not to say that there was no Israelite eloquence or "high style", only that genres remained embedded in the social contexts which generated them.(14)

Secondly, the Israelites' attitude toward their language may also help explain the tendency of some Israelite/Jewish texts of the later part of this period to slip back and forth not only between styles, but also between languages. Ezra-Nehemiah (c.350 B.C.E.) and Daniel (compiled c. 165 B.C.E., but containing, especially in the first half of the book, material perhaps a hundred and fifty years older) notoriously begin in Hebrew, shift into Aramaic (in one case to quote an alleged Achaemenid document, in the other to quote a speech), and then "get stuck" in Aramaic, continuing the narrative in that language. Only later do they return to Hebrew.(15) Such a phenomenon, whatever other explanations there might be for it, is possible only in a context, perhaps primarily scribal, which is not only thoroughly bilingual, but also unselfconscious about what language it happens to be using.(16)

In sum, the Israelites shared a language but tended not to consider it an essential component of their corporate identity. In this respect it was the Israelites, not the Greeks, who approximated to the norm in the eastern Mediterranean of the Iron Age. This view of language lies at the root of some of those characteristics of Israelite writing which are most striking for the western reader: the apparent absence of literary self-consciousness, the poverty of Hebrew literary and prosodic terminology, the incomplete differentiation of poetry and prose, and the unproblematized bilingualism of some of the works.



The bilingualism of such texts as Ezra-Nehemiah and Daniel hints at one of the most significant and least examined cultural developments in the history of the ancient Near East: the changing use of Aramaic. Before c.800 B.C.E., Aramaic was the language of several nomadic tribes living at the edge of the Syrian desert; subsequently, until 332 B.C.E., it was used as an administrative language in the Assyrian and Achaemenid Empires, and in this latter period it gradually replaced, almost completely, a large number of spoken languages in the eastern Mediterranean. Apparently no one compelled people to stop speaking their native languages; nor are there likely to have been Aramaic schools in every village comparable to the Qur'anic schools which slightly over a millennium later helped bring about the replacement of Aramaic, Coptic and Greek by Arabic. Yet by 300 B.C.E. or slightly later, a whole host of local languages, Hebrew, Ammonite, Moabite, Edomite, the unknown languages of the Lebanese, Syrian and south-eastern Turkish hinterland (some of which may in fact have been dialects of Aramaic, while others may not have been Semitic at all), and in many places possibly Phoenician(17) - not to mention the various dialects of Akkadian, too - had lost currency and were spoken, if at all, only by a small number of people.(18) It is true that Aramaic is fairly closely related to the Canaanite (but not Akkadian) languages, including Hebrew, which it replaced; but a comparable situation would be to imagine the Greeks simply giving up Greek and switching to Thracian.

I cannot explain the mechanics of the Aramaization of the Eastern Mediterranean, though the Assyrian and Chaldaean practice of large-scale transfer of populations surely played a role in the Aramaization of Mesopotamia, and it would be reasonable to suppose it to have been a factor in Syria-Palestine, too.(19) However it happened, I would suggest that it tends to confirm one of the conclusions of the previous section, namely that national and tribal groups in the Near East generally did not consider language an essential component of their group identity.

But how thorough was the Aramaization of Palestine? The language of common speech is of course irrecoverable, and evidence for the written language is poor; moreover, the written material is obviously difficult to use as evidence for spoken language, since in some cases writing may reflect no more than scribal practice. And in all cases writing is necessarily related to speech in highly complex and sometimes highly attenuated ways. Besides, literacy was presumably limited along social lines, and ancient Palestinian society was rigidly stratified according to class and status, so that we can scarcely do more than guess at the language(s) used by the lower levels of society.

There have in fact been several different such guesses. Until the middle of this century, the common view about the relative importance of Hebrew and Aramaic was roughly as follows: by the third century B.C.E., Hebrew had given way to Aramaic, certainly for common speech, and probably for most other purposes, too. The language of later Hebrew literature is artificial - the result of efforts by native speakers of Aramaic and/or Greek to write in a dead classical language. This consensus was challenged early in the century by the semiticist M. H. Segal, who argued that the language of the Mishnah, the earliest compilation of rabbinic law, published (in some sense) c.200 C.E., was written not, as had previously been thought, in an Aramaic-calque language (that is, in a version of Hebrew whose peculiarities stemmed from mechanical translation from Aramaic), but in a formalized version of the Hebrew vernacular commonly spoken in Judaea until the Bar Kokhba Revolt (132-5 C.E.), and still used by the children of Judaean emigrants to Aramaic-speaking Galilee, among them the Rabbis, in the late second century.(20)

Initially, Segal's argument convinced few outside a circle of Zionistically inclined Hebraists, and it soon required modification when it became clear from fragments of early manuscripts of the Mishnah found in the Cairo Genizah that its language was far more heavily Aramaized than Segal, who worked from late manuscripts and printed texts, had supposed. Nevertheless, the discovery of the Dead Sea Scrolls at Qumran in 1947, which included many documents written in the third-first centuries B.C.E. in a Hebrew sharing features with the language of both the Pentateuch and the Mishnah, convinced many that Segal had basically been right: it was now possible to trace a more or less steady evolution in written Hebrew, from early biblical to late biblical (e.g., Chronicles, Ezra-Nehemiah, Ecclesiastes) to Qumranic and finally rabbinic. This evolution could not be explained solely as a consequence of the progressive influence of spoken Aramaic and Greek on a classical literary language. It seemed more reasonable to suppose that the language of these texts reflected and was influenced by a Hebrew vernacular still reasonably vigorous. Most have agreed with Segal that this vernacular disappeared a few generations after the Bar Kokhba Revolt, which put an end to large-scale Jewish settlement in Judaea, though it has very recently been argued, none too compellingly, that Jews in northern Palestine spoke Hebrew into the Middle Ages.(21)

As a historian, I may challenge this consensus of semiticists and sociolinguists only with diffidence; and it must be admitted that the arguments from Hebrew literary writing seem compelling, at least to the non-professional. Yet the consensual view seems to me to require major revision, if not complete reversal, for the following reasons.

First, the consensus breaks down when it comes to determining who exactly the speakers of colloquial Hebrew could have been. The evidence for Hebrew in non-literary - epigraphical and documentary - writing is slender, though not entirely nonexistent. In the great corpora of non-literary writing from pre-70 C.E. Judaea - the hundreds of inscribed ossuaries from first-century Jerusalem and the vast collection of ostraca used at the southern Judaean fortress of Masada in 66-74 C.E. - Aramaic and Greek are used almost to the exclusion of Hebrew.(22) The greatest authority on the languages of Palestine, and the most moderate and nuanced of the Hebrew "survivalists", E. Y. Kutscher, concluded from the ossuary inscriptions that Aramaic - originally a language of administration and trade - was predominant in Jerusalem, the multi-ethnic administrative, commercial and religious capital of Judaea, while Hebrew prevailed in the countryside (though in Kutscher's view some Hebrew was spoken in Jerusalem and some Aramaic in the countryside).(23) But this view cannot be maintained - and not only because it is so transparently motivated by a romantic desire to attribute Hebrew speech to the Judaean "folk". In a suggestive observation, Kutscher himself had noted the use of Aramaic in place-names in the Judaean countryside, presumably dating from the post-exilic period (i.e., after 500 B.C.E.); and the ostraca from Masada, published long after his death, seem to demonstrate the primary use of Aramaic by the Judaean rebels who occupied the site. There is no compelling reason to think that they were all, or even mostly, of Jerusalemite origin.

Secondly, the sociological models presupposed by scholars of socio-linguistic inclination in trying to smooth out the complexities and contradictions of the fragmentary evidence have generally been overly simple. The influential characterization of Jewish Palestine as "diglossic", i.e., using two languages, one for formal and the other for informal communication (the "high" and "low" or "home" language, respectively) as applied to any pre-modern, more or less rigidly stratified society, necessarily implies a social distinction: the elites mostly use the high language, perhaps even in informal circumstances (with appropriate distinctions in register), and peasants, artisans and so on, know and use only the low language.(24) But which language was which in Jewish Palestine? Kutscher had after all recognized the frequency of Aramaic and Greek - "low" languages in most schemes - in the funerary inscriptions of well-to-do Jerusalemites (not to mention Judaean literary writing); but Hebrew, in various registers, both "classical" and "vernacular", was a significant vehicle of literary expression - presumably also an elite activity; and the "vernacular" register of Hebrew was allegedly the language of the Judaean peasantry. The strict "diglossia" model thus fails to explain the ancient Palestinian evidence, because it was developed to describe the linguistic situation in certain more or less industrialized developing nation-states. Such entities are generally characterized by relatively permeable social boundaries, wide dissemination of education, prominence of the media and so on, so that many individuals are proficient in both a common language, learned and spoken at home, and a high language, which is learned in school and used for some purposes by the media.(25) Jewish Palestine, by contrast, had a mixed elite of temple staff and large landowners, who interacted with each other, with the country's foreign rulers, and with the general populace in various and complex ways - a patchwork of competing ethnic and religious groups, and a profound yet often bridged division between city and country. Normal as this may have been for the ancient Mediterranean world, it was quite different from the societies from which sociolinguists generate their models; in Jewish Palestine the preconditions for diglossia in the strict sense did not exist.(26)

A third failure of the consensual view is its unwareness of ideological factors in the use of languages. There are no doubt many reasons for this blindness: the tendency among some linguists (as also among many social scientists) to privilege observation over explanation; the tendency of positivist historians to treat writing as unmediated "evidence" and so pay little attention to content, context, social function and other inconveniences; a related scholasticism, which causes confusion between the beliefs expressed in a classical literary corpus and those actually held in a society; and the ideological position of some Zionist scholars, whose zeal to establish Hebrew as the Jewish national language ironically prevented them from acknowledging the ideological uses to which the same language may have been put in the past.(27)

However generated, this blindness to ideology seems to me critical, for it is precisely from the third century B.C.E. onwards that the Hebrew language began to be ideologized, so that its use was no longer a matter of indifference, but came to acquire symbolic weight and social importance. The literary works, coins, documents and so forth in which Hebrew is used cannot be considered simply as "evidence" of a linguistic situation: what they are in fact evidence of is the ideological function of the Hebrew language in ancient Jewish society, and it is as such that I will consider them in what follows. There I will assume that although Hebrew may have been a spoken language in some circles of Palestinian Jewish society in the third century B.C.E. and following, there is no reason to think that these circles included large numbers of the Judaean "folk". All of the suppositions underlying the belief that Judaean villagers preserved the Hebrew language for centuries after the fall of the kingdom of Judah in 586 B.C.E. - their isolation, their cultivation of "purity", and their ferocious resistance to foreign influence, even apart from conventional assumptions about the importance of language for national identity - are questionable. Indeed, the real story of rural Judaea is demonstrably one of flux and stasis, contact and withdrawal, accommodation and resistance, and obviously has far more complex linguistic implications.

In sum, the evidence from Palestine in the sixth to the third centuries B.C.E. is consistent with the view that the Jews, like almost all other national and tribal groups in the Levant and Mesopotamia, generally came to adopt Aramaic as their normal means of communication. The contrary view is based on a simplistic reading of the evidence, characterized by insensitivity to its content and context, and correspondingly to the social and political functions of languages in pre-modern societies. In what follows I will provide a different account of the uses of Hebrew in Hellenistic and Roman Palestine.

Despite the ascendancy of Aramaic and, later, of Greek throughout the Near East, the survival of Hebrew after 300 B.C.E. was not a unique phenomenon. New texts continued to be composed in Akkadian, Old Egyptian and Phoenician down to the end of the Hellenistic period (31 B.C.E.), and in the case of Old Egyptian well beyond this date.(28) If we can trust a hostile testimony quoted in a papyrus of the third century C.E., Idumaeans in Egypt continued to use their ancestral language in their temples almost four centuries after their homeland had been conquered by the Judaeans.(29) Akkadian, hieroglyphic and Phoenician writing, too, are found especially, though not exclusively, in temples. It seems likely that the liturgy of the Jerusalem temple was performed in Hebrew, despite the fact that (like the other languages just mentioned) it was no longer commonly spoken. Some psalms included in the canonical Psalter, which was probably used in the temple liturgy, were composed in the third century B.C.E. or even later, and sectarian groups continued to compose Hebrew psalms for their own liturgical use - which may have reflected that of the Jerusalem temple - into the first century C.E.(30) But while little is known about the non-liturgical use of the pre-Aramaic languages of the Near East, perhaps because of the fragmentary character of the evidence,(31) we do possess, as indicated above, a significant and wide-ranging body of Hebrew (in addition to Aramaic and Greek) writing produced in Hellenistic and Roman Palestine. It would thus be a mistake to think of Hebrew in this period as a purely liturgical language, like the archaic Latin of the revived Arval Brethren in early Imperial Rome, or Old Slavonic and Syriac in modern oriental churches. How then can we account for the language's position?

A brief discussion of the policies of the Achaemenid emperors is a good starting-point. The Achaemenids, as suggested earlier, promoted Aramaic; whether this promotion consisted of its explicit establishment by Darius I as the official language of administration, or the mere encouragement of a language which was already in widespread scribal use, will scarcely have affected the cultural consequences of the promotion, which were considerable and may in a general way be compared to those of the non-exclusive use of Greek for administrative purposes in Hellenistic and Roman Egypt.(32) Among the specific consequences of Aramaization was the rise of an international aristocratic Aramaic literary culture in Mesopotamia, Syria, Asia Minor and, to a lesser extent, Egypt. This culture was distinct from its cuneiform predecessor in that scribes copied Aramaic literary texts on papyrus and leather, not on clay and wax tablets, and did so not purely for their own training, but apparently to satisfy the needs of a market of readers.(33) Aramaic thus came to compete with earlier languages on all levels: it became the language not merely of all sorts of government business and public life, but also of common speech and upper-class literary expression; and the end of Achaemenid rule, and so of government patronage of Aramaic, did not lead to any discernible resurgence of the pre-Aramaic languages, for Aramaic was already established as a language of speech and high culture, and was in any case partly replaced by a new international language, Greek.

But other Achaemenid policies functioned to preserve the pre-Aramaic languages. The emperors liked to be thought patrons of native temples where, as suggested above, the old languages were often still used. For example, Cyrus had come to Babylon in 539 B.C.E. as restorer of the temple of Marduk following the alleged depredations of Nabonidus,(34) and the Achaemenids are also known to have authorized the collection and promulgation of the laws of subject nations; the Demotic Chronicle provides a report of the codification of Egyptian law at the initiative of Darius I.(35) In Judah too, the central temple in Jerusalem - in which the Hebrew language was used - was rebuilt with government support, which it received to the exclusion of other traditional holy places; and the so-called "Law of Moses", more or less our Pentateuch, was - whatever the age of its sources - now finally compiled in the Hebrew language (and also in an Aramaic recension, like the Egyptian code?), and imposed on the Judahites with Achaemenid support.(36)

These developments had complex and ramified effects on language, some of which I will now try to untangle. The temple, and especially the Law, came to be the central symbols of Jewish nationhood - so potent that, if we can believe Josephus, at the time of the Jewish revolt against Rome in 67 C.E. the mere public display of a scroll of the Law was enough to make men take up arms.(37) These two symbols were closely associated too, for the Law itself expected the priests of the temple to be among its authoritative interpreters and the executors of its imposition on the people of Judaea;(38) as far as we can determine, the priests did play an especially important role, at least before the destruction of the temple in 70 C.E., and the high priest was recognized as the final legal authority.(39) The transformation of the temple and the Law of Moses from institutions of real but limited importance into the central items in the symbolic language of ancient Judaism must not be thought self-explanatory - after all, the Egyptian law, also codified at imperial initiative and invested with imperial support, seems never to have played a role comparable to that of the Torah in Judaea.(40) Yet we know too little about the history of Palestine from 400 to 200 B.C.E. - a period crucial to the emergence of the temple and the Torah as ideologically central to be able to suggest a complete explanation. I will indicate only a few factors which I consider to have been important for this development.

First, from Cyrus to Nero, almost every imperial ruler of Palestine supported the temple and the Law.(41) This consistent support (probably absent in Egypt)(42) for these centralizing institutions allowed their curators to impose their values and practices on the Judaeans without serious competition from radical dissenters, who either left the country, withdrew to the desert, or behaved very discreetly. It so functioned by empowering the central authorities to establish administrative, judicial and other networks throughout the country, embodying views and practices that (however roughly) approximated to their own, in part because small-town strong men, who necessarily wielded the real power on the local level, were (ideologically speaking) forced to meet the authorities half-way.(43)

The uniqueness of the temple and the Law gave them a peculiar force as centralizers of ideology - even though it was a uniqueness more theoretical than real. Several shrines outside Jerusalem dedicated to the worship of Yahweh maintained a shadowy existence for centuries after the construction of the Jerusalem temple in the late sixth century B.C.E.,(44) and the Law as actually practised in the Judaean countryside, and even in Jerusalem, was varied and often strikingly different from that laid down in the Pentateuch.(45) Yet the non-Jerusalemite shrines were all, or almost all, located outside Judaea - that is, outside the jurisdiction of the high priest. And in the official ideology, local, non-Pentateuchal varieties of the Law were nevertheless part of a larger conception of the Law, provided that their practice could somehow be reconciled with Pentateuchal prescriptions: the ideological uniqueness of the Law was thus preserved by co-optation.(46)

Surprisingly, the emergence of temple and Torah as the central symbols of Judaism cannot be adequately explained, or even traced. That said, it seems obvious that the "Antiochan persecution" (starting in 167 B.C.E.) - in which the Seleucid emperor prohibited observance of Jewish Law and handed over the Jerusalem temple to a group of reformist priests who abolished the distinctively Judaean elements of the cult - and the subsequent, and ultimately successful, Judaean revolt (167-152 B.C.E.) did much to simplify and consolidate the symbolic system of the Jews.(47) It may be significant that the earliest story to presuppose that the Torah-scroll itself possesses symbolic importance is in the account of the persecution in 1 Macc. 1:56-7, composed perhaps forty years after the events it describes: in the course of the persecution, Torah scrolls were not confiscated and discarded, but ripped to shreds and burnt - something not easily done to parchment. The absence of this information in the parallel account in 2 Maccabees casts doubt on the episode's historicity, but not on the significance of the claim as an indicator of the ideological position of its post-revolt, pro-Maccabean author.

The emergence of the temple as ideologically central is somewhat easier to trace, since admiration for it and its staff is the theme of a long series of ancient Jewish literary works, from Chronicles (c.350 B.C.E.) to 2 Maccabees (c. 100 B.C.E.) to Josephus's Jewish War (c.80 C.E.), and is reflected also in works of pagan writers, from Hecataeus of Abdera (fl. 300 B.C.E.) onwards.(48) That this centrality was not simply the fantasy of a small group of authors themselves of priestly background is demonstrated by a series of events from the feuds of the late sixth and fifth centuries B.C.E. (both among the Judaeans and between them and their neighbours) over access to and control of the temple, to the widespread enthusiasm generated by the emperor Julian's plan, in 361 C.E., to rebuild it; and it is further confirmed by the fact that the post-Hasmonean temple, at least, was a massive economic "black hole", sucking in and consigning to oblivion (often in fact to the strong-boxes of rapacious Macedonian and Roman generals) vast quantities of surplus Jewish silver - not all of it, we must suppose, given under duress.(49)

For my purposes, these developments are important because the Hebrew language was closely associated with these two central symbols, Torah and temple, and so came itself to have a certain symbolic force: it became not the national language of the Jews, but the language whose representation symbolized Jewish nationhood. The earliest indication of this development may date back to the middle or later fourth century B.C.E. Most documentary writing from Achaemenid Judaea - inscriptions on coins, seals, bullae (stamped pieces of clay used to seal papyrus or parchment documents) and so on - is Aramaic, though some probably quite early bullae, made at a time when Hebrew was still commonly spoken, bear Hebrew inscriptions in Aramaic script. All Achaemenid Judaean coins give the name of the province in its Aramaic form and in Aramaic script, but the latest type of Achaemenid coin features two examples bearing inscriptions in Hebrew and in Palaeo-Hebrew script, one of which, interestingly, gives the name of the high priest, and the other, that of a governor who may also have been high priest.(50) By the early third century B.C.E., though, Hebrew alone is used in inscriptions on the coins, written in Palaeo-Hebrew script; and the name of the province now has its old Hebrew form. Official stamps of Hellenistic Judaea, of either the early third or early second centuries B.C.E., likewise use the old Hebrew script. However, it is difficult to know how these inscriptions functioned; it may be worth pointing out that the iconography of the coins is not terribly distinctive - under the Achaemenids it consisted of a repertoire of motifs borrowed in part from Athenian coinage and then, under Ptolemy, imitations of Alexandrian issues. Both sets of coins feature a large number of anomalous types which are difficult to interpret but not distinctively "Judaean", at least not in any obvious way.

The interpretation of the coinage of the Hasmoneans is characterized by a different set of complexities. The Hasmoneans were the family who led the revolt against the Seleucids starting in 167 B.C.E., ruled in Palestine 152-37 B.C.E., and revived autonomous coinage in the 120s or 110s B.C.E., about one hundred and fifty years after it had been abolished. As I suggested above, the revolt which was the dynasty's raison d'etre had tended to magnify the symbolic centrality of the Law and the temple; this was perhaps accompanied by the first explicit and unambiguous uses of Hebrew as a national symbol - at least, such a use of the language was retrospectively attributed to the rebels. The quasi-official chronicle of the revolt and the rise of the Hasmonean family, 1 Maccabees, was composed in archaizing Hebrew;(51) the author of 2 Maccabees (an account of the revolt composed in Greek and unconnected with 1 Maccabees) emphasized, with an uncertain degree of accuracy, that revolutionaries and martyrs of the persecution used Hebrew in some circumstances.(52) The attribution of symbolic importance to Hebrew - whether by the rebels themselves or by their successors - may help explain why the earliest Hasmonean coins, minted under John Hyrcanus I (reigned 134-104 B.C.E.), bore legends exclusively in the Hebrew language and in the presumably increasingly incomprehensible Palaeo-Hebrew script, accompanied by a distinctive iconography no longer simply imitated from Hellenistic royal issues and Greek city coins.(53) Yet John's son Alexander (reigned 103-76 B.C.E.) minted coins with inscriptions in three languages: Hebrew (in Palaeo-Hebrew script), Aramaic and Greek, with no concomitant variation in iconography. What might this mean?(54)

The role of Hebrew in the two Judaean revolts against Rome is less ambiguous. The silver coins of both the Great Revolt (66-74 C.E.) and the Bar Kokhba Revolt bear inscriptions exclusively in Hebrew and in Palaeo-Hebrew script, at a time when ability to read this script must have been very rare indeed, even among those literate in Hebrew;(55) the inscriptions are accompanied by appropriate images, usually connected to the temple cult. There seems little doubt that the language functions on these coins practically as a talisman - an important element in the iconography of these powerful and ubiquitous expressions of Judaean national defiance.(56)

I suppose, then, that for most of the Jews of Palestine, Hebrew was important mainly as an item in a complex of symbols - a fact to which Judaean revolutionaries gave a fairly unambiguous material expression but which was, if less conspicuous, no less true in more peaceful times. These symbols no doubt shaped the Jews' consciousness in various unknowable ways and created among them the feeling that Hebrew was somehow "their" language, but in practice Hebrew impinged on their lives rather little. This is because those contexts in which Hebrew was used were remote from most people's regular experience: a passage in Josephus's Jewish War creates the impression that copies of the Pentateuch could not usually be found in Judaean villages(57) - not surprisingly, given the undoubted expense of parchment scrolls.(58) This implies that there was in most villages no regular time or place for reading the Law, at least until after 70 C.E., when synagogues - where, it must be assumed, a portion of the Law was usually read weekly - became common in rural Palestine.(59) It is also unlikely that, at least before 70 C.E., elementary education in the Law (and so in the Hebrew language) was provided for most boys, though it was certainly common in some circles; the late rabbinic tales about the establishment of a "state" school system in the first century are probably idealizations.(60) Whatever the typical Judaean peasant knew of the contents of the Pentateuch, he knew from oral report, presumably in Aramaic.

The transformation of the Law and the temple into important symbols may be supposed to have enhanced their practical significance as, respectively, the civic and religious constitution of the Jews and the place where (among other things) it was authoritatively interpreted. The Pentateuch was unique among ancient law-codes, national epics and the like, in that it required general acquaintance with itself (a fact whose significance was much exaggerated by ancient Jewish apologists) and was in theory, unlike the temple, not the exclusive province of a class restricted by birth.(61) In practice, this meant that most Jews knew something of the Law as applied in daily life. But despite the non-esoteric character of the Law, its practical importance, its complexity, and the technical impossibility of widespread mastery of it in the conditions of an agrarian subsistence economy, combined to generate a powerful, though loosely organized and to some extent fragmented, curatorial class, responsible for preserving and interpreting it. The priests, or the curators of the temple - a hereditary class of high status in its own right - were, as suggested above, given by the Law itself a special role in its own interpretation and execution. But the public character of the Jewish Law - as opposed to Egyptian or Babylonian - meant that access to curatorship of it was not restricted to priests, but was open to all - all, that is, with enough wealth to finance study. Village scribes, local arbitrators, and so on - probably for most Jews the most conspicuous representatives of the class - were not overwhelmingly likely to have been priests. Thus, knowledge of the Law, even when not accompanied by priestly birth and political connections, was a path to prestige. Even in the absence of evidence we would have to suppose that among some Judaeans knowledge of the Law was zealously cultivated. Hebrew, as the language of the priestly curators of the temple and of the mixed (priestly and lay) curators of the Law became, in short, a language whose use both distinguished a class and determined admission to it - a class whose legitimacy was in turn presumably confirmed by the language's symbolic power.

The function of Hebrew in the ideology of the curatorial class is well illustrated by a passage from the Book of Jubilees, an imitation and expansion of the book of Genesis allegedly narrated by an "angel of the (divine) Presence", composed in Hebrew in the second century B.C.E. (it survives now in full only in a translation into Ge'ez, or Old Ethiopic):

And the Lord God said to me [the angel], "Open [Abraham's] mouth and his ears so that he might hear and speak with his mouth in the language which is revealed because it ceased from the mouth of all the sons of men from the day of the fall [of Babel?]". And I opened his mouth and his ears and his lips and I began to speak to him in Hebrew, in the tongue of creation. And he took his father's books - and they were written in Hebrew - and he copied them. And he began studying them thereafter. And I caused him to know everything which he was unable to understand. And he studied them in the six months of rain.(62)

It would be tiresome, and is in any case impossible here, to trace the effects of this development in the masses of literature produced by this class from the third century B.C.E. to the third century C.E.(63) But some observations may be made. Almost all surviving Palestinian writing from this period (not all of which, by the way, was originally composed in Hebrew) is informed by familiarity with the Pentateuch, and much of it actually imitates the Pentateuch and the other biblical books: Jubilees imitates Genesis, Ben Sira Proverbs, the Temple Scroll Deuteronomy, 1 Maccabees the Deuteronomic histories, and so on. That is to say, one of the consequences of the centrality of the Pentateuch and the concomitant rise of a curatorial class - for which the literature just mentioned is important evidence - is the pervasiveness of literary classicism, that is, the tendency to emulate a more or less discrete body of writing which has come to be thought uniquely valuable and significant.(64) The self-consciousness which classicism implies is, in fact, a novel and defining characteristic of Judaean literature from the third century B.C.E. onwards; it also conforms with the growing importance of Hebrew in the same period as a symbolic commodity and is, like it, an aspect of the ideologization of the language which I have previously discussed.

Imitation is only one possible response to the development of a classical literature; in the Jewish case, as in the Greek, it was accompanied by the rise of detailed study, explication and commentary - in short, scholasticism. Jubilees, for example, contains elements of both classicism and scholasticism, since it is simultaneously an imitation and expansion of Genesis, and an attempt to work out the details of biblical chronology. In the first century B.C.E., if not earlier, the sectaries of Qumran produced commentaries on biblical books, a genre later taken up by Philo, the Rabbis and the Church Fathers. In some cases, Jewish classicism and scholasticism functioned not only as an expression of the ideology of the curatorial class, but probably also as a reaction to Greek classicism and scholasticism: for example, the insistence of Ben Sira, writing c. 190 B.C.E., that the Law is the source of all wisdom has been plausibly supposed to address the growing popularity of Greek education among well-to-do Jerusalemites, while such works as Eupolemus's History of the Kings of Judaea, written in Greek a generation after Ben Sira by a member of his class, may be a different manifestation of a similar impulse.(65)



The curatorial class did not disappear after the two Judaean revolts, but was reconstituted along different lines; the most obvious contrast with the pre-70 C.E. curatorial class is the reduced importance of the priesthood in the new, rabbinic, class that replaced it.(66) The literature produced by the Rabbis is correspondingly different from its antecedents. It is totally inner-directed: it barely acknowledges that there is a world outside that marked by attention to rabbinic concerns; all trace of classicism is absent here, replaced by scholasticism, presumably reflecting the attribution of a higher degree, or at any rate a different sort, of sanctity to the classical literature. But these formal differences between rabbinic and pre-rabbinic literature should not be allowed to obscure the fact that the Rabbis saw themselves as continuators of the old pre-destruction curatorial class. One way in which they expressed this continuity was by using, and promoting the use of, Hebrew, in some contexts even for speech, though it is normally supposed that in the third century, Aramaic began to replace Hebrew as the main language of rabbinic instruction.(67) It is possible (though not unquestionably legitimate) to construct from scattered comments in the rabbinic literature a full-blown ideology of Hebrew, some of whose details are continuous with, and expand, hints found in pre-rabbinic writing.(68) Thus, some rabbinic texts describe Hebrew as the language of creation, and so of God and the angels,(69) and thus of prayer and study, and to some extent of speech;(70) in some circles, such as those of the author of the Sefer Yezirah, the knowledge that the universe was created by means of the Hebrew language led to elaborate and complex speculations about the relationship between the Hebrew language and physical reality, and such speculations may in turn be connected with the facts that late antique Jewish magical texts, like the Hekhalot books and the Sefer Harazim, tended to be composed in Hebrew rather than Aramaic (in the case of the Sefer Harazim, in elegant classicizing Hebrew), and that Hebrew words and phrases can be found in Greek and Coptic magical texts of the same period.(71)

A striking aspect of the rabbinic ideology of Hebrew is that rabbinic texts most often use a previously rare expression to refer to the language: leshon haqodesh.(72) Though commonly translated "holy tongue", the expression probably originally meant "temple language".(73) It is likely that at some point in the history of the rabbinic movement, its original meaning was forgotten, though the Rabbis can never have been wholly unaware of such linguistic parallels as sheqel haqodesh, whose reference to the temple was unmistakable. Be this as it may, it is equally likely that when the term came into use, the Rabbis were aware of its meaning. They themselves thus associated Hebrew with the temple, one of the two main loci of its use by the curatorial class before 70 C.E. Their designation of Hebrew as the leshon haqodesh may furthermore reflect awareness that their Hebrew differed to some extent from biblical Hebrew - an awareness for which there is some additional evidence.(74) There is good reason to think that the Rabbis' name for their language was accurate. Of the hundreds of Semitic texts discovered on Masada, the only ones written in Hebrew are tags indicating that certain goods had been set aside as gifts to the temple; their editor observed that their language is strikingly similar to that of the rabbinic texts.(75) It seems fair to conclude, then, that the decision to compose the Mishnah (the earliest rabbinic document, c.200 C.E.) in Hebrew - and in a type of Hebrew which was especially associated with the temple, as the Rabbis themselves knew - constituted an act of appropriation, an assertion of rabbinic control over what was symbolically central in Judaism.

Despite these elements of self-conscious continuity in rabbinic ideology, the transition from priest to Rabbi was accompanied by significant changes. The temple now existed exclusively as a symbol, and the Pentateuch was no longer, strictly speaking, the constitution of the Jewish nation; at least no government recognized it as such. Admittedly, the values of the curatorial class had to some extent been internalized by 70 C.E. - hence the frequency of temple and Torah iconography in post-70 C.E. funerary and synagogal art,(76) and the apparent importance of the synagogue, which probably involved the practice of public reading of the Pentateuch. Indeed, the text of the Pentateuch, whether in Hebrew or in close Aramaic or Greek translation, was probably somewhat better known in rural Palestine in the third and fourth centuries C.E. than it had been in the first. Significant as this was, the absence of institutional support for the Law fragmented and marginalized its curators. Some familiarity with the text of the Torah may now have been hard to avoid, but expertise conferred little clear advantage, and formal education in it may have been less common than before 70 C.E., though there is no way to be sure.

Hence, the Rabbis were so zealous about appropriating symbols precisely because they lacked real authority over them. Nor was any other group of self-appointed "curators" in any stronger position. The two failed revolts against Rome shattered the symbolic world of Palestinian Judaism. The shards, including Hebrew, were still extant, but displaced. As for Hebrew, it is uncommon in Palestinian epigraphy of the Roman Imperial period; even in synagogues, where one would most expect to find it, inscriptions are overwhelmingly in Aramaic and Greek.(77) Indeed, in the synagogue-based Judaism of the second to the fourth centuries, Hebrew had little role to play. Even if the Pentateuch was read and the prayers recited in the language, which is not attested, competence in such tasks provided little prestige. The leaders of the synagogues were their patrons, who had little need of learning, and the patrons of the patrons were the Patriarchs, who after 70 C.E. had originally been close to the Rabbis, but gradually pulled away; by the middle of the fourth century C.E., the sons of the Patriarchs studied Greek rhetoric, not Jewish Law.(78) This situation changed, in circumstances completely unknown to us, only after the Christianization of the Empire, when gradually the Rabbis joined the patrons as leaders of the Jews, and Hebrew's importance revived.



Though the relevance of the linguistic situation of the diasporic Jews to that in Palestine is undeniable, conditions in the Diaspora were sufficiently different from those in Palestine as to require a separate treatment here. Before considering the position of Hebrew in the Jewish Diaspora, a few general points need to be made. Though it is probable that those Jews living in Palestine always constituted only a minority of the Jewish people, it is usually thought that by the first century B.C.E. there were many more Jews living outside Palestine than in it; they were especially heavily concentrated in Egypt and the Fertile Crescent, and could be found in large numbers throughout Asia Minor (including remote and basically unurbanized regions like Phrygia and Galatia), in smaller numbers in Greece and the adjacent regions and, starting in the first century B.C.E., in ever-increasing numbers in Rome and other cities of Italy and of North Africa. About some of these Jewish settlements we know next to nothing. We are, for instance, almost entirely in the dark about the apparently vast Jewish community of Babylonia between c.400 B.C.E., when cuneiform documents mentioning Judaeans peter out, and c.250 C.E., the period when the Babylonian Talmud may begin to convey some trustworthy information. About most of the remaining Diaspora settlements our ignorance is profound, and what little we do know is often strikingly anomalous.

In attempting to understand both the great size of the diasporic Jewish population, and the anomalousness of much of the surviving evidence about it, it is necessary to bear in mind the circumstances of the development of Diaspora Judaism: it was not just the result of emigration from Judaea and natural increase. Such a supposition would be, in ancient conditions, a demographic absurdity. The basic facts are simple, and in rough outline generally agreed: until around 100 B.C.E. the Jewish population of Palestine cannot have exceeded 100,000 (and this assumes an implausibly high population density for Judaea), while it is likely that by the turn of the era there were no fewer than 2,000,000-3,000,000 Jews in the Diaspora. Much of the demographic bulk of the Diaspora can only have been the result of the spread of the Israelite religion (which was not identical with Judaism), and later of Judaism, among large numbers of pagans - initially by small numbers of Israelite mercenaries, traders, emigrants and so on. The communities which resulted from this dissemination gradually and haltingly assimilated to Judaism as the Jerusalem temple and its staff became influential, and contacts between Jerusalem and the Diaspora more frequent. Local peculiarities of course persisted, and new ones developed, but an examination of the one relatively well-attested Diaspora community, Egypt, makes the general trajectory of the development clear.(79)

All this is by way of explaining that the Jewish Diaspora cannot be discussed as if it were a single entity, and some parts of it cannot be discussed at all. But the considerations listed above also suggest that by the period of Roman rule, for which we are best informed about the Diaspora (excluding Babylonia), some common features had developed and can be detected, at least in the larger and better attested places, like Alexandria, Rome and the cities of Syria and Asia Minor.

For my purposes, the most significant of these common features are the following. By the first century B.C.E., most diasporic Jews worshipped the God of Israel in prayer-houses, synagogues or some other sort of gathering. What went on in these is generally unknown, but it apparently did not include sacrifice. There was thus in most Diaspora communities no cult and no functioning priesthood, though there may have been individuals of priestly descent: most diasporic Jews sacrificed only when they visited Jerusalem.

Another common feature is that larger Jewish communities were often allowed by the Macedonian or Roman rulers some trappings of autonomy, usually including permission to refrain from participation in the municipal cult, to avoid liturgies and other public duties on the Sabbath, and perhaps to have their own courts, which presumably were expected to judge according to Jewish Law (it ought to be said, however, that municipal authorities did not always recognize these privileges). It must be emphasized that these communities were not autonomous in the sense that medieval Jewish communities, or pre-70 C.E. Judaea, were. The ancient communities were apparently completely voluntary organizations, and the communal leaders had no actual jurisdiction over their constituents.(80)

We should thus expect the major Diaspora communities to resemble structurally the Jewish villages of post-destruction Palestine, with the added complication that the Diaspora communities were located in overwhelmingly Greek-speaking, and later in the West Latin-speaking, environments (I continue to exclude Babylonia from the discussion). That is, we should expect there to have been a fragmented and marginal curatorial class mostly subordinate to the actual, patronal, heads of the communities. Therefore, we should also expect Hebrew to have played, apart from the tailsmanic and evocative, only a small role in the lives of the diasporic Jews. In practice, our expectations are confirmed, but only in part: Hebrew did often function talismanically, but rather surprisingly, it played an even smaller role in the Diaspora than in Roman Palestine. The curators of the Law, who preserved the language in Palestine after 70 C.E., largely gave it up in the Diaspora at a very early date.

The Jews of the Diaspora usually spoke the languages of their environment, as far as we can tell. The surviving epigraphic evidence for the use of Aramaic by the Jewish community in Alexandria - a handful of epitaphs and one or two post-Alexander ostraca mentioning Jews - all date from c.300 B.C.E.: that is, at the very beginning of the history of the community.(81) In fact, the vast Jewish community of Alexandria was not entirely typical in respect to language. Down to the Great Revolt the city was the main target of Palestinian immigration, so Aramaic, and perhaps Hebrew too, never entirely disappeared from the city. But evidence for their use is so limited that it is fair to conclude that the prevailing linguistic pattern among Alexandrian Jews is comparable to that common among modern immigrant groups in the U.S.A. and Europe: within two generations, the ancestral language disappeared from use almost completely.(82) It is worth remembering that the most learned Alexandrian Jew known to us, Philo, seems not to have known Hebrew, and furthermore was not always careful in his writings to distinguish between Hebrew and Aramaic, both of which he called "Chaldaean".(83) Other Diaspora communities are less well known, but the nearly exclusive use of Greek (and Latin) in their epigraphic remains before the fifth century C.E. suggests that linguistically they resembled the Alexandrian Jews except that they were more linguistically uniform (to the extent that they were not goals of immigration from the east).(84)

Such linguistic assimilation may be a familiar phenomenon in modern times, but in antiquity, foreign ethnic corporations often retained their languages for generations; this was an especially common pattern in pre-Hellenistic times. The Greeks of Naucratis and Memphis,(85) and the Boeotian community Alexander the Great discovered near Babylon,(86) are the best-known examples (non-Greek examples are, significantly, harder to find). Several factors are responsible for Jewish linguistic assimilation. Given the origins of Diaspora Judaism, the fact that the Jews spoke Greek was in many cases not strictly speaking the result of linguistic assimilation; rather, it was the result of the religious assimilation of pagan adherents of Yahweh to Judaism. The general use of Greek and Latin in the communities of Alexandria and Rome, where many or most Jews were descendants of Palestinians, probably indicates, first, that Jews living in Palestine and their descendants living abroad displayed identical unconcern for language as a marker of identity, apart from the fact that many immigrants probably arrived already speaking Greek as a first language; and secondly, that general social and political patterns had changed. The classical city and its near eastern counterparts did not assimilate foreigners (witness the case of the Greeks of Naucratis and Memphis), while the Hellenistic city at least made the possibility available to some categories of foreigners. Linguistic assimilation, a prerequisite for broader integration, was now the norm.

But in important ways Diaspora Jews deviated from normal patterns. Though most foreign ethnic communities in Hellenistic Greek cities adopted Greek, they often retained their native languages for cultic and some other public communal purposes: Phoenician communities in the third century B.C.E. - a time when Tyre and Sidon themselves were ostensibly thoroughly Greek-speaking - occasionally honoured prominent members with inscriptions in Phoenician.(87) Especially striking is a case already mentioned - the apparent survival of the Edomite (or Aramaic) language c.200 C.E. among the priests of Apollo-Qos in Hermopolis, Egypt, three hundred years after the temple's foundation, by a community which even in the first century B.C.E. seemed to be losing its Idumaean character. Yet the Jews in the Greek-speaking Diaspora made little or no use of Hebrew for liturgical purposes and the Pentateuch itself was commonly read only in Greek translation (and not, as sometimes in Palestine, in Hebrew and in translation).(88)

The proper place to begin discussion of this peculiarity is with the question of the translation of the Pentateuch into Greek, normally thought to have occurred in Alexandria in the third century B.C.E.(89) The earliest and fullest account of the traditional tale about this translation is found in the so-called Letter of Aristeas, probably composed c. 130 B.C.E.(90) Ptolemy II Philadelphus wanted to complete his Library, so he decided to get a translated copy of the Law of the Judaeans. Since none was extant, he invited to Alexandria seventy-two of the most learned Judaeans, six from each tribe [sic!],(91) presided over by their high priest, to prepare a translation from the most accurate copy of the Hebrew text. They miraculously completed it in precisely seventy-two days, handed over a copy to the king and another to the local Jews, who invoked a curse against anyone who would alter even a single letter. For ever after the Alexandrian Jews celebrated this day as a festival.

This story, clearly the mythical aition of the Jewish festival which celebrated both the second giving of the Law and the Jews' alliance with the Ptolemies, has been rejected by almost all modern scholars and replaced by a different story: the Alexandrian Jews, having forgotten Hebrew, sensibly decided to translate the Pentateuch so that it could be read and understood by those attending the synagogues. I would argue that this modern story is no less an aetiological myth than the ancient one, though a rationalist one, reflecting and intending to justify the use of the vernacular in both Protestant churches and Reform synagogues of the nineteenth and twentieth centuries. The advantage of the ancient fiction over the modern one is that the former acknowledges the problematic character of the translation, representing it not simply as the sensible thing to do under the circumstances, but as a difficult and abnormal act - indeed, a second giving of the Law.(92)

But the fact remains that it happened; and it was a highly consequential development, in that it opened the synagogues and the Jewish sacra to outsiders and so played a critical role in the rise of the theosebeis - pagan adherents of Jewish practices - and so in the spread of Christianity, thought to have begun precisely in these margins of the Diaspora Jewish communities.(93) Why did it happen?

The temple and the Torah were not without symbolic importance for the Diaspora Jews. Their funerary iconography, which is remarkably uniform throughout the Roman Empire, is filled with evocations of them: the menorah, the incense-shovel, the ark opened to reveal the scrolls within are the most common images in both Jewish catacombs and synagogue mosaics.(94) As long as the symbolic world of the Diaspora Jews overlapped in this way with that of the Jews of Palestine, the Hebrew language, as a subsidiary symbol, retained a talismanic function: some Italian and Egyptian Jews ended their relatives' epitaphs with the single word shalom, often misspelt and clumsily carved, sometimes written in Greek letters, though others may have used Aramaic words and "biblical" Greek tags for the same purpose.(95) Philo, ignorant as he was of Hebrew, acknowledged its symbolic importance by repeatedly invoking Hebrew ("Chaldaean") etymologies; that these etymologies were often mistaken was presumably unknown to his audience and did not detract from the power of the rhetorical gesture.

Yet the actual life of the Diaspora communities, even before 70 C.E., was comparable in many ways with that of the Palestinian villages in the second to the fourth centuries C.E. As suggested above, the communities were voluntary organizations, usually tolerated but not supported by the government. Unlike other ethnic communities, they centred not on temples (maintained in some cases by taxation and government gifts, and run by priestly guardians of the ancestral laws) but on synagogues, that is, informal prayer-houses maintained and run by those who could afford to do so and who were capable of maintaining friendly relations with local authorities - in short, well-to-do patrons. While curatorial classes did exist in the largest communities, they were, as in Palestine after 70 C.E., of secondary importance; unlike the later Palestinian curators, they seem generally not to have felt themselves closely connected with the priesthood of the Jerusalem temple (though some Diaspora curators may have been of priestly descent) or with the pre-70 C.E. Palestinian legal authorities - which is not to say that they were never in contact with or influenced by them (and vice versa). They were a homegrown class, who were experts, not in the Hebrew Bible of the Jerusalem temple, but in their own, Greek, Bible of the Diaspora synagogues and law-courts, and whose classicizing and scholastic writing was also mainly Greek.(96) Thus, while archontes (rulers), archisynagogoi (rulers of the congregation) and other titles masking wealthy patrons are commonly found in Roman-Jewish inscriptions, nomodidaskaloi (teachers of the Law) and mathetai sophon (pupils of the wise, i.e., scholars of the Torah) are very rare.(97) To the extent that they existed at all, they were subordinate to the leaders, functionaries rather than authorities. This diasporic realignment of the relations between the curatorial and patronal classes may help explain the position of Hebrew there. To explain its subsequent revival is, as in Palestine, more difficult, and I will not attempt it.


Let me now summarize my main arguments. The Israelites and Jews never considered the ability to speak Hebrew to be a central element of their corporate identity. Their literature was thus, before c.300 B.C.E., characterized by unselfconsciousness about language. Furthermore they, like most other national and tribal groups in the Near East, responded to new conditions created by the policies of the Assyrian, Neo-Babylonian and Achaemenid empires by gradually adopting Aramaic for common speech - a process presumably more or less complete by the end of Achaemenid rule. Yet the policies of the Achaemenids, and of most of their successors, also favoured the survival of the native languages of the Near East. For the Jews, imperial support for the Jerusalem temple and the Pentateuch, in conjunction with other factors now impossible to trace, worked to transform these institutions, with both of which Hebrew was closely associated, into the central symbols of Jewish corporate identity. The Jews thus came to regard Hebrew as a subsidiary national symbol, capable of evoking the central ones - a fact which has left physical traces in the form of inscriptions on autonomous Judaean coins, seals and bullae.

But the temple, as the institution which guaranteed Yahweh's protection of his people, and the Torah, as the officially recognized constitution of that people, also had immense practical importance, so that there developed around them overlapping and closely related curatorial classes. For these, Hebrew functioned as a social marker, a means of distinguishing themselves from the rest of the population. The effects of this linguistic self-consciousness are still discernible in the voluminous literature which this class produced, which in this respect resembles contemporary Greek literature more than that of the Iron Age Israelites. The openness of the curatorial class meant that mastery of Hebrew was not only a social marker, but also an important path to prestige. Study of Hebrew and the Law were thus presumably widespread among certain classes of Judaeans.

The Judaean revolts against Rome resulted in the destruction of the temple and the removal of the Law from its position as the constitution of the Judaean nation. But their symbolic centrality had by now been thoroughly internalized, so that in the second century C.E. a curatorial class of a rather different sort emerged - the Rabbis, who in many respects, including their attitude to Hebrew, were self-conscious continuators of pre-70 C.E. curatorial ideology. However, the Rabbis were marginal; knowledge of Hebrew conferred little advantage. So Hebrew retained for the most part a purely talismanic function: representation of it served to evoke Jewishness.

This was also true, though in a rather attenuated way, in the Diaspora, where the Jewish communities even before 70 C.E. also had a basically patronal structure. There, however, even the curatorial classes for the most part failed to preserve Hebrew; and though the language could sometimes serve as a talisman of Jewishness, so also could other languages, especially Aramaic and "biblical" Greek.

In this account of the social and political history of the Hebrew language I have emphasized the position of Hebrew in the structure of ancient Jewish society. Emphasis on structure requires schematic presentation, which in its turn requires simplification. I have excluded from the discussion such anomalous and poorly attested varieties of what may have been Judaism in the Diaspora as the guild of worshippers of the Most High God in (of all places) Hellenized south Russia,(98) or the most obscure of the many sectarian organizations which grew up in first-century C.E. Palestine itself. These groups often fail to fit the patterns I have described in this paper, in that for many of them there is no evidence that they defined themselves mainly in terms of the symbolic complex of temple and Torah. For my purposes then, if more were known of them they would be useful as counter-examples - of how Hebrew functioned among Jews for whom the language's association with the temple and the Torah was of little interest.

In practice, though, there are only two such groups about which we can say anything at all. One is the military settlers of Leontopolis, Egypt. This community seems originally to have consisted of the legitimate Judaean high priest, Onias IV, and his associates, who fled Jerusalem in the 160s B.C.E. as a result of events surrounding the Maccabean Revolt. The Ptolemaic rulers of Egypt gave Onias land in the Nile Delta and allowed him to establish a temple and military settlement; in this temple, run by the legitimate Jerusalemite high priestly family, the pre-Maccabean ritual of the Jerusalem temple was apparently carefully preserved.(99) Yet the material evidence for the settlement - a number of tombstones dating from a hundred and fifty years after its foundation - seems to show that by then the temple played little role in the life of the community. The epitaphs are all composed in Greek but, more strikingly, employ the conceptual language of Hellenistic popular morality and religion - apart from the fact that no Greek god but Hades is mentioned. There is no mention of identifiably Jewish institutions, no allusions to the Law or Jewish ritual, no painting or carving of the standard shofar (ram's horn) or lulav (palm-frond), and significantly (I would claim) not a single word of Hebrew.(100)

The other case, that of the earliest Christians, is far more complex, because no Christian writing comes from the (very brief) time when Christianity was an exclusively Jewish sect, and it is notoriously dangerous to work backwards from the Gospels, composed between 70 and 100 C.E., or even the Pauline Epistles, composed in the 50s C.E. Furthermore, not everyone would agree that the earliest Christians defined themselves not around unique legal interpretations and cultic practices (as did other, better-known, Jewish sects, like the Pharisees or Essenes), but around a myth of salvation, and this is not the place to argue the point. It is sufficient for the present purpose to observe that there is no evidence that Hebrew played any role at all in the life or self-definition of the earliest Christians, and that the Semitic statements attributed to Jesus in the Gospels are, to the extent that their language can be determined, in Aramaic, not Hebrew. These cases are at least consistent with my argument that Hebrew was part of an ideological package, because they seem to show that Jewish communities which had not adopted the two main parts of the package, the temple and the Torah, had no use for Hebrew either.

If the essential validity of my schematization is therefore tentatively confirmed, what does it show? That we should not take for granted a simple relationship between language and national identity, and that imperial domination had complex and varied effects on the symbolic worlds, on the self-definition, of the ruled. The Jews, for example, were created in large part by the administrative practices of the Persian and Macedonian rulers of the Near East: the rulers' decision to mediate their power through the Jerusalem temple and the Jewish Law played a crucial role in the consolidation of concepts and symbols which constituted Judaism.(101)

Furthermore, the case of the Jews might be helpful in making sense of the much less abundant evidence for the other dominated national, ethnic and tribal groups of the ancient eastern Mediterranean and Near East. It might, for example, help us account for the completely unexpected survival of Canaanite mythological lore in the Phoenician History of Philo of Byblos, a Greek writer of the second century C.E., or the durability of literary creativity in hieroglyphics in Egypt until the rise of Christianity.(102) In sum, it may help us to understand aspects of the native cultures of the Hellenistic and Roman Near East from the inside - cultures for the most part hidden from us beneath the shimmering, misleadingly uniform surface created by the conjunction of Greek urbanity and Roman might.

Seth Schwartz Jewish Theological Seminary of America, New York

1 It is conventional, and justified, to apply the word "Jew" to inhabitants of Judaea and their co-religionists elsewhere only after the fall of the Babylonian empire (539 B.C.E.).

2 Homer, Iliad, ii.816-77, at 1. 867.

3 To which I am indebted for some of the observations that follow: see S. Sauneron, "La differentiation des langages d'apres la tradition egyptienne", Bulletin de l'Institut Francais d'Archeologie Orientale, 1x (1960), pp. 31-41; C. Zaccagnini, "The Enemy in the Neo-Assyrian Royal Inscriptions: The 'Ethnographic' Description", in H.-J. Nissen and J. Renger (eds.), Mesopotamien und seine Nachbarn: 25e rencontre assyriologique internationale, 2 vols. (Berlin, 1982), ii, pp. 409-24.

4 Zaccagnini, "Enemy in the Neo-Assyrian Royal Inscriptions", pp. 410-11, p. 411 n. 10.

5 Cf. Ps. 119:1; Ezek. 3:5-6; Ps. 81:6.

6 For translations of and commentaries on this difficult passage, see H. G. M. Williamson, Ezra, Nehemiah (Word Biblical Commentary, xvi, Waco, 1985), pp. 391-402; J. Blenkinsopp, Ezra-Nehemiah (London, 1988), pp. 361-6. More problematic is Isa. 19:18: "On that day there will be five towns in the land of Egypt speaking the language of Canaan and swearing by Yahweh-of-Hosts".

7 Or, "As for you, ought it not be something unheard of to do. . .".

8 The Achaemenid governor of Samaria, of unknown ethnicity, but in all likelihood Israelite.

9 Blenkinsopp's and Williamson's failure (ad loc.) to notice the uniqueness and isolation of this verse is typical.

10 See E. Ullendorf, "The Knowledge of Languages in the Old Testament", Bull. John Rylands Lib., xliv (1961-2), pp. 462-3. For an account of the relationship between Hebrew and its "Canaanite" neighbours Phoenician, Ammonite, Moabite and Edomite, see A. Saenz-Badillos, A History of the Hebrew Language (Cambridge, 1993), pp. 38-44; for north-west Semitic (which consists of the Canaanite and Aramaic families) as a continuum, see ibid., p. 36.

11 My discussion here is indebted to J. Kugel, The Idea of Biblical Poetry (New Haven, 1981), esp. pp. 59-95.

12 For the gradual emergence of language as an important criterion of identity among the Greeks (or at least some of them), see most recently E. Hall, Inventing the Barbarian (Oxford, 1989), pp. 6-13; G. Nagy, "On the Death of Sarpedon", in C. Rubino and C. Shelmerdine (eds.), Approaches to Homer (Austin, 1983), pp. 189-91; G. Nagy, Best of the Achaeans (Baltimore, 1979), pp. 115-17.

13 Aside from the well-known satires of pagan religious practice and belief in 1 Kings, Deutero-Isaiah (that is, Isa. 40-55) and various psalms, it has been attractively suggested that the book of Jonah satirizes (N. B., not parodies) Yahwist prophecy: see M. Smith, Palestinian Parties and Politics that Shaped the Old Testament, 2nd edn (London, 1987), p. 123. Others see Jonah (the fictional character, not the book) as a deliberate parody of the prophet (not the book of) Jeremiah: H. Bloom, Ruin the Sacred Truths (Cambridge, Mass., 1989), pp. 22-4.

14 T. Cole, The Origins of Rhetoric in Ancient Greece (Baltimore, 1991), distinguishes eloquence from rhetoric (in a Greek context) in part on the basis of the latter's self-consciousness; but his restriction of this characteristic to post-Platonic writing has been widely criticized: see reviews by R. Martin, Classical Philology, 1xxxviii (1993), pp. 77-84; D. A. Russell, Fl Hellenic Studies, cxii (1992), pp. 185-6.

15 There are brief surveys of the language problem in J. J. Collins, The Apocalyptic Vision of the Book of Daniel (Ann Arbor, 1977), pp. 15-21; P. R. Davies, Daniel (Sheffield, 1985), pp. 35-9.

16 Achaemenid scribes, even modest ones, needed a high level of bilingualism, since from the time of Darius I all public documents were customarily written in Aramaic; the scribe was expected to be able to compose an Aramaic text from dictation in a different language and, conversely, to read out such a text in the language of his audience. Significantly, the Book of Nehemiah itself reports this practice: Ezra, a scribe by profession, had his associates read out the "Law of Moses" mephorash, that is, a Hebrew text was read in Aramaic - or vice versa: Neh. 8:1-8. Like Ezra himself, many Jewish writers (and readers?) of the fifth-fourth centuries B.C.E. must have come from such circles. See M. Dandamaev and V. Lukonin, The Culture and Social Institutions of Iran (Cambridge, 1989), pp. 113-16; J. Naveh and J. C. Greenfield, "Hebrew and Aramaic in the Persian Period", in The Cambridge History of Judaism, i, ed. W. D. Davies and L. Finkelstein (Cambridge, 1984), pp. 115-16.

17 The Aramaization of Phoenicia may have been slower and less thorough than that of Palestine, Syria and Mesopotamia: see E. Lipinski, "Geographie linguistique de la Transeuphratene a l'epoque Achemenide", Transeuphratene, iii (1990), pp. 105-6.

18 For a good recent survey, see J. C. Greenfield, "Aramaic in the Achaemenian Empire", in The Cambridge History of Iran, 7 vols. (Cambridge, 1968-91), ii, ed. I. Gershevitch, pp. 698-713.

19 See H. Tadmor, "The Aramaization of Assyria: Aspects of Western Impact", in Nissen and Renger (eds.), Mesopotamien und seine Nachbarn, ii, pp. 449-70. In Mesopotamia, though, the Assyrian and Babylonian kings had settled vast numbers of "Aramaeans" and other western Semites, whereas many settlers in Syria-Palestine must have been of East Semitic and Iranian-speaking background: see B. Oded, Deportations and Deportees in the Neo-Assyrian Empire (Wiesbaden, 1979), pp. 26-32.

20 For surveys of the question, see N. Waldman, The Recent Study of Hebrew (Cincinnati, 1989), pp. 91-8; J. Barr, "Hebrew, Aramaic and Greek in the Hellenistic Age", in The Cambridge History of Judaism, ii, ed. W. D. Davies and L. Finkelstein (Cambridge, 1989), pp. 79-114; New Documents Illustrating Early Christianity, ed. G. H. R. Horsley, 5 vols. (North Ryde, 1981-9), v, pp. 19-26; E. Schurer, The History of the Jewish People in the Age of Jesus Christ, rev. and ed. G. Vermes et al., 3 vols. (Edinburgh, 1973-87), ii, pp. 20-8. M. H. Segal's most accessible statement of his position is to be found in his A Grammar of Mishnaic Hebrew (Oxford, 1927), pp. 1-20.

21 See S. Fraade, "Rabbinic Views on the Practice of Targum, and Multilingualism in the Jewish Galilee of the Third-Sixth Centuries", in L. Levine (ed.), The Galilee in Late Antiquity (Cambridge, Mass., 1992), pp. 253-84; for a balanced account of the emergence of the post-Dead Sea Scrolls consensus, see J. Barr, "Which Language Did Jesus Speak?", Bull. John Rylands Lib., 1 (1970-1), pp. 9-29.

22 Of the hundreds of ostraca from Masada, the Hebrew prepositive article appears six times in epithets and nicknames, and Hebrew forms appear eleven times in indications of filiation: see J. Naveh, Masada, i: The Aramaic and Hebrew Ostraca and Jar Inscriptions (Jerusalem, 1989), pp. 8-9. I discuss the Hebrew jar inscriptions at pp. 33-4 below. Of the 177 Jerusalem ossuary inscriptions found in Corpus inscriptionum judaicarum, ed. J.-B. Frey, 2 vols. (Rome, 1936-52), ii, nos. 1210-1387, 24 use Hebrew forms; a similar proportion is also found in the "Dominus Flevit" ossuaries, published after the appearance of Frey's corpus: Gli Scavi del Dominus Flevit, ed. B. Bagatti and J. T. Milik, i (Pubblicazioni dello Studium Biblicum Franciscanum, xiii, Jerusalem, 1958), pp. 70-109. In all cases this is, as in the Masada texts, a matter of prepositive articles in epithets, and use of the Hebrew forms ben (son of), bat (daughter of), or 'eshet (wife of) instead of the Aramaic bar, berat, 'atat. Longer texts, both in Jerusalem and at Masada, are, with the sole exception of the Masada jar inscriptions, in Aramaic (especially at Masada) or Greek (especially in Jerusalem). There may be some justice to Naveh's argument that precisely such small matters as onomastic habits reflect spoken language more accurately than do longer texts, but it is not self-evident that this is so.

23 E. Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll (I Q Isa(a) (Studies on the Texts of the Deserts of Judah, vi, Leiden, 1974), pp. 8-15, 89-95.

24 That is, the situation is unlikely to have been diglossic in the strict sense, for diglossia involves the use in certain public situations of a language which is not actually spoken as a "home" language by anyone, though the two languages are closely related and the high language may be widely understood: see C. A. Ferguson, "Diglossia", Word, xv (1959), pp. 336-7. It was, however, certainly diglossic in the looser sense common in post-Ferguson discussion, in that it was characterized by the use of two or more languages - or dialects, or registers of one language - in various contexts. For discussion, see A. Hudson, "Diglossia: A Bibliographic Review", Language in Society, xxi (1992), pp. 617-18. The utility of so broad a conception of the phenomenon as a principle of social classification is not obvious, for diglossia in this sense is a feature of all but the simplest societies.

25 My characterization of the emerging nation-state is indebted to E. Gellner, Nations and Nationalism (Oxford, 1983), pp. 19-38. See Ferguson, "Diglossia", p. 331, on the importance of schools in spreading the high language. Ferguson's own description of the conditions in which diglossia occurs is quite mistaken: ibid., p. 338. Fishman provides a more satisfactory account: J. Fishman, "Bilingualism With and Without Diglossia; Diglossia With and Without Bilingualism", Jl Social Issues, xxiii (1967), pp. 29-38.

26 The multilingual model, which often complements the diglossic model, supposes that many or most Judaeans could get by in two or even three spoken languages: see, for example, B. Spolsky, "Jewish Multilingualism in the First Century", in J. Fishman (ed.), Readings in the Sociology of Languages (Leiden, 1985), pp. 35-50. No doubt among some male members of the upper class trilingualism of a limited sort (in Greek, Aramaic and Hebrew) was not uncommon, and certainly there are no grounds for excluding the possibility of widespread multilingualism a priori. It is a well-attested phenomenon in other societies; note, for example, the case of the Indians of the Vaupes, in Colombia, who are reported all to speak at least three languages fluently: J. Jackson, "Language Identity of the Vaupes Indians", in R. Bauman and J. Sherzer (eds.), Explorations in the Ethnography of Speaking, 2nd edn (Cambridge, 1989), pp. 55-6. But there is no good reason to think that most Jews in ancient Palestine were multilingual.

27 Though not invariably; Rabin, Rosen and Yadin, to mention only the most important, at least acknowledged the role of Hebrew in the ideology of the various Judaean revolts: C. Rabin, "Hebrew and Aramaic in the First Century", in S. Safrai and M. Stern (eds.), The Jewish People in the First Century, 2 vols. (Compendium rerum iudaicarum ad Novum Testamentum, i, Assen, 1974-6), ii, pp. 1007-39; H. B. Rosen, "Die Sprachsituation im romischen Palastina", in Die Sprachen im romischen Reich der Kaiserzeit (Beihefte der Bonner Jahrbticher, xl, Cologne, 1980), pp. 215-39; Y. Yadin, Bar Kokhba (London, 1971), p. 124; and see pp. 27-8 below. But none of these scholars ever attempted to work out the less obvious implications of Hebrew's significance in revolutionary propaganda. James Barr's consistent denial of the obvious may be mentioned as an extreme example of the neglect (in this case wilful) of ideological factors: Barr, "Hebrew, Aramaic and Greek in the Hellenistic Age".

28 For Akkadian, see the discussion in S. Sherwin-White and A. Kuhrt, From Samarkhand to Sardis: A New Approach to the Seleucid Empire (London, 1993), pp. 149-61; for Egyptian, see the texts collected in M. Lichtheim, Ancient Egyptian Literature, 3 vols. (Berkeley, 1975-80), iii; for Phoenician, see F. Millar, "The Phoenician Cities: A Case Study in Hellenisation", Proc. Cambridge Philol. Soc., xxix (1983), pp. 55-71, esp. pp. 60-3 (isolated Phoenician words and letters on coins to the end of the second century C.E.).

29 See Griechische Papyri im Museum des Oberhessischen Geschichtsvereins zu Giessen, i, ed. E. Kornemann and P. Meyer (Leipzig and Berlin, 1910-12), no. 99; F. Zucker, Doppelinschrift spatptolemaischer Zeit aus der Garnison von Hermopolis Magna (Berlin, 1938), p. 13.

30 For a survey, see S. Cohen, From the Maccabees to the Mishnah (Chicago, 1987), pp. 64-5; for more tietail, see Schurer, History of the Jewish People, rev. and ed. Vermes et al., iii, pt 1, pp. 187-96; ibid., pp. 452-5 (on the Qumran "Thanksgiving Hymns").

31 The Hellenistic Akkadian administrative documents, discussed by Sherwin-White and Kuhrt, From Samarkhand to Sardis, pp. 149-61, which demonstrate that in the Seleucid Empire it took at least a century for Greek to be established as the exclusive language of administration, may at least hint at what is missing.

32 See Dandamaev and Lukonin, Culture and Social Institutions of Iran, pp. 112-13; Greenfield, "Aramaic in the Achaemenian Empire", p. 698; K. Hopkins, "Conquest by Book", in Literacy in the Roman World (Jl Roman Archaeology, suppl. iii, Ann Arbor, 1991), pp. 1313-58.

33 See A. L. Oppenheim, Letters from Mesopotamia (Chicago, 1967), pp. 42-53; the important remarks of E. J. Bickerman(n), The Jews in the Greek Age (Cambridge, Mass., 1988), pp. 51-65; S. Brock, "Three Thousand Years of Aramaic Literature", Aram, i (1989), pp. 17-18. On the association of Aramaic with papyrus and leather, see Greenfield, "Aramaic in the Achaemenian Empire", p. 698.

34 For discussion and literature, see P. Briant, "The Seleucid Kingdom and the Achaemenid Empire", in t'. Bilde et al. (eds.), Religion and Religious Practice in the Seleucid Kingdom (Studies in Hellenistic Civilization, i, Aarhus, 1990), pp. 53-4, 58-60.

35 For the text and translation of this account, see Die sogenannte demotische Chronik des Pap. 215 der Bibliotheque Nationale zu Paris, ed. W. Spiegelberg (Leipzig, 1915), pp. 30-2; Dandamaev and Lukonin, Culture and Social Institutions of Iran, p. 125. The laws were officially published in Demotic, for the use of native judges, and in Aramaic ("the writing of Ashur"), for Persian officials.

36 That construction of the temple and imposition of the Law were not uncontested is demonstrated by the partisan accounts in the books of Ezra and Nehemiah. On the connection between the Demotic law and the Law of Moses, see Bickerman(n), Jews in the Greek Age, pp. 29-32; for a discussion of the Achaemenid context of the Pentateuch, see J. Blenkinsopp, The Pentateuch (London, 1992), pp. 239-42.

37 Josephus, Autobiography, 134-5.

38 See, for example, Deut. 17:8-13.

39 See S. Schwartz, Josephus and Judaean Politics (Leiden, 1990), pp. 69-70; E. J. Bickerman(n), Studies in Jewish and Christian History, 3 vols. (Leiden, 1976-86), ii, pp. 69-74; Bickerman(n), Jews in the Greek Age, pp. 140-7.

40 Though various versions of Darius's law-code may, if it is somehow related to a Demotic law-code of the third century B.C.E. found in Hermopolis, have survived into the high Roman Empire, and may have functioned in some circles as a kind of national symbol. On the "Hermopolis code", see G. Mattha, The Demotic Legal Code of Hermopolis West (Cairo, 1975); a fragment of a Greek translation of this code, probably made in the third century B.C.E. but copied in the late second century C.E., is published in The Oxyrhynchus Papyri, xlvi, ed. J. R. Rea (London, 1978), no. 3285; the suggestion about the ideological function of the text in the Roman period is made by J. Meleze-Modrzejewski, Les Juifs d'Egypte (Paris, 1991), pp. 89-91.

41 The only certain exception is the Seleucid emperor Antiochus IV, who suspended the Law between 167 and 164 B.C.E. The emperors Ptolemy II, III and IV, who ruled in Palestine 285-200 B.C.E., may have done nothing either to support or to undermine the native institutions of Judaea.

42 On the Ptolemies and Egyptian law, see H. J. Wolff, "Law in Ptolemaic Egypt", in A. E. Samuel (ed.), Essays in Honor of C. B. Welles (Amer. Studies in Papyrology, i, New Haven, 1966): pp. 67-77. In the third century B.C.E., the Ptolemies empowered Egyptian priests to judge according to Egyptian laws, but gave neither the priests nor the laws exclusive jurisdiction over any inhabitant of Egypt: the Egyptian law was voluntary, an imperial concession intended to spare the feelings of native subjects. In the second century B.C.E., the system collapsed.

43 For an account of this process, see Smith, Palestinian Parties and Politics, pp. 121-4.

44 For a brief survey, see ibid., pp. 68-70.

45 For an example, see T. Ilan, "Premarital Cohabitation in Ancient Judaea: The Evidence of the Babatha Archive and the Mishnah (Ketubbot 1.4)", Harvard Theol. Rev., lxxxvi (1993), pp. 247-64.

46 See Smith, Palestinian Parties and Politics, pp. 130-43. Alternatively, "the Law" may be thought of as a large and unsystematic body of practice, customary and prescribed, including both the Pentateuch and other normative corpora - a body in which at some point, perhaps not much earlier than 250 or 200 B.C.E. (and even then not for all Jews), the Pentateuch came to dominate: see J. Blenkinsopp, Wisdom and Law in the Old Testament (Oxford, 1983), pp. 74-5. Such a formulation is admirably complex but seems to me to downplay, perhaps wrongly, the role of the Achaemenid emperors.

47 See 1 Macc. 1:41-64; 2 Macc. 6. The most important single work on the background of the persecution and revolt is still E. J. Bickerman(n), Der Gott der Makkabaer (Berlin, 1937); more recent literature is clearly and briefly discussed in D. Harrington, The Maccabean Revolt: Anatomy of a Biblical Revolution (Wilmington, 1988). For the effect of the persecution and revolt on Judaism, see S. Cohen, "Religion, Ethnicity and 'Hellenism' in the Emergence of Jewish Identity in Maccabean Palestine", in Bilde et al. (eds.), Religion and Religious Practice in the Seleucid Kingdom, pp. 204-23.

48 See M. Stern, Greek and Latin Authors on Jews and Judaism, 3 vols. (Jerusalem, 1974-84), i, passim.

49 There is, surprisingly, no good single treatment of the history of the temple; for a general survey, see Schurer, History of the Jewish People, rev. and ed. Vermes et al., ii, pp. 227-313. The temple as an economic "black hole" is well discussed by M. Goodman, The Ruling Class of Judaea (Cambridge, 1987), pp. 51-75.

50 On the coins, see Y. Meshorer, Ancient Jewish Coinage, 2 vols. (New York, 1982), i, pp. 14-17; L. Mildenberg, "Yehud-Munzen", in H. Weippert, Palastina in vorhellenistischer Zeit (Munich, 1988), pp. 721-8; D. Barag, "A Silver Coin of Yohanan the High Priest and the Coinage of Judaea in the Fourth Century B.C.E.", Israel Numismatic Jl, ix (1986-7), pp. 4-21. On the seals, bullae, etc., see the survey in E. Stern, Material Culture in the Land of the Bible in the Persian Period (Warminster, 1982), pp. 202-13.

"Palaeo-Hebrew" is the script, closely related to Phoenician, used for Hebrew during the Israelite and Judahite monarchies. Starting in the sixth century B.C.E., Aramaic script began to replace Palaeo-Hebrew for Hebrew texts, though scattered use of the old script continued to the first century C.E. among Judaeans; Samaritans still use Palaeo-Hebrew script. The best and most intelligent discussion of late Palaeo-Hebrew and its significance - and one of the best general discussions of the language situation in ancient Palestine - is in a Hebrew book: J. Naveh, 'AI Heres Vegome': Ketovot 'Aramiyot Ve'ivriyot Mitequfat Bayit Sheni, Mishnah Vetalmud [On Sherd and Papyrus: Aramaic and Hebrew Inscriptions from the Second Temple, Mishnaic and Talmudic Periods] (Jerusalem, 1992), pp. 11-36.

51 See J. Goldstein, I Maccabees (Garden City, 1976), p. 14.

52 See 2 Macc. 7, 12:37.

53 The fullest collection is to be found in Meshorer, Ancient Jewish Coinage, i. He dates the first Hasmonean coinage to Alexander's reign, but recent archaeological discoveries have made it overwhelmingly likely that John was the first to mint: see D. Barag, "Jewish Coins in Hellenistic and Roman Time", in T. Hackens et al. (eds.), A Survey of Numismatic Research, 1985-1990, 2 vols. (Int. Soc. Professional Numismatists, special pubns, xii, Brussels, 1991), i, p. 106. Hasmonean coins were all small-denomination bronze, intended only for local use. Seleucid royal and municipal silver coins continued to circulate extensively in Hasmonean Palestine.

54 I will refrain from speculation here, and simply point out that the coins of the Hasmoneans' successor, Herod, were inscribed exclusively in Greek, with a gradual return, especially under his successors, to "normal" (i.e., Greek-style) iconography, including portraits of monarchs.

55 The survival of one or two Judaean texts from the first century C.E. in the old script, and the use of a few Palaeo-Hebrew letters as stone-cutters' marks and on food-rationing tags from Masada no more demonstrates general literacy in Palaeo-Hebrew than the use of Greek letters by mathematicians proves general literacy in Greek nowadays: see Naveh, 'Al Heres Vegome', pp. 11-23. On the difficulties the engravers of the coins of both revolts had with the Palaeo-Hebrew legends, see L. Mildenberg, The Coinage of the Bar Kokhba War (Aarau, 1984), p. 22.

The standard collections of the revolutionary coins are the outdated, but not yet replaced, work by L. Kadman, The Coins of the Jewish War of 66-73 C.E. (Tel Aviv, 1960) - see especially his comments on the symbols on the coins, pp. 83-95; Mildenberg, Coinage of the Bar Kokhba War.

56 There is some evidence that Bar Kokhba insisted on the use of Hebrew in his revolutionary administration: see the discussion in Schurer, History of the Jewish People, rev. and ed. Vermes et al., ii, pp. 27-8. In contrast, the documents produced by the rebels of the Great Revolt encamped at Masada are overwhelmingly Aramaic: see n. 22 above. On the language of the Bar Kokhba coins, see Mildenberg, Coinage of the Bar Kokhba War, pp. 69-72.

57 Josephus, Jewish War, ii.228-31.

58 This is not the only possible interpretation of the passage, and it may be worth remembering that substantial collections of books could be found in some Egyptian villages; however, Greek books in Egypt were written on (relatively inexpensive) papyrus while, if we may extrapolate from the Dead Sea Scrolls, Jewish holy books, as early as the second and first centuries B.C.E., were written on parchment: cf. Mishnah, Yadayim, iv.5 and Megillah, ii.1. See M. Goodman, "Sacred Scripture and 'Defiling the Hands'", Jl Theol. Studies, new ser., xli (1990), p. 105; M. Haran, "Book-Scrolls at the Beginning of the Second Temple Period: The Transition from Papyrus to Skins", Hebrew Union Coll. Annual, liv (1983), pp. 111-33. Nevertheless, the implication of Josephus's Against Apion, ii. 175, that Torah scrolls could normally be found in Jewish settlements, if it is not pure idealization, may reflect conditions in the Diaspora, not Palestine: Schwartz, Josephus and Judaean Politics, p. 23.

59 Before 70 C.E., archaeological evidence is all from fortresses, and literary evidence mostly concerns cities. It is not clear how seriously one should take the references to synagogues in the villages of Nazareth and Capernaum in the Gospels (Mark 6:2 and parallels; Mark 1:21 and parallels), composed after 70 C.E. in diasporic environments. John, the only Gospel possibly composed in Palestine, mentions synagogai far less frequently than the others, and the word may mean "assemblies" or "gatherings", not "synagogues" (e.g., John 18:20): see W. Schrage, "Synagoge", in G. Kittel and G. Friedrich (eds.), Theological Dictionary of the New Testament, trans. G. W. Bromiley, 10 vols. (Grand Rapids, 1964-77), vii, pp. 830-2. On the synagogue in general, see Cohen, From the Maccabees to the Mishnah, pp. 111-15; L. Levine, "The Second Temple Synagogue", in L. Levine (ed.), The Synagogue in Late Antiquity (Philadelphia, 1987), pp. 7-31. Even the Mishnah, compiled c. 200 C.E., apparently supposes that villagers would have to travel to the nearest market town to hear the reading of the Scroll of Esther on the festival of Purim: Megillah, i.1, with the comments of Rashi ad loc.

60 For a brief discussion, see Cohen, From the Maccabees to the Mishnah, pp. 120-3.

61 See A. Baumgarten, "The Torah as a Public Document in Judaism", Studies in Religion, xv (1985), pp. 17-24.

62 Book of Jubilees, xii.25-7.

63 A good survey may be found in Schurer, History of the Jewish People, rev. and ed. Vermes et al., iii. The standard collections of the material are Apocrypha and Pseudepigrapha of the Old Testament, ed. R. H. Charles, 2 vols. (Oxford, 1913); Old Testament Pseudepigrapha, ed. J. H. Charlesworth, 2 vols. (Garden City, 1983-5). For the Dead Sea Scrolls, see G. Vermes, The Dead Sea Scrolls in English, 4th edn (London, 1995).

64 See Cohen, From the Maccabees to the Mishnah, p. 192.

65 See M. Hengel, Judaism and Hellenism, 2 vols. (London, 1974), pp. 138-54 (who characteristically goes too far); on Eupolemus, see Schiirer, History of the Jewish People, rev. and ed. Vermes et al., iii, pt 1, pp. 517-21.

66 The most successful account of rabbinic curatorial ideology is S. Cohen, The Three Crowns: Structures of Communal Politics in Early Rabbinic: Jewry (Cambridge, 1990).

67 Though what may be rabbinic tombstones continued to be inscribed in Hebrew after 200 C.E. See N. Avigad's comments in B. Mazar et al., Beth Shearim, 3 vols. (New Brunswick, 1971), iii, pp. 52-65; Naveh, 'Al Heres Vegome', p. 25.

68 Though, as usual in rabbinic literature, contradictory opinions are expressed too - e.g., that the angels understand only Aramaic: Babylonian Talmud, Shabbat, 12b. Surveys of rabbinic comments about Hebrew may be found in K. Treu, "Die Bedeutung des griechischen fur die Juden im romischen Reich", Kairos, xv (1973), pp. 132-4; Fraade, "Rabbinic Views on the Practice of Targum". For an attempt to construct a rabbinic theory of language, mainly from the late antique midrash Genesis Rabbah, see H. Eilberg-Schwartz, "Who's Kidding Whom? A Serious Reading of Rabbinic Word Plays", Jl Amer. Acad. Religion, lv (1987), pp. 765-88. In many of the passages, Hebrew is expressly promoted at the expense of Aramaic. Furthermore, the later rabbinic documents, in which the comments about Hebrew are mostly found, themselves move between Hebrew and Aramaic with an unconcern comparable to that of the compilers of Ezra-Nehemiah, though to what extent this editorial insouciance reflects actual rabbinic practice I have not tried to determine. It may be worth noting that by late antiquity, when the rabbinic texts were redacted, distinctively Jewish dialects of Aramaic had emerged which their speakers may have thought as suitable as Hebrew for expressions of Jewish identity. Also striking and difficult to evaluate is that midrashic texts often provide Greek "etymologies" for biblical Hebrew words - admittedly, an exegetical strategy. In any case, it is probable that Jewish self-consciousness about Hebrew never in antiquity produced the same sorts of results as the Greek (for example, systematic grammar and prosody). These developed only in the Middle Ages.

69 Genesis Rabbah, 18; Palestinian Targum, ad Gen. 11:1. Hebrew also appears as the language of creation in the passage from the Book of Jubilees quoted at p. 30 above. For discussion, see L. Ginzberg, Legends of the Jews, 7 vols. (Philadelphia, 1968), v, pp. 205-6; it seems a straightforward inference from the Genesis account, in which God creates the universe by speaking in Hebrew (Gen. 1:2). Two ancient Jewish works composed in Greek take it for granted that Greek was the language of creation: see Sibylline Oracle, iii. 24-6; 2 Enoch, xxx.13.

7o Babylonian Talmud, Sotah, 49b; Palestinian Talmud, Megillah, i.11, 71b.

71 Most of these texts are now available in translation: for Sefer Harazim, see Sepher HaRazim: The Book of Mysteries, trans. M. Morgan (Chico, 1983); the Hekhalot books are now being translated into German under the supervision of P. Schafer: Ubersetzung der Hekhalot-Literatur, ii-iv (Texte und Studien zum antiken Judentum, xvii, xxii, xxix, Tubingen, 1987-91); for the magical papyri, see Greek Magical Papyri in Translation, including the Demotic Spells, trans. H. D. Betz (Chicago, 1986). Isolated words and phrases of unquestionably Hebrew origin are common, and some long strings of so-called nomina barbara may be garbled transcriptions of Hebrew: G. Scholem, Jewish Gnosticism, Merkabah Mysticism and Talmudic Tradition (New York, 1960), pp. 94-100. My characterization of Sefer Yezirah as linguistic speculation is indebted to a public lecture by Professor J. Dan of Jerusalem given at Cambridge in winter 1993.

72 The only pre-rabbinic attestation of the term is in a recently published Dead Sea Scroll fragment officially designated 4Q 464. This text is dated by its editors to the first century C.E., but its character is unclear: see E. Eshel and M. Stone, "Leshon Haqodesh Be'aharit Hayamim Le'or Qeta' Miqumran" [Leshon haqodesh in the End of Days in the Light of a Fragment from Qumran], Tarbiz, lxii (1993), pp. 169-77.

73 It is formed on the paradigm of such biblical phrases as "sheqel [a weight of silver] haqodesh", "hin [a liquid measure] qodesh" ("sheqel / hin according to the temple standard"): see J. Greenfield, "Languages of Palestine, 200 B.C.E.-200 C.E.", in H. Paper (ed.), Jewish Languages: Theme and Variations (Cambridge, Mass., 1978), p. 149. Shinan's suggestion that leshon haqodesh and its occasional Aramaic equivalent lishan bet qudsha' mean "synagogue language" seems to fall on the fact that in no extant source is the synagogue called haqodesh or bet qudsha': A. Shinan, "'Lishan Bet Qudsha" Betargumim Ha'aramiyim Latorah" [Lishan bet qudsha' in the Aramaic Targums to the Torah], Bet Miqra, lxvi (1976), pp. 472-4.

74 See, for example, various passages in the Babylonian Talmud: Qiddushin, 2b; Avodah Zarah, 58b; Hullin, 137b.

75 See Naveh, Masada, i, nos. 441-61, with comments ad loc.; Naveh, 'Al Heres Vegome', p. 24.

76 See n. 94 below.

77 An interesting exception to this rule, perhaps related to Hebrew's status as "cosmic" language, as exemplified in such works as Sefer Harazim, are the inscriptions labelling the signs of the zodiac on the synagogue floor mosaics of Byzantine Palestine, collected in B. Lifshitz, Donateurs et fondateurs dans les synagogues juives (Paris, 1967); J. Naveh, 'Al Psefas Va'even: Haketovot Ha'aramiyot Veha'ivriyot Mibbatei Keneset Ha'atiqim [On Stone and Mosaic: Aramaic and Hebrew Inscriptions from Ancient Synagogues] (Jerusalem, 1978). Many of the inscriptions in the latter probably date from the fifth to seventh centuries C.E., or even later, though their dating is problematic: ibid., pp. 4-6. An additional zodiac mosaic has recently been found at Sepphoris.

78 See M. Stern, Greek and Latin Authors on Jews and Judaism, ii, p. 596. A recent account, concentrating on the rather different (?) conditions of the third century, is M. Goodman, "The Roman State and the Jewish Patriarch in the Third Century", in Levine (ed.), Galilee in Late Antiquity, pp. 127-42; see also S. Cohen, "The Place of the Rabbi in Jewish Society of the Second Century", ibid., pp. 157-73 and, especially, L. Levine, "The Sages and the Synagogue in Late Antiquity", ibid., pp. 201-22, where the relations between Rabbi and synagogue are characterized much as here.

79 My understanding of the rise of the Diaspora is derived from Smith, Palestinian Parties and Politics, pp. 70-2. There is little evidence for this outside Egypt, but the demographic problem makes Smith's hypothesis (or something like it) inevitable.

80 For a summary, see Cohen, From the Maccabees to the Mishnah, pp. 108-11 (where the extent of communal autonomy is exaggerated); on the absence of the Personalitatsprinzip in ancient kingdoms and the essentially patronal character of the government's concession of some trappings of autonomy to non-citizen communities, see n. 42 above.

81 See Jewish Inscriptions of Graeco-Roman Egypt, ed. W. Horbury and D. Noy (Cambridge, 1992), nos. 3-5. The other Jewish inscriptions from the same Alexandrian necropolis are in Greek. For the ostraca, see P. Grelot, Documents arameens d'Egypte (Paris, 1972), nos. 12-13.

82 Similarly, Thompson concludes from the onomastic evidence that the Idumaean settlers at Memphis were "hellenized" (i.e., had come to use Greek for most purposes) in three generations: D. Thompson, Memphis under the Ptolemies (Princeton, 1988), p. 100. For the continued use of Semitic languages for some purposes by Judaean immigrants no demonstration is necessary; but it probably explains the Nash Papyrus, a Hebrew manuscript containing several biblical verses, allegedly written in the second century B.C.E.: see W. Albright, "A Biblical Fragment from the Maccabean Age: The Nash Papyrus", Jl Biblical Lit., lvi (193v), pp. 145-76. It may also explain a story recorded by Philo in Against Flaccus, 39: when the Jewish king, Agrippa I, visited Alexandria in 38 C.E., the pagan "mob" dressed a fool as the king and acclaimed him mockingly as marin - faux-Aramaic for "our lord". They were perhaps imitating Alexandrian Jews, though this is not Philo's understanding of the incident. On the question of the linguistic assimilation of the Egyptian Jews, see Meleze-Modrzejewski, Juifs d'Egypte, pp. 64-71.

83 Note that the Letter of Aristeas, par. 11, warns against this confusion, suggesting that it was common in Alexandria. For a discussion of the consensus on Philo's knowledge of Hebrew, see L. Grabbe, Etymology in Early Jcwish Interpretation: The Hebrew Names in Philo (Atlanta, 1988), p. 63. Josephus also occasionally conflates Hebrew and Aramaic, calling both "Hebrew". Both Philo and Josephus could, however, distinguish the two languages. James Barr takes Josephus's confusion as evidence that the Jewish Aramaic vernacular was heavily Hebraized and thus a distinctively "Jewish" language: Barr, "Which Language Did Jesus Speak?", pp. 25-9.

84 The main collection of this material, incomplete and now out of date, is Corpus inscriptionum judaicarum, ed. Frey; the first volume has been reprinted with an important introduction by B. Lifshitz (New York, 1975). For Roman inscriptions, Frey is now superseded by H. J. Leon, The Jews of Ancient Rome (Philadelphia, 1960), and by U. Fasola, "Le due catacombe ebraiche di Villa Torlonia", Rivista di archeologia cristiana, lii (1976), pp. 7-62; for Egypt, by Jewish Inscriptions of Graeco-Roman Egypt, ed. Horbury and Noy; for Cyrene, by Corpus judischer Zeugnisse aus der Cyrenaika, ed. G. Luderitz (Wiesbaden, 1983); for the western Roman Empire, excluding the city of Rome, by Jewish Inscriptions of Western Europe, ed. D. Noy (Cambridge, 1993).

85 Hellenomemphites (descendants of Archaic Greek settlers in Memphis) were still using Greek in the fourth century B.C.E., having arrived in the late sixth century, though they were in various ways Egyptianized and presumably in many cases bilingual: Thompson, Memphis under thc Ptolemies, pp. 95-7.

86 Diodorus Siculus, Library of History, xvii. 110.4-5; cf. Herodotus, Histories, vi.119.

87 See Millar, "Phoenician Cities".

88 See V. Colorni, "L'uso del greco nella liturgia del giudaismo ellenistico e la novella 146 di Giustiniano", Annali di storia del diritto, viii (1964), pp. 18-80.

89 For a survey of modern views on the origins of the Septuagint, see S. Jellicoe, The Septuagint and Modern Study (Oxford, 1968), pp. 59-73.

90 See E. J. Bickerman(n), "Zur Datierung des Pseudo-Aristeas", in his Studies in Jewish and Christian History, i, pp. 123-36.

91 "Aristeas" seems strangely unaware that ten of the twelve tribes had disappeared in the eighth century B.C.E.; the reference to twelve tribes is odd even in a work of fiction.

92 The miraculous nature of the translation is played up in later Jewish and Christian literature, but is already present in Aristeas: see the list of post-Aristeas accounts in H. B. Swete, An Introduction to the Old Testament in Greek (Cambridge, 1914), pp. 12-14. For a recent attempt partially to rehabilitate the traditional account, see Meleze-Modrzejewski, Juifs d'Egypte, pp. 84-90. But the traditional account is problematic because it fails to account for the widespread acceptance of the Septuagint as a substitute for the Hebrew text. The continuing insistence on the miraculous and perfect nature of the translation suggests that the Septuagint long remained controversial among the Jews, as it is known to have later become among Christians: see S. Rebenich, "Jerome: The 'vir trilinguis' and the 'Hebraica veritas'", Vigiliae Christianae, xlvii (1993), pp. 50-77.

93 See Cohen, From the Maccabees to the Mishnah, pp. 55-9.

94 The fullest repertoire of material is E. R. Goodenough, Jewish Symbols in the Greco-Roman Period, 13 vols. (New York, 1953-68), esp. vols. ii-iii; but see the comments and corrections of M. Smith, "Goodenough's Jewish Symbols in Retrospect", Jl Biblical Lit., lxxxvi (1967), pp. 53-68.

95 I owe this last observation to William Horbury. Some Jewish communities in the Latin-speaking West may have retained Greek as a liturgical language, though it is impossible to be certain: Colorni, "Uso del greco nella liturgia del giudaismo ellenistico".

96 I refrain from going into greater detail because it is often difficult to tell whether a given piece of ancient Jewish literature was composed in Palestine or the Diaspora. It is certain, as suggested at p. 30 above, that Palestinians wrote in Greek, and possible that Semitic books were written by Jews of the Greek-speaking Diaspora, though it is impossible to specify which.

97 The patronal control of the Diaspora communities is observed by Cohen, From the Maccabees to the Mishnah, pp. 111-15; for more detailed discussion, see T. Rajak, "The Jewish Community and its Boundaries", in J. Lieu, J. North and T. Rajak (eds.), The Jews among Pagans and Christians in the Roman Empire (London, 1992), pp. 9-28.

98 See Corpus papyrorum judaicarum, ed. V. Tcherikover, 3 vols. (Cambridge, Mass., 1957-64), i, pp. 94-6.

99 Josephus, Jewish Antiquities, xiii. 62-73.

100 Jewish Inscriptions of Graeco-Roman Egypt, ed. Horbury and Noy, nos. 29-105.

101 Admittedly, this claim is reductionist but, I think, justified not only per se, but also as a corrective to the inflationary rhetoric of most discussions of the subject.

102 On Philo of Byblos, see H. W. Attridge and R. A. Oden, Jr, Philo of Byblos: The Phoenician History (Catholic Biblical Quart., monograph ser., ix, Washington, D.C., 1981), esp. pp. 3-9; on hieroglyphics, see Lichtheim, Ancient Egyptian Literature, iii.
COPYRIGHT 1995 Oxford University Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1995 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Schwartz, Seth
Publication:Past & Present
Date:Aug 1, 1995
Previous Article:The land-family bond in England.
Next Article:The culture of children in medieval England.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |