Greek education and cultural identity in Greek-speaking Judaism: The Jewish-Greek historiographers

Abstract

The style of the Jewish-Greek historiographers Eupolemus and Demetrius has often been evaluated as “bad Greek.” This is generally seen as evidence of their lack of education. The negative views on the language of Demetrius and Eupolemus are illustrative of a broader issue in the study of Hellenistic Judaism: language usage has been a key element in the discussion on the societal position of Jews in the Hellenistic world. In this article, I assess the style of the historiographers in the context of post-classical Greek, and conclude that their language reflects standard Hellenistic Greek. The linguistic analysis then becomes a starting point to reflect on the level of integration of Jews in the Greek-speaking world as well as to consider the nature of Jewish multilingualism in the late Second Temple period.

Introduction

The Greek writing style of Demetrius the Chronographer, who is traditionally situated in the third century B.C.E. and considered the oldest Jewish-Greek author known to us, ¹ is described as “shapeless” ² and “awkward.” ³ We find even more negative evaluations of the work of Eupolemus, a Jewish historian dated to the second century B.C.E. ⁴ His style is said to be “inferior,” ⁵ even “miserable,” with “clumsy sentence construction.” ⁶ The basis for these harsh criticisms is the presumed use of Hebraistic constructions, ⁷ which are generally understood to be caused by bilingual interference. Especially for Eupolemus, Greek is assumed to be his second language. ⁸ Scholars often suggest that Demetrius’s and Eupolemus’s lack of literary prowess is the result of a limited Greek-language education, ⁹ or reflective of a Jewish dialect. Ben Zion Wacholder even states that “only the presumed existence of a distinct Judaeo-Greek dialect renders Eupolemus’s Greek tolerable.” ¹⁰

The negative views on the language of Demetrius and Eupolemus are illustrative of a broader issue, namely that of the evaluation of the level of Greek used by Jews in the late Second Temple Period. The style of Jewish writings in Greek has often been dismissed: the Greek of the Septuagint, too, for example, was long seen as substandard. Its character was explained mainly on the basis of either a specific Jewish-Greek dialect or a poor mastery of Greek on the part of the translators. ¹¹ In this way, language usage has been a key element in the discussion on the societal position of Jews in the ancient world: the supposed low-level Greek used by Jews in the Hellenistic world has been linked primarily to their social status as outsiders.

While the hypothesis of a Jewish-Greek dialect can no longer be maintained, ¹² the question remains whether and particularly to what extent Jews in the Hellenistic era would have been able and willing to participate in the Greek education system. This matter relates to our conceptualization of the cultural dynamics between Hellenism and Judaism: it is a question both of the level of integration of Jews in the Greek-speaking world and of the consequences of the use of Greek for Jewish cultural identity.

There has been a move in recent scholarship toward understanding the style of the Septuagint as reflective of post-classical Greek, and appreciating the Septuagint translators as linguistically proficient. ¹³ This gives rise to the question whether the same could be true for Jewish-Greek writers such as Eupolemus and Demetrius. In this contribution, I reassess Demetrius’s and Eupolemus’s language usage and offer some considerations on their position as Greek-speaking Jewish authors.

The evidence, I: features of post-classical Greek

We may first reconsider the evidence cited in support of the claim that Demetrius and Eupolemus wrote substandard, Hebraistic Greek.

Jacob Freudenthal, in his seminal work on Alexander Polyhistor, cites two clauses from the same paragraph (PE 9.34.2) to illustrate that Eupolemus’s style is “incorrect and tasteless”: περὶ δὲ ὧν γράφεις μοι, περὶ τῶν κατὰ τοὺς λαοὺς τοὺς παρ’ ἡμῖν “concerning the matters about which you wrote to me, specifically your request for some of our people,” and ὑπὲρ ὧν ἂν αὐτὸν ἐρωτήσῃς τῶν ὑπὸ τὸν οὐρανὸν πάντων “any question under heaven which you might ask him.” ¹⁴ While Freudenthal does not make it explicit what the stylistic issue is, these examples seem to pertain to the use of the resumptive pronoun in a relative construction. Wacholder includes this feature and adds the following elements as indicative of Eupolemus’s poor style: the use of the verb ποιέω (with reference to PE 9.34.6 without further comment); the frequent use of the connector καί (considered to be reflective of the paratactical style of Hebrew); and the “abused objective” ὅλος (i.e., used in the same way as the Hebrew uses כל‎). ¹⁵

As evidence of low-level Greek in Demetrius’s writings, the author’s restricted vocabulary and use of ususual constructions are often mentioned, generally without citation of concrete examples. ¹⁶ We can, however, identify elements in his writings similar to those mentioned with regard to Eupolemus’s supposed stylistic poverty, such as the use of the resumptive pronoun (e.g., in PE 9.21.5). In other words, for both authors, reference is made to a limited number of features that are singled out because they somehow, often intuitively, seem striking to a modern reader. These evaluations, however, rely on modern assumptions about the grammatical and stylistic stability of Greek across time and space. ¹⁷ As the fields of classical and biblical studies have gained a more nuanced understanding of the nature of Greek in the post-classical period, we are invited to re-evaluate the evidence and compare it to contemporary literary and non-literary sources.

Let us first look at Eupolemus’s use of ποιέω. Wacholder describes its use in PE 9.34.6 as an example of “pidgin Greek.” ¹⁸ In this passage, the verb is used repeatedly in the context of manufacturing. In Greek, ποιέω can be applied to the construction of anything material, such as buildings, as well as to a smith’s work (LSJ—see, for example, Homer, Il. 1.608; 7.222). Eupolemus uses ποιέω precisely in these two contexts. While one may have an opinion about the aesthetic nature of the repetition of ποιέω in this fragment (especially throughout P.E. 9.34.6–11), one cannot maintain that its usage in this passage would be indicative of bilingual interference, given that it lies within the standard semantic range of the verb. Holladay mentions another usage of ποιέω that he considers “crude” or “unusual,” namely, the construction καλῶς ποιέω + participle, such as at PE 9.34.3. ¹⁹ This construction is, in fact, common in contemporary papyri, as we see in BGU 6 1238 (Arsinoïtes, third century B.C.E.); BGU 10 1996 (location unknown, 241 B.C.E.); BGU 8 1786 (Herakleopolites, 50 B.C.E.), to mention but a few. These parallels indicate that Eupolemus’s usage of the construction is in no way abnormal.

Second, an increased use of parataxis is also typical of post-classical Greek. ²⁰ While a high frequency of καί can be the result of Semitic interference in translated texts, we cannot conclude that a Greek text in which parataxis is dominant, such as in the cases of Eupolemus and Demetrius, is indicative of Hebrew influence or of limited fluency in the Greek language. Tellingly, we observe that Eupolemus uses the variation that the Greek language offers by alternating between polysyndeton and asyndeton. We can contrast the use of polysyndeton in τόν τε χρυσὸν καὶ ἄργυρον καὶ χαλκὸν καὶ λίθον καὶ ξύλα κυπαρίσσινα καὶ κέδρινα “gold and silver and bronze and stone and cypress and ceder wood” (PE 9.30.8) to the asyndetic χρυσίον, ἀργύριον, χαλκόν, λίθους, ξύλα κυπαρίσσινα καὶ κέδρινα “gold, silver, bronze, stone, cypress, and ceder wood” (PE 9.30.6).

Third, resumptive pronouns are not unusual in Greek either. They already appear in the works of classical authors. Willem Bakker has given an extensive overview of occurrences from Homer through the classical period to Hellenistic Greek. ²¹ The use of this construction, too, seems to increase as time and linguistic development progresses into post-classical Greek. Eupolemus’s and Demetrius’s use of resumptive pronouns is, therefore, reflective of contemporary language usage.

The use of ὅλος in the sense of πᾶς in the work of Eupolemus can also be explained easily when taking into consideration other Greek writings. ὅλος in the sense of πᾶς occurs mainly in later Greek, particularly in papyri such as P. Oxy. 14 1770 (Oxyrhynchos, third century C.E.) and P. Abinn. 25 l. 15 (Philadelphia, 346 C.E.) as well as in literary sources such as Nonnus, Dionysiaca 47.482 (fifth century C.E.) and Sophronius of Damascus, Anthol. Pal. 7.679.5 (circa 630 C.E.). There are, however, some instances of ὅλος in earlier literature which already suggest this usage of ὅλος, such as Sophocles, Ajax 1105, and Menander, Pk. 225. The occurrence of this form in Eupolemus’s writing, therefore, need not be seen as unnatural Greek. Moreover, Εupolemus varies his word choice to denote “all,” as well as the position in which he places the adjective πᾶς. See, for example, ἡ δύναμίς μου πᾶσα in PE 9.32.1; τῶν ὑπὸ τὸν οὐρανὸν πάντων in PE 9.34.2; παντὸς τοῦ ἱεροῦ in PE 9.34.11; and τὸ δὲ σύμπαν χρυσίον in PE 9.34.16. ²² Eupolemus thus clearly knows different ways to express the same notion.

To summarize, when we re-evaluate the primary evidence, we can conclude that those features that have been cited as evidence for Semitic interference on the part of Demetrius and Eupolemus are, in fact, normal features of post-classical Greek. Comparative linguistic analysis helps us to refrain from “the tendency to equate oddities too readily with bilingual influences.” ²³

The evidence, II: features of educated Greek

Now that we have established that there is no clear evidence for bilingual interference in their writings, we can proceed to attempt to gain insight into the level of Greek Eupolemus and Demetrius used. Was it, in fact, substandard Greek? Since illiteracy was common in antiquity, these authors’ ability to read and write indicates that they would, in fact, have participated in a Greek-language education. ²⁴ While it is impossible to glean from a text the exact level of education a writer may have obtained, ²⁵ we are invited to look for indications of some degree of educated Greek usage on the part of our authors.

For the texts under consideration, we have to rely on indirect traditions of transmission, in particular, the works of Eusebius and Clement of Alexandria, ²⁶ who often cite passages preceded by the formula Δημήτριός φησι or Εὐπόλεμός φησι, “Demetrius/Eupolemus says.” For this reason, I have selected language features that do not depend on the sentence construction as indirect speech. This increases the likelihood that the feature under consideration may be attributed to the authors themselves rather than to later editorial activity. ²⁷

• In Hebrew, the order of the nomen regens and the nomen rectum in a construct state is fixed. In Greek, however, there is relative freedom when it comes to the position of the genitive vis-à-vis its governing noun. ²⁸ When a translator or writer follows the word order of the Hebrew, the Greek will stereotypically follow the same pattern, namely, the governing noun followed by the genitive. When a translator or writer positions the genitive before the governing noun, they make use of a typically Greek linguistic feature. ²⁹ We encounter the prepositioning of the genitive a number of times in the works of Demetrius (e.g., τὴν ἑαυτῆς παιδίσκην “her own slave” in PE 9.21.3; τῆς Χαναὰν γῆς “the land of Canaan” in PE 9.21.8; τοῦ Ἡλιοπόλεως ἱερέως “the priest of Heliopolis” in PE 9.21.12) and Eupolemus (e.g., τῇ τοῦ θεοῦ βουλήσει “by the will of God” in PE 9.30.2; τὸν τούτου υἱόν “his son” in PE 9.30.3; εἰς τὴν Οὐφρῆ νῆσον “to the island of Ofir” in PE 9.30.7).

• Variation in the placement of the cardinal number before as well as after the noun is another feature that indicates familiarity with Greek, ³⁰ since in Hebrew, the cardinal is usually placed before the noun. We see such variation repeatedly in the works of Demetrius (e.g., PE 9.21.3: ἐν ἑπτὰ ἔτεσιν “over the course of seven years” vs. PE 9.21.6: ἔτη ἕξ “six years”) and Eupolemus (PE 9.30.8: ὄντι ἐτῶν ιβʹ “who was twelve years old” vs. τῶν δώδεκα φυλάρχων “the twelve rulers”). Regarding the use of numerals, I would also like to draw attention to Demetrius’s use of the dual in μηνῶν δυοῖν “two months” in PE 9.21.8 (bis). The dual began to disappear even before the post-classical period. ³¹ If this particular usage is not the result of an Atticizing tendency on the part of copyists of Demetrius’s work, it represents an indication of the latter’s educational background.

• Another sign of educated language in post-classical Greek is the use of particles, which became increasingly infrequent. ³² Our authors use a variety of particles, such as τε and τε (. . .) καί (e.g., for Demetrius, PE 9.21.3; 9.29.15; for Eupolemus, PE 9.26.1; 9.30.2; 9.30.8), δέ and particularly μέν (. . .) δέ (e.g., for Demetrius, PE 9.21.9; 9.29.15; for Eupolemus, PE 9.30.8; 9.32.1), πρῶτον μέν (. . .) ὕστερον δέ (e.g., Eupolemus, PE 9.34.13), γάρ (e.g., Demetrius, PE 9.29.16), οὖν (e.g., Demetrius, PE 9.29.16), and δή (e.g., Eupolemus, PE 9.30.5).

• In classical Greek, a subject in neuter plural will have a verb in the singular. This is the so-called σχῆμα Ἀττικόν. ³³ In post-classical Greek, we encounter a gradual move toward supplementing a neuter plural subject with a verb in plural. In Eupolemus, we find an example of the σχῆμα Ἀττικόν: ὅπως χορηγῆται τὰ δέοντα “so that their needs will be supplied” (PE 9.34.3). This indicates clearly that Eupolemus is educated in Greek beyond the standards of non-literary post-classical Greek.

• Separation of an attribute from its noun, known as hyperbaton, is a typically Greek feature. ³⁴ It can be found in Eupolemus, PE 9.30.5: ἀξιοῦν τὸν θεὸν τόπον αὐτῷ δεῖξαι τοῦ θυσιαστηρίου “he asked God to show him a place for the altar,” where the genitive τοῦ θυσιαστηρίου is separated from its governing noun τόπον.

I could add numerous examples of additional features that signify educated Greek usage, but the above should suffice to reframe our perception of Demetrius and Eupolemus as proficient post-classical Greek authors and to allow us to move on to the bigger question: How do we understand the style of Demetrius and Eupolemus?

Education and identity in Greek-speaking Judaism

The above analysis has shown that the language of Demetrius and Eupolemus is at home in a post-classical environment. There is no solid ground for the argument that their language indicates that either of them was bilingual. Their use of Greek is not rudimentary. On the contrary, it contains significant subtleties indicative of an educated background. Why, then, have these authors been dismissed by scholarship in the past?

The context in which Eupolemus and Demetrius are to be situated is that of Hellenistic Judaism. The study of Judaism in the Hellenistic period has been significantly influenced by a scholarly paradigm that regards Judaism and Hellenism as monolithic cultures in opposition. The background to this paradigm lies in a philosophy of language dating to the nineteenth century that identifies language, culture, and ethnos. ³⁵ In this view, access to Greek-language education for Jews “must have been purchased with the betrayal of Judaism.” ³⁶ As a result of this underlying framework of thought, scholarship has not been able clearly to position authors such as Eupolemus and Demetrius. The assumption is that, as Jews, they would not have penetrated the Greek education system to a significant extent and could only have been “Hellenically half-educated,” thus writing in “pidgin Greek.” ³⁷ Alternatively, if these authors’ facility with the Greek language was not necessarily considered to be problematic, the view is often expressed that Jewish-Greek authors were writing for apologetic or propagandistic purposes ³⁸ —in other words, they were writing Greek, not because of an internal Jewish development of expanding literary traditions across languages, but because these works were presumably aimed primarily at the non-Jewish world. This view of a Judaism/Hellenism dichotomy represents an oversimplification of the complex social and cultural interactions between Judaism and the Greek world into polar opposites. Hellenism, defined as the phenomenon of the spread of “Greek” culture, is multifaceted: it can be seen in terms of the Greek language, literature, art, politics, or ideas, for example. “To class all of these as one cultural system is to ignore the differences between them and the likelihood that one person may be in favor of one aspect but not another.” ³⁹

The literary evidence shows us that Jews created a substantial body of Greek-language literature. Eupolemus and Demetrius are part of a wider movement: we have at our disposal translations of scripture into Greek (commonly referred to as the Septuagint), as well as numerous compositions, such as deuterocanonical writings, pseudepigrapha, apocrypha, and fragments of several Hellenistic Jewish authors, ⁴⁰ which in addition to our historiographers include writers such as Ezekiel the Tragedian, Philo the Epic Poet, Aristoboulus, and Artapanus. While we cannot at this moment go into defining each of these individual categories, what this group of texts shows is an engagement with Jewish traditions in the Greek language in a variety of genres of prose and poetry. The start of this literary engagement is generally located in the early third century B.C.E. ⁴¹ We cannot exclude the possibility that Jews may have already been using Greek prior to this, but we do not have evidence for it. Salt-tax papyri from the mid-third century B.C.E. record Jewish names as Hellenes, indicating that Jews were learning Greek, at least in Egypt. ⁴² The earliest papyrological evidence we have for Jews writing literature in Greek consists of second century B.C.E. papyri of the Septuagint Pentateuch and is found in both Egypt and Palestine (P. Ryl 458 and 4Q122, respectively). For the fragmentary Jewish-Greek authors specifically, we can only be relatively certain of a terminus ante quem of the first century B.C.E. This date is based on the fact that these fragments have been transmitted as citations by Christian writers which in turn go back to the work of Alexander Polyhistor, a (pagan) historian dated to the first century B.C.E., who quoted extensively from Jewish-Greek authors in a now-lost work titled On the Jews. ⁴³

Keeping these caveats regarding the delineation of a Hellenistic Jewish-Greek literary corpus in mind, the evidence attests to the fact that Jews in the Hellenistic period established a literary output in Greek alongside Hebrew and Aramaic in a variety of modes and genres. Certain Jews, in any case, had no problem using the Greek language and literary forms. The use of Greek provided an opportunity for Jews to expand their own literary traditions in different languages. In other words, Judaism in the Late Second Temple period is multilingual. The most straightforward evidence for Jewish multilingualism in the Hellenistic era is the fact that the Dead Sea Scrolls contain texts in Hebrew, Aramaic, and Greek. The presence of Greek in both documentary and literary documents at Qumran has been undervalued. ⁴⁴ Particularly interesting in the framework of the present study is 4Q127, which seems to represent the genre of “rewritten scripture” in Greek ⁴⁵ and may provide an interesting parallel for our historiographers, whose work is often related to this genre, too. ⁴⁶ The scrolls suggest that the activity of Eupolemus and Demetrius is at home in a Jewish environment without a need for the assumption that their work was directed outward. ⁴⁷ Jewish literature in Greek needs to be integrated further into our understanding of the variegated nature of Judaism in the ancient world.

The above linguistic analysis, on the basis of contemporary comparative material, has shown that the Greek of Eupolemus and Demetrius is not in any way clumsy. Their language is parallelled closely in papyri and thus reflects standard post-classical Greek. This indicates that Eupolemus and Demetrius had participated in the Greek-language education system to a meaningful extent. That they did not write in the high register of literary Greek associated with a historian such as Polybius may, in this view, not necessarily be evidence for a limited level of education, but reflective of their position as Hellenistic authors for whom it was acceptable to write in this style and register of Greek. The literary output of the Hellenistic era was large, yet only part of it has been preserved—usually what is assumed to have been the best works, often by canonical authors such as Polybius or Apollonius of Rhodes. ⁴⁸ We do not have access to the full range of Greek literary production from the Hellenistic era. While we may not have the most appropriate comparative material for the Jewish historiographers at our disposal, the above analysis based on the available evidence has shown that Greek-speaking Jews would have used Greek as well as their non-Jewish contemporaries. As much as Eupolemus and Demetrius are insiders in the Jewish tradition, they are also insiders within the Greek-speaking world.

Conclusion

In past scholarship, Demetrius’s and Eupolemus’s position in the Hellenistic world and in Jewish society has been extrapolated from their supposed (lack of) linguistic skill. This negative evaluation, in turn, has been dominated by scholarly assumptions regarding “standard” Greek style as well as the conceptualization of the relationship between Hellenism and Judaism as monolithic cultural constructs. However, a reassessment of the historiographers’ language usage in the context of post-classical Greek has led to the conclusion that they were, in fact, proficient Greek writers. Although we may not be able to determine with certainty what level of education each author would have attained, the subtleties of their language usage indicate a significant familiarity with Greek. They wrote in a standard register of the Greek language because it was acceptable to them and to their audience. New line of all fragmentary Jewish-Greek authors, Eupolemus and Demetrius have been evaluated most negatively as Greek writers. That even their language is representative of standard Greek suggests that the same may hold true for other Jewish-Greek authors, too. These writings need to be contextualized within a broader movement in Second Temple Judaism which uses Greek as a means to develop their own cultural traditions. This broader movement is thus one of positive engagement with aspects of Hellenism. ⁴⁹ Hellenism can, in this view, be most adequately described as a toolbox used to construct the Jews’ own evolving identity in a new societal environment of cultural interaction. At the most basic level, this tool is the Greek language. ⁵⁰

Jews writing in Greek represent a peculiar example of hybridity. Their output is inherently located on the crossroads of two literary systems: that of Jewish literature in the Hellenistic period which has a tradition in Semitic languages and for which Greek is the novel language of literary productivity, as well as that of Hellenistic literature which has a rich tradition of its own and which became a mode of expression for Judaism. Hellenophone Jews are part of the cultural diversity that characterizes Hellenism. As much as Jewish literature attests to the development of a Jewish tradition in Greek, it also attests to the different ways in which Hellenistic Greek literature could be shaped to fit specific contexts. Jewish literature in Greek reflects Jewish engagement with the Greek world as an expansion of the Jewish literary tradition across languages.