Septuagint Origins: The State of the Question

Development of Previous Scholarship

Several recent publications reaffirm or refine previously held views, usually by providing additional data or responding to criticisms. Kreuzer has published a number of essays since 2010, based on the theory first published in 2004. In fact, his 2004 essay was reprinted in 2015 (Kreuzer 2015a) as part of a volume of his collected essays. In the meantime, Kreuzer also provided a more general introduction to the Septuagint in the LXX.D commentary (2011), which incorporated much of the same material and was translated into English (Kreuzer 2015b). In his latest discussion on this topic (2016; English translation 2019), Kreuzer does not develop the theory extensively, but restates, for the most part, that the Septuagint was created for internal needs at first, and that those are difficult to ascertain. Nevertheless, and adapting Rösel’s view (1994), he argues that Genesis would have been revised shortly after its initial production for circulation within educated circles to legitimize the Jewish community. This effort would have taken place in a climate of intellectual and political rivalry, perhaps in the context of the Mouseion. Though the Aristeas narrative is not historically reliable in terms of its main points, Kreuzer thinks it nevertheless provides an accurate portrait of the cultural-political environment of this period and suggests the need to bring (at least part of) the Jewish Scriptures into dialogue with Greek and Egyptian traditions. Among newer data put forward by Kreuzer is Usener’s analysis of the language of the Septuagint (2011, 2016), which, in his estimation, demonstrates contact between its translators and the Greek philosophical tradition. He also cites Niehoff’s study on Homeric scholarship in Alexandria (2011) as another piece of evidence for contact between the translators and Alexandrian culture and education, ‘the traces of relevant knowledge [being] just too numerous’ (2019: 10 n 9).

Rösel’s most recent publication on the matter argues that ‘it is hardly conceivable that the translation came into being without the agreement or support of the Jewish community in Alexandria’ (2021: 255). Several aspects imply extensive preparations, whether that be the consistency by which key religious terms are translated, or the very few instances where the source text appears to have been misunderstood. Following Aejmelaeus (2013; 2018) and her arguments to this effect (see §Related Developments), Rösel argues that these observations could point to a period of oral translation in which such terms were defined within the Jewish community, which preceded the production of written texts. The use of an early lexicon or bilingual word list is also likely, as claimed by Lee (2018, see below). That similar ‘sophisticated aids’ were developed for the study and interpretation of Homer in the Mouseion context suggests that this was the model the Jewish translators used for their task (Rösel 2021: 254; see also Rösel 2016a). Likewise, the use of Egyptian colloquial terms, as well as adaptations to an Egyptian context, points to an Egyptian origin for the translators. But their familiarity with Greek philosophy, as far as Genesis is concerned, suggests a level of education and social context other than scribes with a military background (Rösel 2021: 255-57). Individuals with this level of education would most likely be found in Alexandria, or at least in a milieu closely connected to it. Moreover, that Genesis would have been translated first precludes the theory of legal need.

Kreuzer and Rösel’s assessments are closely connected in that they both see Genesis as a Jewish history circulating in Alexandria’s intellectual milieu, perhaps as a response to Egyptian competing accounts (e.g., Manetho). But, as Rösel points out, the apologetic nature of such literary efforts, and the accompanying impetus ‘to resolve inconsistencies and contradictions found in the biblical text’, can also be understood as primarily directed inward (2018: 11). The same motivation can be observed in other works of this period, such as Daniel 1–6, Esther, and Judith, whose apologetic purpose is analogous yet directed at the Jewish community. In a similar vein, Schaper’s work on Exodus leads him to see in this translation (and the Pentateuch more broadly), not a religious or legal translation, but rather a people’s history. However, rather than situating this history within a scholarly context, the colloquial nature of the Greek leads him to align with Frankel in describing it as a Volksbuch, popular literature intended to entertain (Schaper 2014: 44-46; see also Frankel 1851: 79). Another variation of this view would see the Pentateuch translation stored in the library after its creation, as part of an effort to collect law codes and other documents (see the thorough discussion in Honigman 2017).

Likewise, van der Kooij develops the view he has promoted in the past few decades on the importance of the priestly class. In an extensive essay (2012a; see also 2012b; 2013), he attempts to demonstrate how this is reflected in the translation of the Pentateuch, which would elevate the priestly class, and especially the high priest as the prime exegete of the law. Such an emphasis would only underscore, in his estimation, the theory that the translation project would have been conducted under priestly authority. Any other scenario would produce a text which would be considered a rival version (van der Kooij 2012: 56, 58). He goes on to argue that the Pentateuch was conceptualized as a single entity, a unit that would have been translated as part of the same project, even if Genesis was perhaps translated before the remaining books within that unit (2012: 57-58). The differences in translation approach observed in various parts of the Pentateuch should be interpreted in conjunction with the lexical stock that is common to all. This combination of differences and similarities would simply reflect the work of several individuals working as a team. As for the location of this activity, van der Kooij does not consider the Alexandrian community to be sufficiently developed to support an endeavor of this scale. Instead, the translators must have come from Jerusalem, where the required expertise in the reading and interpreting of the Torah can be found (2012: 59). As for the linguistic profile of the translations, van der Kooij suggests that Egyptian vocabulary found throughout the Pentateuch can be explained by the spread of the Greek language throughout the Ptolemaic empire, even in Judea, where the use of similar vocabulary could be expected. One must also account for the accidental nature of the evidence and the fact that Egyptian sources are better attested. On the other hand, the translators’ familiarity with Homer and Greek rhetorical devices betrays their status as ‘scholar scribes’ who may possibly have spent time in Alexandria, on the model of Ben Sira’s grandson: ‘a stay of a number of years may have given them the opportunity to improve, if necessary, their knowledge of Koine Greek, and to be in contact with Greek scholars like Hecataeus of Abdera and Demetrius of Phalerum’ (2012: 60-62). Their translation’s particular style is connected to its contents, a legal document, which can be expected to be rendered in this way. This style bears similarities with what is found in contemporary legal documents and contracts.

The connection with Jerusalem or Palestine more broadly was also repeatedly argued by Tov (1997: 185-86, 2010: 21-22, 2022). In more recent publications (e.g., 2020, 2022), he cites as evidence the presence of interpretative traditions shared by the Septuagint Pentateuch and Palestinian halakhic exegesis. This can be seen in the shared exegetical traditions concerning the genealogies of Genesis 5 and 11 (Tov 2014a, 2022). Tov also points out that the Greek translations reflect textual variants that align with Hebrew texts of Palestinian provenance. An example of such commonalities is the joint harmonizing tendencies observed between the Septuagint’s putative source texts and the Samaritan Pentateuch, which points to a textual stratum common to both, as well as pre-Samaritan texts found in Palestine (Tov 2022; see also 2016, 2018). This would suggest that the Hebrew source texts, and perhaps even the translators, originated from Palestine, in keeping with the basic outline of the Aristeas narrative. As with Rösel, Tov (2014b) argues that Genesis was translated first. His argument is based on an analysis of the vocabulary and translation technique employed throughout the book, which shows traces of internal development as the translator progresses before reaching a more settled method.

Herakleopolis

Dorival concluded his survey by discussing a significant discovery of recent decades that could shed light on the Egyptian Jewish community during the Hellenistic period: the Herakleopolis politeuma papyri. The first complete edition and translation of this collection was provided by Cowey and Maresch (2001), while a new edition was published recently as part of Corpus Papyrorum Judaicarum IV (Hacham and Ilan 2020). These 21 documents, dating from the second half of the second century BCE, attest to the existence of an administrative arrangement which provided this Jewish community a limited degree of self-governance. References to ‘τὸ εἰθισμένον τοῦ ἀποστασίου τὸ βυβλίον’ (‘the customary divorce deed’) in a petition concerning the dissolution of a betrothal (P.Polit.Iud. 4) harkens to a similar phrase in the Septuagint version of Deut 24.1–3. Moreover, the mention of a decision, ‘κατ̣ὰ τὸν νόμον’ (‘according to the law’), has revived the thesis popularized largely by Mélèze-Modrzejewski (1997, 2001, 2011) to the effect that these petitions refer to, and even appeal to, the Septuagint Pentateuch as legal authority. Dorival proposes, therefore, that the translation project is motivated by the Jewish communities’ legal needs. However, he goes on to qualify this thesis: the Ptolemaic legal system would not allow such an independent initiative. The different legal codes that were composed in Greek were created at the king’s initiative. He could also intervene directly through his khrematistai (Dorival 2010: 44). Therefore, even in this scenario, the king would have been involved in some way, if only to sanction the undertaking.

Considerable attention has been devoted to these texts since, notably in the research of Kugler (e.g., 2011, 2013, 2022b, see also Sänger 2018, 2019; Szántó 2024:175-200), with a focus on the various types of connections between the petitions and biblical law. In his recent monograph (2022a), Kugler argues that the collection shows that Egyptian Jews relied on both the prevalent Greek laws and their ancestral law to argue their case. After transcribing, translating, and commenting on each document, he attempts to show that petitioners would choose to present their case in this venue because of the Jewish legal norms that would apply in the evaluation of their case. How this connects to Septuagint origins is more difficult to ascertain, however. Kugler claims that these petitioners were familiar enough with at least a few sections of the Torah ‘to use specific passages in making their legal arguments’ (2022a: 255). However, he does not discuss how this might have operated or how it required a Greek version of the Torah. As some have pointed out, the connections between these petitions and Jewish legal norms are not as obvious as Kugler would like them to be (Kreuzer 2023: 180; Maurais 2024b). This is because, in some cases, a similar Greek or Egyptian legal norm can be found. In other situations, the connection to Jewish legal norms (here in their Septuagint version) is superficial in wording and distant in terms of underlying principles. Moreover, the abstract legal norms in view do not need to exist in their Greek form to influence these communities. These legal principles could alternatively be based on oral customs or traditions only loosely connected to the Torah texts, as argued in Altmann’s recent study of these texts (2022).

Though this theory has the advantage of working from extant data (Aitken 2021: 12-13), the politeuma papyri would achieve very little, then, in illuminating the origins of the Greek Pentateuch, at least in terms of motivations. Nevertheless, these documents could potentially demonstrate how the Septuagint Pentateuch was subsequently employed. They also remain helpful in a more indirect way, by demonstrating the integration of Jews in Egyptian culture, at least in the context of the second century BCE. The Jewish politeuma remains a structure firmly embedded in its Egyptian social context: non-Jews present their petitions there, and the legal norms observed do correspond in large part to Greek legal norms.

Studies on the Social and Scribal Context

Additional evidence besides the Herakleopolis archive has emerged and sheds light on the social context of Egyptian Jews. Under this rubric, I want to highlight recent studies on the social situation of Jews in Ptolemaic Egypt in the third century BCE, especially as it relates to their occupations, scribal activity, and the use of Greek. A brief survey of Jewish presence in Egypt in this period is provided by Capponi (2021), while a far-ranging discussion of the various types of evidence can be found in Szántó’s recent monograph on the topic (2024). Aitken (2020: 400) posits that ‘the finds of the past century allow for a reasonably confident reconstruction of the social setting in Egypt that could have generated a translation of the Pentateuch’. These can illuminate the range of possible motivations, location, and even dating of the translation project (Aitken 2021: 9-10).

More directly connected to the issue of Septuagint origins is Honigman’s recent study of identity markers in the documentary papyri of the third-century BCE Egyptian chora. Working from an analysis of tax rolls from the Fayum (Arsinoite nome) in Middle Egypt, she argues that Jews were present in large numbers in some villages and regularly categorized as Hellenes for tax purposes. Building on earlier discussions by Aitken (2011a; 2011b), Carleton Paget (2014: 118-19), and others, she argues that this status likely required a command of the Greek language and represented an outward mark of Greekness. It also entailed a higher standing in the community, though such individuals would refer to themselves as Ioudaioi in other contexts. Many would have been established there for more than three generations. Nevertheless, this minority continued to use identity markers such as Jewish personal names (Honigman 2023b: 205-07). Moreover, as the third century progresses, more Ioudaioi are counted among military recruits, the privileged cleruchic class endowed with a piece of land (Honigman 2023b: 207-08). Intermarriage with non-Jews also appears widespread in this context. This mixing of identities suggests the existence of a privileged Jewish minority, well integrated into Egyptian culture throughout Egypt already in the third century. Similarly, Egyptians sometimes used a Greek name, particularly in relation to the privileged tax-exempt status that some held (Honigman 2023b: 212-13).

The topic relates to Septuagint origins in that the privileged tax status was likely connected to the holding of an official function in the royal administration. Though, as Honigman surmises, there is no direct evidence of Jewish and Egyptian scribes serving together in this context, the everyday interactions observed between these privileged groups in remote regions suggest that they very well could have done so (Honigman 2023b: 214). Given that literary and non-literary translations were produced in such settings, these ‘are also by far the most plausible place where Greek-speaking Judean and Egyptian scribes were likely to meet and mingle’ (Honigman 2023b: 213). Consequently, this context offers a plausible explanation for the similarities in translation practices between official documents and those found in the Septuagint Pentateuch, as Aitken (2016) seeks to demonstrate. Moreover, such interactions among the scribal elite likely led to the circulation of Jewish traditions among Egyptians and vice versa.

In another recent essay on the topic, she suggests that the translation of the Pentateuch, especially in its ‘manipulation of the genealogies’ of Genesis, reflects a competition between these Jewish and Egyptian elites (Honigman 2023a: 4-5). She argues from likely allusion by Manetho to the Exodus narrative, and Demetrius ‘the Chronographer’ and his work On the Kings of Judea, that not only the Pentateuch, but also the books of Kings and the other so-called historical books were available in Greek by the end of the third century at the latest (see also Honigman 2025). The motivation for translating these books would have included, on this theory, ‘the need to engage in the competition about which people are the oldest’, and therefore most apt to promote their god or gods to the Ptolemaic court. This is not to say that the translation was made at the royal court. She argues that it could very well have been the work of Jewish scribes employed by the administration, as Aitken has proposed. But instead of situating them in the Fayum, she would rather locate them in Alexandria (Honigman 2023a: 6).

In his 2020 article on the Ptolemaic provenance of the Septuagint, Aitken also points out that Jews could be found in many locations in third-century Egypt, in numbers sufficiently large to build prayer houses, for example. He also points to the Fayum, a prosperous area settled by Greeks and others who often came from outside Egypt, where immigrants would have been in regular contact with Greek speakers (Aitken 2020: 401-402). This situation could also account for the presence of Aramaic loanwords and spellings in the Septuagint, a phenomenon previously explained by positing that its translators were descendants of Aramaic-speaking Jews. Though, as he surmises, contact between Aramaic and Greek is more probable in Judea rather than Egypt, Aitken points to the presence of Judean immigrants who mastered Greek as a second language in Egypt as another likely explanation (Aitken 2020: 401, 410-411; see also Joosten 2010, 2017; on Joosten, see §Research Contributions and the Ethics of Citation below). There is no need, under this hypothesis, to posit that translators worked in Judea or came expressly from there to perform this task. In fact, their native speaker competence in Greek (to be discussed in the next section) implies that they have been in Egypt for some time (Aitken 2021: 14-15).

Moreover, and despite the accidental nature of the evidence, the use of Greek in Egypt appears to grow in importance as the third century progresses. This is especially noticeable toward the middle of the third century in administrative correspondence, likely stimulated by Ptolemaic legislation to this effect (Aitken 2020: 403). In contrast to the earlier period under Ptolemy I where Demotic remained the language of administration, Greek became more pervasive in this context from 260 BCE onward due to the emergence of a greater number of trained scribes (Aitken 2020: 404-405). Aitken posits that the growing number of educated Jews writing in Greek represents an important pool of individuals from which translators could be found (Aitken 2020: 404). Drawing on studies on scribal education in this period, he points out that ‘Septuagint translators were most likely comparable to the more skilled of the Egyptian bureaucratic scribes. . . having acquired enough rhetorical skills and learned enough of classical literary vocabulary to use in their work’ (Aitken 2014b: 23). Similar scribal features between Egyptian translations and Septuagint texts, such as the occasional use of Greek particles going out of use in this period, also suggest similar scribal practices. Moreover, the quality of the Egyptian scribes’ linguistic output is on par with similar documents of this period written by Greeks, despite their training in both Demotic and Greek (Aitken 2020: 407-09). This provides an intriguing parallel to the possibility of Jewish scribes being trained in Hebrew but who were also able to write competently in Greek. It also suggests a setting where this could have taken place. If Jewish scribes followed the Egyptian practice of scribal training within temple contexts, it is possible to imagine similar activities by Jewish communities in the context of their prayer houses, where a copy of their Hebrew scrolls would be stored (Aitken 2020: 403-404).

Putting all of this together, Aitken suggests that the translation of the Pentateuch would make more sense in the period after the broader adoption of Greek as an administrative language and the emergence of scribal education leading to the situation described in the tax papyri, that is, toward the middle or second half of the third century BCE (Aitken 2020: 405). And as it becomes obvious, the centrality of Alexandria as a setting for the translation is relativized when considering the spread of Greek education throughout the Egyptian chora (as posited, long ago, by Frankel 1841: 9-11). Another consequence of the similarity in translation approaches between Egyptian texts and those attested in the Greek Pentateuch is that the translation style in itself can tell us very little about the translators’ objectives (Aitken 2021: 11-12). At the same time, Aitken’s findings do not invalidate all theories of Septuagint origins. Wright (2021a: 246) formulates it thus:

If the LXX translations merit favorable comparison with the Egyptian translations studied by Aitken and they thus function similarly, we are no closer to determining the precise Sitz im Leben of the LXX at its point of production. Certainly a functional translation would work well in liturgy or education, for example, and so some theories about the LXX’s origins remain tenable. It also becomes more difficult to take seriously a narrative such as the Letter of Aristeas, the historical veracity of which can be questioned on other grounds, when it imagines a highly sophisticated text, translated at royal behest, to be placed in the Alexandrian library. The question of the LXX’s origins thus remains an open one.

Sociolinguistic Studies

Sociolinguistics is another discipline that has played an important role in assessing these questions over the past few years. Here also, Aitken has been particularly active, while Lee, working within a similar framework, contributed significantly to the discussion with his 2018 landmark study, The Greek of the Pentateuch.

Aitken forcefully argues in the conclusion of his 2014 monograph that ‘language study should form the basis of Septuagint studies’ (2014a: 106). By this, he means not only the type of studies that have been done until now to better situate the Greek of the Septuagint within its historical context, on the excellent model provided by Lee, but also by careful attention to the social component of the linguistic data. A benefit of this approach is that ‘[it] sheds light on the educational levels, literary sensibilities, and social interactions of the silent individuals behind the texts’ (Aitken 2014a: 107). Investigating the Pentateuch translators’ social identity and location using linguistic data is not new (see Kreuzer and Rösel above). But this trajectory was pursued much more intently by Aitken within the scope of the extant Egyptian documentary evidence.

Aitken’s investigation of the register and style of these documents led him, in conjunction with Lee’s work, to identify the Greek of the Septuagint Pentateuch as ‘educated Greek’, a register that often corresponds with what is found in Egyptian administrative documents (Aitken 2014b; Lee 2018: 271-75). In other words, the Greek of the Pentateuch reflects the linguistic usage of the postclassical period and ‘largely reflects the vernacular from Egypt, a non-literary standard familiar from documentary papyri’ (Aitken 2020: 406; see also Aitken† and Dhont 2023; Ross 2023a). Consequently, Aitken argues that the quality of the Greek is not to be attributed to incompetence or an easy technique (as suggested by Barr 1979: 300; see also Aejmelaeus 2007: 63-63), but is a conscious stylistic choice based on common translation practices throughout Egypt (Aitken 2016; 2021: 17-18). Such a similarity suggests that the translators were from Egypt. This interpretation diverges significantly from that recently proposed by Ziegert (2017). Ziegert, following Rajak (2009), posits that the ‘literal’ translation style represents a form of resistance, a way of reaffirming Jewish identity in response to Greek cultural influence. It would also demonstrate the translators’ respect for their source text and its language. For Aitken, similarities with Egyptian translation practices undermine this type of explanation.

But Aitken’s investigations were not limited to documentary sources. His work on epigraphic material also demonstrates its relevance for understanding the variety of genres and registers employed in postclassical Greek, a variety that is reflected in the vocabulary choices of the Septuagint translators (Angelini and Vergari 2023; Aitken 2014a). Consequently, and in contrast to the approach taken by Rösel and Kreuzer, the fact that translators were familiar with Homer and other ‘classics’ is not understood as reflecting an effort on their part to participate in scholarly discourse in the context of the Mouseion, but rather that Greek scribal education included the memorizing and studying of Homer and other Greek poets (Aitken 2021: 12). In this context, the variety of phenomena observed, from so-called Hebraisms to Homerizing features, is not contradictory, but points to the deliberate use of various linguistic registers and a particular style (Angelini and Vergari 2023: 20-21). In fact, this combination of features is not to be explained as a confusion of linguistic varieties or as an effort to assert their Jewish identity (in reaction to acculturation, as per Rajak 2009). Rather, the ‘eclectic style’ that makes up Septuagint translations is the result of its translators attempting (with varying degrees of success) to communicate an ‘educated’ style (Aitken 2020: 406; see also Angelini and Vergari 2023: 22). Thus, the study of the Septuagint’s language confirms a particular stage in the history of Greek and a register that is ‘not of the highest literary standard, but not without literary features’ (Aitken 2021: 11). According to Aitken, ‘this would support the idea that the translators had spent much time in Egypt, descended from Jews in Egypt or immigrants that had been present for many years, rather than newcomers with little knowledge of the language’ (Aitken 2020: 407). Lee’s conclusions are very similar (2018, 2020).

Another corollary is that the vocabulary employed also points to the translators’ milieu, with both Septuagint translators and Egyptian scribes using words in ways that are attested in this period. These would include, for example, the Egyptian loanword θῖβις for ‘box’, or words employed in their later postclassical sense, such as σῶμα to designate a ‘slave’. This would also support the argument for locating the Septuagint translators among the scribes employed by the Ptolemaic administration, rather than assuming they were mercenaries (Aitken 2020: 409-10).

Egyptian influence on the translator’s vocabulary has been the object of sustained inquiry in recent years. In her doctoral thesis, Recalcati (2024) examined the influence of the Egyptian milieu on personal names, toponyms, geographical images, job titles, weights and measures, animal names, realia, and theological accents. Her study confirms that the Egyptian social context exerted discernible influence on the translation of Genesis, Exodus, and Leviticus, within the scope of her research. She notes that ‘Greek lexemes appear to be used with the aim of giving a higher degree of precision in adherence with the Egyptian reality’ (Recalcati 2024: 253-54). These imply knowledge of the geography of the contemporary Mediterranean region and familiarity with job titles employed in Ptolemaic Egypt (see also the shorter discussion in Pfeiffer 2016). In her conclusion, she notes that the translators’ familiarity with their setting is most obvious in the realm of everyday elements and Egyptian indigenous culture. These findings would therefore question the translators’ connection with a scholarly milieu such as that of the Mouseion (Recalcati 2024: 261-62).

On the issue of dating, Lee (2018: 273) is less precise than Aitken but concludes:

The linguistic evidence points to a date early in the Ptolemaic period but cannot establish a terminus ante quem earlier than the middle of the second century BCE. The incompleteness of our evidence means that no conclusion can be certain, but as things stand, the many connections between the Greek of the Pentateuch and the Greek of the third century BCE documents corroborate a date in that century.

As for the location and motivations for such activities, Aitken suggests that the translators could most likely be found in the Fayum. They would need no royal patronage, being financially independent, and possibly familiar enough with their Hebrew Scriptures ‘through translating it and discussing it in Greek in a schoolroom setting, perhaps in a local proseuche’ (Aitken 2020: 411). Such discussions would have facilitated the translation by establishing equivalents for many key terms in advance. It is also possible that some passages had already been partially translated in this context.

In Aitken’s estimation, the needs of the Jewish community remain the most likely explanation. He offers two possibilities (2020: 412-414): (1) the translation could be seen as an important component of Jewish heritage for Egyptian Jews, perhaps in the context of the redaction of competing histories such as those of Manetho and Berossos. On the analogy of Egyptian priestly scribes, it is also possible that this task was undertaken by the personnel of the proseuchai and preserved there, thus not in conflict with theories positing a schoolhouse or liturgical setting. (2) The translation could also be motivated by the cultural value attached to the use of Greek by Egyptians, and not solely because of the loss of Hebrew, though such loss is likely to some extent (see also Alfaro 2024). The flourishing of Greek made it a natural activity and afforded the possibility of diffusing one’s work in this language.

Aitken’s forays in these areas stand out, since they explore the issue in many directions. Yet, the most concerted effort to argue a similar case from the linguistic evidence is found in Lee’s 2018 monograph (see also Lee 2020). After examining the Septuagint Pentateuch from the perspective of linguistic variation, educated language, Greek idiom, and variation in translation technique, Lee concludes that its translators would have received a formal education in Greek, were fully proficient in this language, and professionally trained for this task. The hebraistic style often encountered would therefore represent the outcome of a deliberate stylistic choice (Lee 2018: 271-75; see also Dhont 2023).

The consequences of these investigations for our understanding of Septuagint origins are many. If scribes of the Ptolemaic administration located throughout Egypt possessed the ability to perform such a task, then not only is Alexandria not essential as a location, but royal sponsorship is entirely unnecessary to materially (and financially) support such an enterprise (Aitken 2021: 15; Angelini and Vergari 2023: 19). The Greek Pentateuch’s status as a groundbreaking project and cultural innovation is also downplayed. Instead, the Septuagint is viewed as part of a long tradition of Jewish translation activity, with the Pentateuch specifically embedded in its broader Egyptian sociocultural context (Aitken 2019).

Related Developments

Other efforts are ongoing to relate the Greek Pentateuch to its Egyptian milieu. Under the rubric of historical linguistics, studies continue to build on Lee’s work to situate the Greek Pentateuch primarily in terms of dating. Of note here is William Ross’s study on donkey words, which extends Lee’s study of this vocabulary by examining previously unavailable documentary evidence (Ross 2023b). These confirm the range that Lee and Evans proposed based on their lexicographic and syntactic analysis.

Another line of inquiry is found in the work of Korytko (2023) on the covenant code (Exod 21-23). Laws about the death penalty are examined in light of similar cases and formulations in Ptolemaic law codes. In his estimation, the translation shows awareness of these laws and seeks to adapt some aspects of biblical law to the prevailing Ptolemaic legal norms. The careful comparison of legal formulation between biblical and Ptolemaic texts raises intriguing questions as to the sociocultural context in which the Exodus translator lived, as well as the motivations for such adaptations. Such findings would support the view that Septuagint translators are connected with the administration in some way.

Also building on an analysis of the Septuagint’s linguistic data, Aejmelaeus (2018: 12-13) follows Lee and others in positing an Egyptian provenance for the translators. In an approach recalling Aitken’s, she argues that the only reliable source for properly understanding Septuagint origins is the translation itself, its language, style, and translation technique (2018: 12; see also Wright 2011, 2015). Her main argument, however, is that linguistic data suggest a community-based development that took place in an oral context. She remarks that the specialized terms employed for religious and theological concepts shed light on the question of Septuagint origins. Some are clearly of Aramaic origin (e.g., σάββατα, ‘Shabbat’), while others are simply transcribed into Greek and do not seem to require explanation from those who employ them. This suggests that the community spoke Aramaic at some point in the past and that such terms remained in use despite the emergence and widespread use of Greek. She also observes that the selection of several Greek equivalents for theological terms (e.g., berit = διαθήκη for ‘covenant’; torah, in the sense of ‘instruction’ = νόμος, ‘law’) also points in a similar direction: these Greek terms were likely employed in Jewish communities’ spoken language before their use in the Septuagint (2018: 13). This would explain the consistency with which they are employed as equivalents throughout the corpus. Other translation choices aim to avoid improper language, such as preventing any implication that God can be seen when visiting the sanctuary. These features further suggest a context of religious use rather than one associated with the royal library.

Regarding the emergence of the written translations that make up the Septuagint Pentateuch, Aejmelaeus suggests that it is best understood as a process that closely parallels the creation of the Targums (2018: 14-16; see also Aejmelaeus 2013: 5-13). These Aramaic translations have been argued to originate from the practice of orally rendering the Hebrew Scriptures into Aramaic alongside the reading of the text. While this activity may go back to the fourth century BCE, significant changes in linguistic environment within the span of a few generations led to the target language shifting to Greek (see also Brock 1980). According to Aejmelaeus, many of the translations’ features described above—particularly the avoidance of theological missteps and the distinctions made in translation, such as between legitimate and illegitimate altars—can be attributed to an oral preparatory phase. However, the variation in the application of such distinctions to particular instances in the Septuagint Pentateuch and the Targums suggests a common tradition and shared translation conventions rather than direct literary influence in either direction.

It is also noteworthy that each translation tradition developed along different trajectories. In Aramaic-speaking communities the close relationship between both languages allowed the Aramaic translation to expand into explanation and exegesis much more overtly, as it continued to accompany the reading of the Hebrew Scriptures. In effect, the audience would be expected to distinguish between Scripture and its interpretation. In the Egyptian Greek context, however, the Septuagint’s general avoidance of exegetical additions and its emphasis on the formal representation of the Hebrew source text suggest that the audience had no such expectations. Instead, its creation appears to be driven by the larger community’s growing inability to understand Hebrew, a situation that required a written version in a much shorter time frame. The motivation for this undertaking was the Jewish communities’ need to understand their Scriptures as a guide for their lives, particularly in Diaspora contexts where access to these texts was important for maintaining their identity (Aejmelaeus 2018: 16).

Two Significant Trends

First, if the prevalent theories in the 2000s focused primarily on Alexandria, its intellectual sphere, and thus the library setting and perhaps even royal patronage, a significant portion of the research in the intervening time has shifted away from this focus. One need only look at an assessment made not long before the beginning of the period under survey: in his 2007 essay, van der Kooij noted that ‘recent research has seen a resurgence of interest in the perspective of the Letter of Aristeas’, later adding that ‘the view that the Greek version of the Pentateuch was initiated by the Ptolemaic court has recently become the prevailing view’ (2007: 292). This gradual change is due in large part to (1) the increased attention given to the Egyptian cultural and literary environment, instead of the Greek Alexandrian setting (Honigman 2023a: 1), and (2) the growing importance of the Septuagint’s linguistic data, which is seen as best interpreted against this background. Aitken is quite explicit in his justification for the foregrounding of these data: ‘Language is a guide, and a surer one than later legends, to the origins and social setting of the production of the translation, but . . . it needs to be examined by more sophisticated methods’ (Aitken 2014b: 125, 2020: 405). Highlighting another aspect of this shift in perspective, Wright also remarks that Aitken saw the Septuagint Pentateuch ‘as more properly Ptolemaic than peculiarly Jewish’ (Wright 2021a: 245). I suspect that ongoing scholarship on Septuagint origins will continue to advance in this direction, incorporating cultural and sociolinguistic studies while also examining the scribal traditions of Ptolemaic Egypt in relation to broader Second Temple Jewish scribal practices. This is a welcome development, as there is much that remains to be explored both within the field of Ptolemaic Egypt and its documents, as well as in the linguistic analysis of the Septuagint. Many of the trails opened up by Aitken should be pursued further and will likely lead to a more solid or nuanced portrait. For instance, Aitken placed much weight on the similarity in translation method between Egyptian non-literary translations and the Septuagint. However, the Egyptian evidence he surveys consists of a few legal documents and a limited number of features. The argument should be developed by examining a greater variety of texts (literary and non-literary) and features (as well as their frequency), especially since there is a developing corpus of scholarship in this area (for instance, Mairs 2018, 2020, 2024; Bowman et al. 2021).

Second, while locating the translators in Alexandria is still common in recent publications, the decentralization of scribal education and activity taking place in this period questions this ‘default’. Honigman argues that while much of the scribal and bookish activity took place in the circles of the Alexandrian court, manuscripts of the Book of Thoth, which share some affinities with Aristeas, have been found in the Fayum. This further suggests the presence of learned individuals, including Jews, in regions such as the Fayum and beyond (Honigman 2023a: 2). She also points out that many Jewish settlements outside of Alexandria, all the way into Middle Egypt, had regular contact with the Alexandrian community (Honigman 2023b: 201). Moreover, the king frequently traveled along the Nile, visiting various regions and assembling his officials (Clarysse 2000; Thompson 2017). In this connection, it is interesting to note that evidence indicates that both Ptolemy I and Ptolemy II made visits to the Fayum (Clarysse 2000: 31, 40-41). The decentralization of the administration, therefore, combined with the mobility of the elite class—whose networks, as Zenon demonstrates, could extend beyond Egypt—suggests that competition for influence and prestige at the court could also occur far from Alexandria (Honigman 2023a, 2023b). This makes it plausible to envision a scenario in which scribes within the administration maintained regular contact with one another, with Alexandria, and with the royal entourage. Consequently, it becomes much more difficult to pinpoint the location of the translation, whether that is Alexandria or elsewhere in Egypt, and the possible interactions with Palestine. This is true even in a scenario where the translation is motivated by a desire to gain influence with the king (Honigman 2021). All things being equal, Alexandria remains more probable, though not necessarily in connection with the library or Mouseion. Future research will surely nuance this picture further.

Avenues of Research

Two other issues stand out among the many lines of inquiry warranting further research. In the first place, there are numerous indications that the Pentateuch translators were very familiar with the corpus they were translating. There are two aspects to this observation:

Second, is it still appropriate to conceive of the Septuagint Pentateuch as a single translation project, even when undertaken by a plurality of individuals, as is often assumed? Lee’s proposal (2018: 173-209) that the shared technical vocabulary points to a form of collaboration between translators has stimulated further discussions in this area but has also received significant pushback. For one, it precludes other plausible explanations, such as the possibility of a multi-stage process where later translators based their lexical choices on an earlier translation, or word lists preserved for this purpose (Rösel 2016b; Fresch 2021: 36-38). A few studies have been published on possible borrowings of later translations (e.g., Deuteronomy) from earlier ones (e.g., Exodus), some entering the realm of probability (Maurais 2024a). Lee’s data can therefore fit a two-stage process, where Genesis and Exodus are translated first, and the other books follow (but not necessarily simultaneously). This explanation would also fit with van der Louw’s recently published thesis (van der Louw 2019) to the effect that the books of Genesis and the initial chapters of Exodus (at minimum) were translated by the same individual.

Although Lee’s thought-provoking observations highlight striking commonalities, significant obstacles remain to viewing the Septuagint Pentateuch as a single translation project. In fact, important changes in the translation approach can be observed between these two groups, with Deuteronomy applying a source-oriented approach more rigidly, overall, than the other books of this corpus (e.g., Aejmelaeus 1982). This is supported by Recalcati’s research (2024). She observes that Leviticus contains considerably fewer instances of Egyptian influence on proper names, toponyms, and other vocabulary, not only due to the nature of the book’s contents, but also because of a different translation approach (2024: 259-60). In fact, besides chapter 11 and its list of animal names, its translator adopted a more ‘cautious approach’, perhaps to preserve the book’s contents from ‘foreign cultural influences’ (2024: 260). Interestingly, her colleague working on Numbers and Deuteronomy comes to similar conclusions (Naveau 2024). In his work on Deuteronomy, Maurais (2022: 166-73) found little concern to adapt legal material to its Ptolemaic context. In fact, its translator ignores the legal terminology that exists for situations similar to the levirate law (Deut 25.5-10) and instead prioritizes harmonization with other Torah texts dealing with a similar issue. This offers a stark contrast to the approach taken in Exodus, as documented by Korytko, where an opposite concern seems to be at work. In sum, differences exist not only in terms of style and translation technique, more generally, but also in terms of attitudes towards the target culture and the translation’s function (as Korytko 2023: 264 concludes, tentatively). This poses a significant challenge to theories that view the Septuagint Pentateuch as a single, coherent project, whether commissioned for a specific purpose or authorized by a particular entity—be it priestly authority, as suggested by van der Kooij, or as a community-driven project, as proposed by Aejmelaeus. I would venture that future studies will likely confirm the existence of at least two distinct ‘projects’. While these projects may originate from similar scribal circles (as per Aitken and Honigman), they reflect divergent approaches and aims, and therefore differences in contexts and readerships. Should this research bear out, it will become increasingly difficult to argue for a singular motivation (or set of motivations), location, or date for the collection (as already suggested by Rösel 1994, 2018: 16). Instead, the Septuagint Pentateuch might be more productively studied as a developing corpus, with new data uncovering additional layers of complexity and offering ample material to engage Septuagintalists for many decades to come.