It has been a little over a year since James Aitken passed away. Anyone who knew Jim knew that he had a few key passions. He loved fine wines, and well-prepared foods (‘not too greasy’). He loved traveling to the continent, especially Germany. He also had a passion for the Septuagint, which he believed deserved more attention. Many of us who studied with him can testify that we found ourselves suddenly in Cambridge, slightly under-qualified, wondering how we got there. It was Jim: he wanted to see us make the contributions he himself could not make in the course of a single career. If we had the right tools in place, he would provide the mentorship.

One neglected aspect of Jim’s work was the value he placed on Greek diachrony. To give you a picture: in our very first conversation he recommended I read GB Shipp, and in our last supervision we discussed the Constantinople Pentateuch. I had just obtained a hard-won copy of it, and Jim quickly fetched a book off his shelf and told me the exact page numbers I needed to begin research. Jim’s work focused heavily on placing the Septuagint in the context of classical and postclassical sources, especially documentary papyri. But prior to his passing, he had begun to focus more on later periods of the Greek language (and praised others, like John Lee, for doing so too).

So as a tribute to Jim, let’s talk about the notion of ‘Hebraisms’ in the Septuagint and how later periods of the Greek language can kindle fresh analysis. Let’s start with Deut. 1:17:

Notice the use of ἀπό for comparative מן (Muraoka 2009: 71; Soisalon-Soininen 1979: 87). Greek speakers reading this are probably thinking, “So what? We say this all the time.” And that’s right (Thackeray 1905: 23):

κοστίζει περισσότερο από ένα αυτοκίνητο.

It costs more than a car.

But Deut. 1:17 is at least 1000 years before the fact. As far as I can tell, comparative ἀπό doesn’t develop in the Greek language until the late medieval period. This makes Deut 1.17 (and others like it: Deut. 17:8; 3 King. 19:7; 4 King. 6:1; Sirach 24:29) very curious. We have in Jewish Greek literature a ‘Hebraism’ that preempts a much later development in the Greek language. What is happening here?

It seems like there are two options: (1) The translator calqued the source text here but not elsewhere (for something incredibly basic!), or (2) It is an instance of proleptic contact-induced change (Heine & Kuteva 2003 and Ziegeler 2016). Proleptic contact-induced change (or replica grammaticalization) describes how a receptor language borrows patterns of grammaticalization or polysemy from a donor language, for the same reasons the donor language developed them naturally. For example, ablative markers are “one of the most common, if not the most common, means of encoding standard noun phrases in comparative constructions” (Heine & Kuteva 2022: 36). Comparative ἀπό appeared in Jewish Greek texts because contact with Hebrew preempted a language-internal process in Greek: ABLATIVE > COMPARATIVE. Even in cases where real change takes place, the term ‘Hebraism’ fails to capture anything interesting about it.

Another alleged Hebraism is the light verb construction with ποιέω:

וַיַּ֧עַשׂ לְאָבִ֛יו אֵ֖בֶל שִׁבְעַ֥ת יָמִֽים

And he made mourning for his father seven days (Gen. 50:10).

καὶ ἐποίησεν τὸ πένθος τῷ πατρὶ αὐτοῦ ἑπτὰ ἡμέρας.

And he made mourning for his father 7 days.

In the article linked above, Marieke Dhont notes this construction, although borrowed from Biblical Hebrew, is found not only in postclassical sources but in Modern Greek as well. Look at Filos (1994):

και ο Iωσήφ έκανε για τον πατέρα του πένθος επτά ημέρες.

And Joseph made mourning for his father seven days.

The light verb construction is present in all periods of the Greek language. And it’s not surprising why. It’s extremely common cross-linguistically:

明日、私 は ギリシャ 語 を 勉強 する。

Ashita, watashi wa Girisha go o benkyou suru.

tomorrow I _TOP Greek _{LANGUAGE OBJ} study do

Tomorrow, I will study Greek.

That a construction was borrowed from the source does not make it a ‘Hebraism’. The final example comes from the so-called ‘finite participle’ in Ex. 15:18 (cf. Ex. 12:37 B; Jdg. 4:16, 13:19, 20, 14:4):

יְהוָ֥ה ׀ יִמְלֹ֖ךְ לְעֹלָ֥ם וָעֶֽד

YHWH will reign forever and ever.

κύριος βασιλεύων τὸν αἰῶνα καὶ ἐπ̓ αἰῶνα καὶ ἔτι. 

The Lord reigning forever, and evermore.

Finite participles are a more curious case. They may be instances of anacoluthon, an online glitch in speech production. The speaker failed to commit to the utterance for some reason––either because they realized the sentence was not true or aborted their discursive strategy. The result is an ungrammatical utterance. That’s unlikely here, since the translator had a yiqtol in front of him, and knew its semantics well.

Two options stand out. One is that the copula has been elided and this token is actually a periphrastic construction (e.g. βασιλεύων ἔσται). Another is that the finite participle is a pragmatic expansion of an absolute construction, resulting in a syntactic blend (Aarts 2007: 187). Or perhaps the answer lies between the two. For our purposes, it is worth noting the early medieval historian John Malalas (6th CE) also has finite participles (375: 84):

ὁ δὲ Περσῶν βασιλεὺς Κωάδης δεξάεμος Ἑρμογένην μάγιστρον μετὰ καὶ δώρων τῆς ἀναγορεύσεως τοῦ βασιλέως ʼΙουστινιάνου ἐν μηνὶ ʼΙουλίῳ.

But the king of the Persians, Koades, receiving [=received] Hermogenes the magus along with gifts on the occasion of the proclamation of King Justinian in the month of July.

As Geoffrey Horrocks notes (2010: 245): “[W]hile Malalas’ Greek is a great deal closer to the spoken language…it is still very much a written style.” If we dismiss the participle in Ex. 15:18, we should dismiss it John Malalas. And no one does that. Clearly something more is happening than a mere solecism (‘a misuse of the participle’). The Constantinople Pentateuch (1547) uses the future-referring particle να (which is typical for yiqtol):

אוֹ קִירִוֹש נָא בָאשִלֵיִיכִי אִיש נֵיוֹנָא פַנְטוֹטֵיש

ὁ κύριος νὰ βασιλεύγῃ εἰς ναιῶνα πάντοτες.

The Lord will reign forever and ever.

I don’t yet have an answer. But I suspect, as Jim often did, it will come from the diachrony of the Greek language itself.