I have added the following article by Ann Taylor to the bibliography at Greek-Language.com:
“The change from SOV to SVO in Ancient Greek.” Language Variation and Change. 6.1 (1994) 1-37.
While the order of major sentence constituents is quite free at every stage in the development of Ancient Greek, the distribution of those constituents is not random at any stage, and one particular constituent order can be shown to be dominant at each stage. Taylor argues that the dominant constituent order was verb-final (SOV) in Homer, but changed to verb-medial (SVO) by the Hellenistic period.
Using the paradigm of Kroch (1989), Taylor constructs two models—one for the verb-final grammar of the Homeric period (before 800 B.C.) and one for the verb-medial grammar of the Hellenistic Koiné (c. 100 A.D.). She describes the intervening period (Herodotus, c. 450 B.C.) as in part like Homer and in part like the Koiné. She shows further that the ratio of these two constituent orders in Herodotus is also supported by an independent measure of the distribution of weak pronouns and clitics.
Thank you for the reference.
How does she deal with the perspective that sees Herodotus as VSO (H. Dik, 1995) and Koine as also VSO (many who worked on Greek discourse in the 70’s-80’s, including myself)?
Taylor’s article was published in 1994, a year before Dik’s work, so I can only speculate about how she might have dealt with her arguments. Since Taylor presents Herodotus as representing a transitional phase in the shift from SOV to SVO, I suspect she may have agreed with much of what Helma Dik had to say, but was less convinced that the evidence pointed definitively to SOV as the dominant pattern.
I am not finished reading Taylor’s article, so I can’t speak with certainty. I can say that she is using a different linguistic paradigm, and that I’m not convinced that her arguments are definitive. While there are abundant examples of SVO clauses in the Hellenistic period, there are still good reasons for seeing the basic pattern as remaining SOV.
By the end of the Hellenistic period the attempt of upper class writers to imitate the earlier Attic writers makes their writing clearly SOV. See Athenasius of Alexandria (b. ca. 296-298 – d. 2 May 373), for example, at the end of this period. But these writers clearly do not represent the common speech of the period.
I’ll see what I think when I finish the article. The copy I have is a PDF made from a very bad photocopy (http://www.ling.upenn.edu/~kroch/hist-pdf/taylor-gr.pdf). It’s tough going!
I don’t know how relevant it is to ancient Greek, per se, but in modern Hebrew there are phrases for which we use verb-subject as the standard order, although in the greater scheme of things it is subject-verb. For example, it is very non-standard to hear someone say גשם ירד בצפון (géshem yarad ba-tsafon), while it is colloquial and standard to hear ירד גשם בצפון (yarad géshem ba-tsafon), for “it rained [lit. fell-rain] in the north.” This phrase remains in the expressed order in almost all instances, though it reverses in a question regarding quantity (“how much rain fell?”), which becomes ?כמה גשם ירד בצפון (káma géshem yarad ba-tsafon?). Word order in Hebrew is still more flexible in this matter than in English.
Word order is something that I struggle with in Greek composition, though I tend to follow Latin’s regular structure (and it feels good to me), holding the verb until the end of the phrase or at least until the standard-order pieces of information are given. This is similar to the structure of the participial phrase in Philemon 4:
εὐχαριστῶ τῷ θεῷ μου πάντοτε [[μνείαν σου ποιούμενος ἐπὶ τῶν προσευχῶν μου]]…
The object (μνείαν σου) appears before the verb (ποιούμενος), which is then followed by an additional adverbial clause as a prepositional phrase. We’ve discussed this verse the past couple of days on CARM’s biblical languages forum, regarding the attachment of πάντοτε and to which verb it belongs (the main verb, εὐχαριστῶ) or the participle (ποιούμενος). My own opinion is that it should be read with the main verb, while the participle should be translated in a clear circumstantial force – thus, “I always give thanks to my God [[when I make mention of you in my prayers]]…”
Anyway, I’ve rambled a bit. All that to say that word order still stumps me. 😉