Aristides Quintilianus has some fascinating instances of ἐκ and εἰς that use the source-path-goal (SPG) construction, like this one in example (1).

1. Μεταβολὴ δέ ἐστι ῥυθμικὴ ῥυθμῶν ἀλλοίωσις ἣ ἀγωγῆς. γίνονται δὲ μεταβολαὶ κατὰ τρόπους δώδεκα· κατ’ ἀγωγήν· κατὰ λόγον ποδικόν, ὅταν ἐξ ἑνός εἰς ἕνα μεταβαίνῃ λόγον ἢ ὅταν ἐξ ἑνός εἰς πλείους ἢ ὅταν ἐξ ἀσυνθέτου εἰς μικτόν ἢ ἐκ ῥητοῦ εἰς ἄλογον ἢ ἐξ ἀλόγου εἰς ἄλογον ἣ ἐκ τῶν ἀντιθέσει διαφερόντων εἰς ἀλλήλους ἢ ἐκ μικτοῡ εἰς μικτόν.
   Rhythmical modulation is alteration of rhythm or of tempo. Such Modulations occur in twelve ways: according to tempo, according foot ratio — when it changes from one ratio to another or from one to several or from incomposite to mixed or from structured to erratic or from erratic to erratic or from one to another of those differing by antithesis or from mixed to mixed (Aristides Quintilianus, On Music, 1.19).

This particular example is also rather significant. Narrowly speaking, this example is significant because abstract (not spatial or temporal) uses of the the SPG construction are relatively difficult to find in post-classical Greek. So much of the documentary papyri involves people going about their lives and interacting with physical objects in space. The best figurative example in the papyri tend to be instances where the SPG construction is involves a part-whole relationship, such as here where someone designates a portion of property from a larger piece land to another person, as in example (2).

2. ἀπογρά(φομαι) … ἐπιστρατηγήσαντος ἣν ἔχω ἄβροχ[νο] καὶ ἐπηντλ(ημένην) περὶ κώμην Παλῶ ἐκ τ(οῦ) Ἄνδρωνος σὺν τῷ Μενεσθ(έως) κλ(ήρῳ) εἰς Πέτσειριν Ἡρακλήου α (ἀρτάβης) ἀβρόχ(ου) (ἀρούρας) βδʹ
   I register … the unwatered and irrigated land which I own: in the area of the village of Palosisin from the holding of Andron with that of Menestheus, to Petsiris son of Heracleüs, rated at 1 artaba, unwatered, 2¼ arurae (P.Oxy 1459 Return of Unwatered Land  A.D. 226).

Here, motion through space is replaced with transfer of ownership.

But the text from Aristides Quintilianus above in (1) does something else for us, too. It gives us an example of usage of the much more uncommon SPG construction, where both prepositions have the same object/Landmark and that landmark is nonspatial. These are super rare and hard to find. And they matter because they have a distinct semantic structure—come to our SBL paper in November if you want to hear about that.

To find these, You rely on proximity searches across text and wade through hundreds/thousands of false hits. Yet, they are relatively easy to find in the LXX & NT since those those texts are well tagged. But there’s no button that automatically provides you with only the abstract nouns. That data analysis hasn’t been done. It’s like pulling teeth. I spent last week going through ~800 tokens from the Perseus Project and never found one in postclassical Greek text, because that stuff is mostly geography and history. Fewer abstract nouns. How’d I find this one from Quintilianus? A journal article cited it and I just happened to be reading int he right place at the right time. The problem is that they didn’t actually give a citation, only the text and the author and title of the work. I then had to ask a friend to search the TLG for the actual reference and they were kind enough to oblige.

But the larger point is that we need better data and we need better data access. We also need better tools for annotating that data openly so that important and useful grammatical patterns can be more readily documented and included in in grammars.