How has the word “pou” (που) been used in Greek, historically, throughout the various dialects?

God bless Khateeb, he’s actually asking me what I found in my PhD. Without me bribing him. And I’ve forgotten to reupload my thesis; Khateeb, remind me to do it if I haven’t done it in the next week.

There will be some jargon here, but I’ll try to keep it high level.

που (< ὅπου “where”) is a relative conjunction (I saw the guy that came here). It can also be used in Standard Greek to introduce some clausal complements, but not all (I am happy that you came, but not I hope that you came. I hear you walking, but not I heard that you walked. You walked in—I already knew that, but not I know that you walked in.) And especially in older vernacular Greek, it was a catch-all conjunction in general: it could be used to mean “when”, “because”, “although”, etc.

If any classicists are reading this: its range of functions is pretty similar to those of the Ancient participle.

The unifying principle seems to be that whatever clause it introduces is taken as given, in some sense. (Some very slippery, hard to express sense.) It is true and real in the world, or it is presupposed, or it is phrased as if it is presupposed.

• Τhere’s a wonderful exception in “the hell I did!”-type statements, where the speaker is making fun of the claim being presupposed by someone else: βρε άει στο διάολο που πήγα. Nicholas, N. 2005. pu-based Greek Rude Negators. Journal of Greek Linguistics 6. 103-150; http://www.opoudjis.net/Work/xez…

What’s the diversity in dialect? High level:

• Lots and lots of dialects have generalised the complement use to all real complements: they can use it to say “I heard that you walked” or “I know that you walked”. Constantinople dialect is the best known one; see http://www.opoudjis.net/Work/com…. There’s no obvious connection between the dialects, and I think it’s just independent drift: it’s the kind of construction that would generalise anyway.
• For the conjunction use, the dialects generally align, but the mainland dialects align much more closely than the island dialects do. Some island dialects do strange things with που as a conjunction, especially moving it away from given clauses, to more hypothetical clauses (e.g. “whenever”).

My irresponsible hypothesis on the latter: mainland dialects are more homogeneous grammatically, because of unimpeded communication between regions and bilingualism: analogy was more pervasive in levelling out eccentric developments, and promoting internal consistency (που always referring to given situations, etc). The islands are more eccentric linguistically in general (including more archaic): they are more isolated, and have not had any bilingualism. So they had less pressure to level out eccentricities.