And one more thing: no folk singer would sing „Μηδέ εξηκο…, μηδέ ακριβο…“. The „e“ sounds with no consonant between them would make him wince. Even if he were singing from a sheet of paper someone had given him—which he wouldn’t have been at that time…—he would add the „n“ sounds between the vowels himself, without a second thought and without asking anyone.
Indeed, such hiatuses do not occur in folk song.
That does not mean he would stick an -ν- in just anywhere. Nor does that ever happen in folk song. Euphonically inserted -ν- sounds do of course occur frequently, precisely to resolve hiatus, but not just anywhere. Specifically, I do not think that “μηδέ” could ever become “μηδέν”.
As for the second hemistich, the correct reading might be “μηδ’ ακριβοπουλούσα” (since I am not convinced by #18’s interpretation either: why should it matter to us if “many” dialects drop the intervening “consonant”? The question is whether in Gortynia they specifically drop the γ from this particular ending, given that we have first established that the ending itself—αγα, clearly not pan-Greek—is used in that area. The “να μην ακριβοπουλείς;” in the preceding line does not make this likely, but of course I do not know the Gortynia dialect…).
As for the first, I don’t know. Perhaps some rewording of the whole phrase.
Quite right. And μηδέ does not become μηδέν: ordinary people know the meaning of these two words very well and would not make such a mistake. However, they can quite naturally say
Μηδέ ‘ν’ ακριβοπούλαγα
The euphonetic ν is placed before ακριβοπούλαγα; it is not added to μηδέ, just as people pronounce
Ν’ ο κυρ βορηάς παρείγγηλε ν’ ούλω τω γκαραβιώνε
(note: ν’ ούλω, not παρείγγηλεν)
Ah, you’ve reminded me of something now!
My late brother spent a lot of time, as a student, with our uncle George, our father’s elder brother. Having heard that our uncle had worked at the Ministry of Press under Metaxas, he asked him what he had done in connection with the ministry’s guidelines for the lyrics submitted for approval before recording could go ahead. His answer: — I corrected Tsitsanis’s hiatuses.
Niko, what you are arguing for is precisely the «anywhere» that we agreed isn’t possible:
There is absolutely no difference in pronunciation between «μηδέν ακριβοπούλαγα» and «μηδέ ‘ν’ ακριβοπούλαγα». And for that reason, an euphonic -ν will never be added after «μηδέ». Unless there really is a recorded example, in which case of course I will retract what I said. But an example with -ν after «μηδέ», not with -ν just anywhere in general, like the others you cited.
As for the other point: of course it says «παρήγγειλεν»; I am at a loss to see how you can allow room for any other assumption.* Euphonic -ν in verbs can be added to the same forms in which it has been added since ancient times. You will never see «παρήγγειλα -ν’» (i.e. first-person singular). You will see «είπεν ότι», «είπαμεν ότι», but not «είπατεν ότι».
All of the above applies to prose as well; these are not special rules of folk song. By contrast, «Ν’ ο κυρ-Βοριάς» is a feature of song, not of the spoken language, and in any case it is unrelated to euphony and the avoidance of hiatus, since before the vowel there is neither another vowel nor anything else. I do not know how it is to be interpreted, but I can accept the possibility that it can be interpreted and therefore exists (if it cannot be interpreted, it does not exist), because I know of a similar case that constitutes a regular pattern: in Kalymnian song, a line beginning with a vowel can acquire an extra μ-, entirely independently of which vowel it is and what the word is. It is simply an idiom of local singing. So there may be another similar case with ν- somewhere else.
*: It is possible that the written form with ‘ν’ was recorded by someone who, without having heard or paid attention to many different varieties of Modern Greek, simply thought, «παρήγγειλεν is ancient and katharevousa, so a Modern Greek speaker wouldn’t say it». But the thing is, they very often do!