Weeell… in the Ottoman Empire (and in the Byzantine Empire before it), identity was primarily credal, organised as Millets (Ottoman Empire). As far as everyone in the Ottoman Empire was concerned, there were:
- Muslims
- Franks (Catholic)
- Romans (Orthodox)
- Armenians
- Jews
See Albanians or Bulgarians in that list? Me neither. In fact, Bulgarians were only able to assert a distinct national identity by establishing a distinct ecclesiastical identity, through the Bulgarian Exarchate.
Ethnicity as we understand it did not factor much in how people understood identity. The Catholics of the Greek islands were ethnic Greeks, as were the Muslims of Crete; that didn’t matter. In the Millet way of thinking, Markos Vamvakaris was as much a Frank as Édith Piaf; and no distinction was to be made between a Muslim from Iraklio and a Muslim from Damascus.
That delayed the establishment of national identities in the Balkans. People were aware of the Albanian language, and had words for Albanians, for example; but Albanians were either Franks, Romans, or Muslims first, and Albanians second. That’s why it was so important for the Albanian nationalists to assert “the religion of Albania is Albanianism”. And why Greeks historically used the odd construction “Turk-Albanians” (who are just Muslim Albanians).
So. There are a lot of longstanding ethnic minorities in Greece, who identified as Romans through the Byzantine and the Ottoman periods, and transitioned to a Greek national identity after Greek Independence. That includes Arvanites (Albanian-speaking) and Aromanians (Romance-speaking), who may not have spoken a word of Greek but fought the Turks, because their identity was Orthodox.
Things with Slavic-speakers got a lot more contentious, of course, and I’ll wisely decide to avoid getting into it. There was some contentiousness with Aromanians too. But I will mention the entertaining case of the Kızderbent Trakatroukides:
- Settled in Northeastern Turkey from southern Bulgaria.
- Unlike the other Bulgarian villages of Northeastern Turkey, did not join the Bulgarian Exarchate but stayed with the Greek Orthodox church.
- Therefore did not join the other Bulgarians in leaving Turkey for Bulgaria in 1919.
- Therefore instead were persecuted as Greeks and fled to Greece in 1923.
- And the majority of them were settled in the village of Polypetro…
- … whose local population spoke Makedonski.
The sum of the anecdotes is: Greek identity may not be ethnoreligious now, but “Roman” identity was ethnoreligious for a long time. Greek identity is in many ways a successor to “Roman” identity, so who ended up called “Greek” was not unrelated to who was called “Roman”.
And of course those historically non-Grecophone populations get very touchy if you tell them they aren’t Greek.
EDIT: Dimitra edited her answer in light of mine. I am now reciprocating.
Let’s set some parameters. There were three names in play.
* Romans (Ρωμιοί), which meant the Rum millet, and included everyone Greek Orthodox. I’ll claim that’s the ethnoreligious identity.
* Hellenes (Έλληνες), which meant the Ancient Greeks, and which the intelligentsia towards the end of the Ottomans started promoting over the other two names. That’s the modern name for Greeks; and though that may not have been the original intent it includes Greek nationals of all creeds (including Jews and Armenians—and though many Greeks may wince to acknowledge it, Muslims as well). I’ll claim that that’s the civic identity. Inasmuch as is there such a thing in Greece.
* Graikoi (Γραικοί), which was used less than the other names, but which I have seen used for ethnic Greeks. (The Aromanian writer Giorgos Exarchos is a quite loyal Greek, but he makes a point of distinguishing Graikoi Hellenes from Vlach Hellenes.) The word of course is just Greek for “Greek”. I’ll claim that that’s the ethnic identity.
Dimitra’s conclusion is that Greeks in Ottoman times were not ethnoreligious, because “If Greek Then Orthodox” did not map to “If Orthodox Then Greek”.
Now the equation “If Roman Then Orthodox” did map to “If Orthodox Then Roman”. Roman was an ethnoreligious identity; and Bulgarians, Albanians and Aromanians were Roman.
It is true that the Graikoi were a privileged group within the Romans. They had control of the Patriarchate (which ran the Rum millet), and they were quite happy for Graikoi and Romans to be conflated, at the expense of the other ethnicities.
BUT the culture they were privileging wasn’t a culture of modern vernacular Greek: it was Hellenic culture (Ancient Greek). And here’s the catch: Graikoi were only somewhat more privileged in their access to Hellenic culture than Bulgarians and Albanians. Remember that people did not always consider Ancient Greek the same language as Modern Greek. Greeks themselves called Modern Greek Romaic until independence; and their Western contemporaries often did too. And if Graikoi had to learn how to write in good Hellenic, well, so could Albanians and Bulgarians. (And of course, they did.)
So the equation Hellenes = Graikoi was disrupted: non-Grecophones had access to Hellenic culture. The equation Romans = Graikoi was disrupted: non-Grecophones were Orthodox. And the equation Hellenes = Romans was disrupted: the true Hellenes were pagans, and people were well aware of the discontinuity between Ancient and Mediaeval civilisation.
But the equation Roman iff Orthodox, of course, was not disrupted: it was a definition.
(It is now that Romios means something different—alignment with Greek identity through low rather than high culture—Hellene being the high culture, of course. Vamvakaris was a Frank not a Roman, but Greeks now will happily claim him as a Romios. I’ve heard Greek Christians claim that any Greek Jew who loves Kazantzidis must be a Romios.)
So:
Was the definition of Roman = Orthodox forced on Graikoi, Bulgarians, Albanians, Aromanians etc by the Ottomans? I suspect it wasn’t, and that it was inherited from Byzantium. If you were Orthodox, you followed the Emperor’s creed; if you were Catholic or Muslim, you were a foreigner, and if you were Jewish, you were a second class citizen. I don’t think the Byzantines overly worried about what your ethnicity was, whether you were Bulgarian (like John Koukouzelis) or Georgian (like John Tzetzes) or Armenian (like half the emperors). As long as you were Orthodox and wrote in Ancientish Greek, you were a good subject of the Roman Empire.
Were the Graikoi or the Bulgarians worried about ethnicity in Ottoman times? Again, I don’t think so. Partly because the term Graikoi was so rarely used to begin with, partly because you don’t hear much mention of the minority ethnicities at all.
I *think* I’m in agreement with Dimitra overall: Greeks as Hellenes are a civic identity now (Greek Jew, Έλληνας Εβραίος, is not a contradiction in terms); Greeks as Romans were an ethnoreligious identity back in the day. And this having been a reasonably recent transition, it has been a little bumpy.
“Greek Muslim”. Έλληνας Μουσουλμάνος. I don’t think I’m wrong in saying that Christian Greeks are still uncomfortable with the concept.
(Google gives me 177 hits for Έλληνας Μουσουλμάνος, vs 2170 hits for Έλληνας Εβραίος. There are 98,000 Muslims in Greece and 5,000 Jews in Greece.)