Ἡλληνιστεύκοντος

This is a complicated story no matter how one tells it and from what vantage point, but I’ve put this off a few months too many. So, following on from the preceding post on the Trakatroukides, let’s take the story from the top. Again, thanks to Butcher of Yore for doing all the research.

After the Greek-Turkish war of 1919–1922 and population exchanges of 1924, thousands of refugees settled in Greece. These refugees were deemed Greek because they were Greek Orthodox, but they did not all speak a form of Greek. As I wrote at the Other Place, a good number of them spoke Turkish: over 100,000 first language speakers out of 1.3 million refugees—191,254 reporting Turkish as their language in the 1928 census (Google Books excerpt below), minus 84,669 indigenous Turks in Thrace (p. 17 of Featherstone & Papadimitriou’s presentation):

But language was not what the Ottomans or their subjects used to define identity.

The villages that were abandoned are now dwelled in by the Muslims that stayed, and one such village is Kızderbent, “Maidenpass”, in the region Greeks call Bithynia, 28 km away from Nicaea, which the Turks call İznik.

View Larger Map

The refugees from Kızderbent, who go by the name of Trakatroukides or Rokatzides, were scattered in Greece. There are two Facebook groups for refugees from the village (1, 2), and per this post on the second, most settled in Valtotopi, Kilkis prefecture; others ended up in the nearby villages of Ryzia, Gerakona, Fillyria, and Polypetro, but also further afield, in Nikomidino, Kallithea (Thessalonica prefecture), New Triglia (Chalkidike), and Roditis (Rodopi): this local newspaper article is on a get-together of the Rodopi Trakatroukides.

View Trakatroukides in a larger map
(Touchingly, the Greek Facebook page for those who have left Kızderbent links across to the Turkish Facebook page, for those who live there now.)

The Greek villagers of Kızderbent call it Κιζδερβέντ, Kizdhervent, and there is a copious variety of transliterations: Kizderved Kisderved Kizdervent Kisdervent Kizderbed Kisderbed Kizderbent Kisderbent.

And Къздервент. Because there’s something notable about the language of Maidenpass. It wasn’t Greek; or at least, it wasn’t just Greek. We have little left of their language—and I’d certainly never heard of it during my preoccupation with Balkan linguistics; but from the Motley Word, a site I reviewed a while back, we have some 500 words. (To see them all, click on Όλες in the interface.) A lot of the words are Slavonic, and a lot of the words are Turkish, and none are noticeably Greek. That could be a selection bias—the Motley Work lexicographer (who is still going!) may just be singling out the Slavonic and Turkish words. But from the evidence I’ve seen elsewhere (see the followup post), I doubt there’s a substantial amount of Greek vocabulary.

There was also discussion of the language of Maidenpass on a thread at Sarantakos’, which triggered this (and which I’ve already posted): the indication there was that there was no mutual intelligibility between Trakatroukika and Greek. From that thread, we also have a sentence of Trakatroukika, a song verse: tákar tákar ke priataláh duí sulísperi zebé khantúm

The indications are that the language of this village was somehow a mixture of Balkan Slavonic, and Turkish. At least one glancing mention, from the website of the municopality of New Triglia, implies Armenian was there as well, but I have not seen a direct confirmation. (I have seen something more indirect, on which see further down.) The Slavonic component is conventionally called Bulgarian, and certainly when Slavists visited Kızderbent in the 19th century, noone was making a distinction between Bulgarian and Macedonian; but there is at least one account saying the village was settled from Ohrid, and the language spoken in Ohrid is not now called Bulgarian.

The name the villagers went by, Trakatroukides, doesn’t give us much of a hint: as TAK anticipated in comments, it derives from traka truka, an onomatopoeia of firecrackers (as the Triantafyllidis dictionary gives it)—so “bang bang”. The onomatopoeia is also associated with grinding knives (carnival song: τράκα τρούκα τη μαχαίρα, a verse that has wandered into Rebetika referring to knives), and a children’s song about two steamships colliding (“The Greek boat has lemons. The Turkish boat has eggs. The boats go ‘bang gang’. The sea is avgolemono“).

The alternate name Rokatzides doesn’t help either: ρόκα means spindle or rocket, arugula (Italian rocca—hence also English rocket; or ruca). It’s tempting to associate “rocket” with “bang bang”, but I’d say that’s an anachronism—the association of spindles with rockets is late and Western.

The web site of New Triglia has this to say on the origins and language of the Trakatroukides of Kızderbent—the Bangbangers of Maidenpass:

There are three views on the founding of Kizdhervent. One claims that Greeks from Ohrid and Monastir [Bitola] migrated to Constantinople seeking a place to settle. The other opinion is that they did not migrate but were exiled because they robbed the nearby inhabitants. The third is that they were prisoners of Mehmed the Conquerer, who transported them to Constantinople and from there to Kis-Han, as he passed by Ohrid returning from Kroja. […]
Later on, because of the privileges the Sultan gave the area, Greek-speakers, Turkish-speakers, Armenian-speakers and Bulgarian-speakers settled the region, and that’s how Kizdhervent was created. The mixture of these peoples is also the reason why its peculiar language came about, spoken by no other village in Asia Minor. For that reason they were called Trakatroukides.
Their customs and religious habits are the same as other Greeks’, and their religion is Christian Orthodox.

The Bulgarian Wikipedia article has a somewhat different story, which is worth giving in detail:

Kăzdervent was the first village of Asia Minor Bulgarians attested by written evidence in the 19th century. [Accounts from 1807 on, almost all of which describe the village as speaking Bulgarian.]
According to stories by the Kăzderventsi the village existed in the 18th century, and is one of 12 villages donated by the Grand Vizier to a lady in the seraglio: she sold them to an Armenian banker, and he sold them back to the Turks one by one.
Konstantin Jireček later wrote of Kăzdervent, and inferred that they originated from Momim Prohod, since that is what Kăzdervent means in Turkish. According to Jireček (1899) the villagers “speak Turkish and half are already on the way to losing their nationality entirely.” Vasil Kănchov writes that the village had 440 houses, and believes that the Kăzderventsi were involuntary migrants from Ohrid.
At the end of the 19th century Kăzdervent was gradually Turcisised linguistically and Hellenised in national consciousness as a result of intermarriage and influence from other Karamanlides’ villages.

So we have in Kızderbent a village which spoke Bulgarian (or Macedonian Slavonic); which ended speaking something between Bulgarian and Turkish (somehow), and was notorious in its region for “talking funny”. (We have a report the Greeks and Turks in the area thought so; did the Bulgarians?) By 1914 however, the villagers did not consider themselves Bulgarians or Turks, but Greeks, and did not follow the Anatolian Bulgarians’ exodus from Turkey; they left Turkey with the other Greeks in 1922.

The history behind that is obviously complicated, and needs a separate post. Here, I’m going to pose some linguistic questions. I don’t have the answers (although I do hope someone who knows their South Slavic does go through the Motley Word list, and the language samples from the next post). For all I know, there have been Bulgarian studies of Trakatroukika; but I’ll speculate anyway.

The most important question is, what sort of a mixture was Trakatroukika? If the situation was something like the New Triglia article paints, with Armenians, Greeks, Turks and Bulgarians cohabiting—and no single language dominating, we might end up with a pidgin. Pidgins arise as contact languages all the time, between people in recurrent but not close contact. For such a pidgin to become the main language of a village though, it would have to make its way into the hearth. That would mean routine intermarriage (possible for the Greeks, Bulgarians, and Turkish-speakers if they were Christian too).

It would also only make sense if multilingualism was impractical. That happens on slave plantations perhaps, which is how pidgins become creoles. But it didn’t happen in Papua New Guinea, because the hearth can be multilingual; and although it may not now look like it, the peoples of the Balkans used to be multilingual as well. So the drastic breakdown of language which you get with pidgins—and the rebuilding you get with creoles—seem to me unlikely for Kızderbent. We’d know whether that had happened of course the minute a Slavist sees an actual sentence of Trakatroukika (like the one I pasted above). A pidgin or a creole would have very little left of the morphology of Bulgarian, and even less of its syntax.

True mixed languages, like Michif—the verbs are Cree, the nouns are French—are even more rare, more spectacular, and certainly more controversial. But they don’t result from language breakdown, but fluent multilingualism: enough villagers speak both languages that not only words, but grammar ends up seeping between the two. It’s worth noting that Wikipedia’s list is short, but includes Cappadocian Greek—because Cappadocian was not just borrowing words from Turkish, but vowel harmony and inflections: elements of the basic machinery of Turkish.

That said, mixed languages are rarely stable anyway, and at least some of them (I’d say including Cappadocian) are an attention-grabbing twist on language shift—language death, if you prefer. That’s certainly how Jireček read Kızderbent: their Bulgarian was mixed with Turkish, not because it was finding a new equilibrium, like Mbugu (Cushitic vocabulary, Bantu morphology) or Erromintxela (Romany vocabulary, Basque grammar). It was mixed because it was yielding to Turkish. Of course, if Trakatroukika really was like Michif or Erromintxela, Jireček may not have realised it, and I don’t know if we’d have enough data now to work it out. This was the time of duelling aphorisms on mixed languages (Müller 1875: “There’s no such thing as a mixed language”; Schuchardt 1882/83: “There’s no such thing as a completely unmixed language”), but linguistics was only starting to be sophisticated enough to realise a mixed language when it saw it.

A scenario like Cappadocian though makes sense, and I suspect (based on no real information) that that’s what happened in Kızderbent as well. The village language in SW Cappadocia has a Greek core, but is bilingual: several villages had both Muslim and Christian inhabitants (in one in fact the Muslims spoke Greek), and Cappadocia was also home to the Turkish-speaking Christian Karamanlides. The Greek-speaking villages themselves were shifting towards Turkish: the presence of Muslim Turks and/or Karamanlides made it that much easier. Because Turkish was gaining ground, and everyone was fluent in Turkish, there are more and more elements of Turkish grammar, and not just Turkish vocabulary, in the language the village spoke. In the Balkans, that same bilingualism meant that the grammars of the Balkan languages ended up looking fairly similar, even if their lexica did not. In Cappadocian, it was read—probably correctly—as a precursor to Greek yielding to Turkish correctly. Substitute Greek with Bulgarian, and what you get sounds plausible, at least.

To put the Cappadocian scenario in play in Kızderbent, it would help if we knew that the Turkish-speakers were on intimate terms with the Bulgarian-speakers. The Bulgarian Wikipedia (building on Kănchov? on that 2001 study on Anatolian Bulgarians?) thinks it was Karamanlides, who were Christian and so could intermarry. I think Bithynia is a bit too far West for Karamanlides, but obviously something Turcicised the Trakatroukides’ language.

Outside the main question, of how Trakatroukika actually worked, are the historical questions of what went into it. For starters of course, the question that Jireček and Kănchov worked on in their visits to the region: was Kızderbent’s Slavonic closer to Momim Prohod or Ohrid? (Or for all I know, a mixture of the two—although I presume these settlements were block moves and not mergers of multiple populations.) From the Bulgarian Wikipedia, I see we know a lot about Anatolian Bulgarian in general, and that it was mostly southern Bulgarian; it was possible to pinpoint some villages to particular towns—including Stengelköy, settled from “Kostur in Southwestern Bulgaria”. (The hyperlinked town on the page is Kastoria, now in Greece, and its Slavonic has not been called Bulgarian for a while; but there are a couple of Kosturs in Bulgaria as well, so I’ll assume lazy hyperlinking. 🙂

It’d also be worth checking whether there is actually any Armenian in the language mix, as New Triglia’s site hints. Bithynia is next door to Phrygia, and Phrygian is suspected of being related to Armenian—or a missing link between Armenian and Greek. Phrygia is a long way from the Armenian heartland, but Armenians did travel west during the Ottoman Empire; and the Wikipedia listing of Armenian dialects, from Acharian’s 1909 Classification des dialects arméniens, includes Nicomedia (İzmit) and İznik (Nicaea). So Armenian was spoken in Bithynia at the time (and there’s confirmatin next post too). The creation story of the village reported in the Bulgarian Wikipedia refers to Armenian involvement.

New Triglia also drops the hint of a Greek contribution, both in the four constituents of the village, and in its reference to “Greeks from Ohrid and Monastir”. (If your reaction is, “*what* Greeks from Ohrid and Monastir?”—well, hold that thought.) As mentioned, there is no evidence of mutual intelligibility with Greek, and the Motley Word lists don’t feature a single word that looks more Greek than Bulgarian and Turkish. The Motley Word is a Greek dialect wiktionary, so Hellenic words should have still been distinctive enough to include. (Of course there’s a selection bias, and I’ll come back to that.)

As a language of culture and trade (and the church), Greek had pervasive influence through the Balkans and into Asia Minor, including where Trakatroukika was spoken; and Greek was spoken by other communities in Bithynia:

• Danguitsis, C. 1943. Étude Descriptive du Dialecte de Démirdési (Brousse, Asie Mineure). Paris: G.-P. Maisonneuve.
• Δεληγιάννης, Κ.Π. 2002. Κουβουκλιώτικα: Το γλωσσικό ιδίωμα των Κουβουκλίων Προύσας. Αδελαΐδα.

From the Bulgarian Wikipedia again, we know both Turkish and Greek were encroaching on Anatolian Bulgarian in general, not just in Kızderbent: “Surrounded by many Greek and Turkish villages, the Bulgarians were forced to adapt and to learn the languages of the majority. According to the recollections of the older sons of Anatolians, almost all young and old settlers in Anatolia also used the Greek and Turkish languages, though many of them were illiterate.” But if Bithynian Greek had an effect on Trakatroukika, I suspect it would have been apparent to me by now.

And finally, “*what* Greeks from Ohrid and Monastir?” Well, that’s a messy question, and it’s best left for The Other Place, where I take all discussion of complications in what being Greek (or Bulgarian, or Macedonian) is about. You can probably already predict what I’ll say: the current Trakatroukides of Valtotopi or New Triglia identify as nothing but Greek—so they’re nothing but Greek. The indication is that they share a bloodline and a linguistic heritage with people who have never been Greek. And one must have a very restrictive notion of national identity, to think that a paradox.

Because of course in the 17th century, or whenever Maidenpass was settled by the Bangbangers, there was no Bulgarian or Macedonian or Greek identity as we now know it; there was just the millet-i Rûm of Orthodox Christians, which encompassed Greeks, Albanians, Bulgarians, Arabs and Romanians alike. People knew they spoke different languages (and very different dialects within those languages), but that was not how they defined themselves: that came later.

So yes, it may be wishful thinking to say that settlers from Ohrid in the 17th century were Greek—particularly as Greek that far north was an urban-only language, and Kızderbent is unlikely to have been settled by townsfolk. But if you’re going to deem them Bulgarian or Macedonian, and not just Slavonic-speaking, you’ll need a very big asterisk; and to their descendants, such an identification is irrelevant. (If you make it alive out of the argument over whether they were Bulgarian or Macedonian—which goes to show how immaterial the question was back then.)

This post has been put off for a long time, and its sequel even longer; but one of its protagonists has just asked for it here, so I will not put it off further. This is a translation of a derailed comment thread last October, over at the Magnificent Nikos Sarantakos’ Blog, through which I discovered the existence of the Trakatroukides of Kızderbent (Kizdervent, Κίζδερβεντ, Къздервент). Kızderbent is a village in Turkey which spoke “Trakatroukika”; Trakatroukika is reportedly a heavily Turkicised Bulgarian (possibly with Greek and Armenian elements). The villagers of Kızderbent considered themselves Greek, and their descendants now live in Greece.

There will be two or three followup posts with more detail that I can find about the village and its language, but I wanted to get this out to tide me over. I’d like to thank here the two protagonists of the thread:

• In the right corner, Cornelius (Κορνήλιος), the polytonicist linguistic reactionary gadfly of the blog—and as it turns out, the grandson of a Trakatroukika speaker. It’s worth noting, for those interested in Greek diglossia, that even if Cornelius is a linguistic reactionary, he can’t quite use Puristic for the Cause in 2010—which underlines how thoroughly Puristic has lost ground by now. (That he is under forty doesn’t help either.)
• In the left corner, A Butcher of Yore (Ένας Χασάπης Από Τα Παλιά). He’s a Butcher Of Yore because, like many other regulars at Sarantakos’, he used to be on the Hellas mailing list in the 90s, where his moniker was Abdullah the Butcher. Butcher of Yore has deep reserves of scholarship on Greek refugees, and I want to thank him for forcing me to look at the Trakatroukides more thoroughly than I intended to.

I’m going to translate the entire exchange, because it really did come out of nowhere, it went in several interesting directions (including whether it’s worth preserving languages), and it has some great pointers. Like I said, I’ll be supplementing this with more material in future posts (including links to a dialect lexicon, and to a conversation snippet on YouTube).

I’ve inserted some annotation in square brackets; I am editing out some of the ribbing and the language politics, because that would tread on sensitivities further than is necessary (including taking some linguists’ names in vain). But the thread is openly accessible (in Greek).

• #20 Cornelius: My grandmother calls jars κιούπι, and γκιούμι for metal vessels. Once a gang of antiquities smugglers was arrested, and I was watching it on the television with my grandmother. They were showing some amphorae and my grandmother said, “ah, we got plenty of them κιούπιa.”
• #21 Voulagx: Cornelius, do you know where γκιούμι comes from? I mean is it a Turkish word, Slavic, or something else?
• #22 Cornelius: I don’t know. My grandmother in her childhood spoke a language with Turkish, Slavic, Bulgarian, Greek words etc in it—and she was of Asia Minor descent, at that. They’d say mátska for cat and kokóska for hen and kútse for dog and ódzak for fire and sófa for couch. Quite unacceptable. How could I not become an admirer of the learnèd tongue after that? Fortunately my grandmother has almost forgotten that language. Her mother spoke it; her father spoke Greek, and was at war for 10 months and 10 years. Of such a generation do I long to be. My grandmother also uses the word ílem: I don’t know what it means, but I know how to use it instinctively.
  [More on Cornelius teaching himself polytonic at 11.]
  Also from a fairy tale my grandmother used to tell, I remember this song which was supposed to be spoken by a snail on a bridge while a cart went past. I don’t know what language it is or even if I’m breaking the words apart correctly:
  tákar tákar ke priataláh duí sulísperi zebé khantúm
• #22 Butcher Of Yore: Don’t say it so loud, Cornelius; no matter where you think your grandmother’s descent is from, Tremopoulos will deem you an Oppressed Macedonian, and then you’re in trouble! matska and kutse, he says—why Gruevski will be looking for you as a Grkoman example to avoid! 🙂
  In all seriousness, if your grandmother’s mother was truly from Asia Minor, the only thing I can think of is that she was one of the Trakatroukides of Kızderbent. But there were very few of them.
  Otherwise, should I presume her mother was a local Slavophone, and her father was from Asia Minor?
• #24 (Me thanking Butcher for the information, and threatening Cornelius with posting about Trakatroukika)
• #25 Butcher Of Yore: Well someone should work on the Trakatroukides before their language dies out (as with the speakers of Cappadocian, Propontis Tsakonian, etc.) Beyond the aforementioned municipality in Chalcidica, you can also head towards Polykastro in Kilkis and find them there.
  They’re organised: [Facebook group], and there’s a thesis on them. [NN: More links in following posts]
• #27, #29, #44 (Me burbling further on what I found on Wikipedia, including the claim that they were Bulgarians) [More on that in following posts too]
• #31 Nikos Sarantakos: (γκιούμι ~ γκιουγκιούμι < Turkish güğüm < Greek κουκούμι(ον) < Latin cucuma) So what is the word? Greek, Latin, Turkish? It turns out to be common property.
• #33 Voulagx: Sarant, it was an honest question, because we also call it γκιούμι in Aromanian (giumi-a, where -a is the feminine article). I like “common property”, but you’re making me wonder: what if the partners start arguing about their percentage of ownership? Thank you.
• #34 SophiaOik: Cornelius, I suspect you’re trying to trick everyone with your grandmother; it’s not as if those parts had so large a population, or were fertile enough, for us to bump into their descendants as easily as that.
• #40 Cornelius: I’m not tricking anyone, I’m telling the truth, and she probably was a Trakatroukis, because I have heard it said, that they spoke “trakatroukika”.
• #44 #45 #46 (Me and Cornelius about me posting about Trakatroukika)
• #48 Cornelius: I used to note whatever my grandmother remembered of that language in a notepad, but where might it be? I’ll ask my grandmother when I go to the village on a weekend. They had two versions of “fish”, balúk and ríba. My mother says that my great-grandmother never learned Greek; so when she would go as a child to my great-grandmother’s village, her grandmother, she must have learned some words; but she doesn’t remember them. But I found this [a lexicon at the Motley Word site], and I think I remember ókhtse, I must have noted it in the notepad.
• #49 Cornelius: In fact they came from the region around Nicomedia to Greece. [Kızderbent is near İznik, Greek Nicomedia.] First they took them to Komotini. [NN: Some Trakatroukides still live in Roditis, 5 km from Komotini.] But my great-grandfather did not want to see mosques, they were sick of them from Asia Minor, and they moved on. It’s all so fitting for me now!
• #50 E-fufutos: You’re carrying such an imperial treasure with you, and instead of preserving it, you exorcise it with caricatures of the Ancients? Haven’t you heard stories of aunts (like mine or “the Aunt From Chicago”), who thoughtlessly threw out their grandmothers’ furniture, carved from massif wood, to buy fake European stuff out of plywood, only to regret it now?
  I hunt after things like that, maybe because I feel the lack more intensely. My folk have their peculiar origins hidden more deeply.
• #51 Maria: Cornelius, if your great-grandmother didn’t speak Greek, then your grandmother must have spoken Trakatroukika. She can’t have needed an interpreter to communicate with her.
• #52 Cornelius: Anyone can consider whatever they want to be a treasure. I don’t consider the leftovers of Ottoman Rule and Bulgarians a treasure. What I consider a treasure are Doric lexical remnants still surviving in the mouths of illiterate villagers. As for old furniture, they are my weakness after old books.
• #52 SophiaOik: I wouldn’t say they’re an imperial treasure, because their vocabulary would be limited to everyday objects with few abstract notions. Such a mixture might have linguistic interest, but such halfway languages seldom survive long, because they have few speakers and are not spoken outside the group that creates them. If Cornelius works on linguistics, he might write a couple of papers on it, maybe even a PhD. If not, he need not feel guilty.
• #54 Nick Nicholas: Let’s lay off people: all treasures end up faded and ignored. If someone doesn’t want to be a hostage to his heritage, well, we have to respect that. I do it too, and all of us who are modernised do. If he prefers grave accents [referring to Cornelius and the polytonic], good for him. I prefer English, after all. [By which I meant, I don’t feel an atavistic obligation to use the language of my forefathers in my everyday dealings.]
  I remember a talk on remedies and folk botany in Indonesia in my linguistics department. All us westerners were saying how cute and picturesque it was. But the Indonesian student was irritated: he did not come to the West from Aceh to keep hearing about primitive nonsense. And of course he was under no obligation to stay “picturesque”.
  But Cornelius kept a notepad and submitted entries to the Motley Word. [I was wrong about the latter.] I’m hardly going to tell him off; quite the opposite.
• #55 Butcher Of Yore: Well, I’ve impressed myself with my correct guess on where Cornelius’ great-grandmother was from. See what obsessing with the Macedonian Issue does to a Southern Greek? Admittedly the Trakatroukides had particularly impressed me when I first found out about them. I’m curious to hear if they really did also have Armenian linguistic antecedents.
  Of course the Bulgarians consider them all their own, Nick 🙂 —they consider the Slavic-speakers of the Albanian Prespa lakes Bulgarians too. (And some though not all of them do call themselves Bulgarians, as their great-grandfathers did.) The same goes for the (Muslim) Gorani of Kosovo. The Bulgarians would even issue them with passports, at least until recently.
  But if you’ve read about Vasil Kanchov [who had researched the Bulgarians of Asia Minor], you immediately realise that these populations had been placed into the game of Balkan nationalisms, since Basil was the ethnographer par excellence of Bulgarian nationalism. (And his work was quite detailed—Lithoxoou treats him as a saint.) So politically Bulgaria clearly tried to stamp them as her own, to use them as a minority, either as leverage and a pressure point, or as a population to exchange (as largely happened). Somewhat like how Rumania used the Rumanian-identifying Vlachs (since there was of course no possibility of advocating for a Rumania reaching the Aegean). [See “Awakening of the Aromanian identity and Romanian sponsorship “]
  What’s interesting to research further is whether the enmity between the Patriarchate and the Exarchate [which marked the ecclesiastical starting point of Bulgarian nationalism] had spread further from Macedonia, Thrace and Constantinople to Bithynia [the region of Kızderbent], and how extreme or even violent it ended up as in Asia Minor. It’s not impossible that those who left for Bulgaria were already pro-Exarchate, and those who stayed behind and were expelled [to Greece] with the population exchanges were pro-Patriarchate—meaning, the national/ecclesiastic split in the Ottoman Balkans had also passed to Asia Minor.
  Trakatroukika have unfortunately gone down the road of all the remaining dialects and languages of the Greek people, though given the small number of its population and its scattering in Greece, it is closer to dying out than the other dialects and languages. It would be good for it to be studied and recorded; I don’t consider it practical to revive it at this late stage, nor do I think it desirable, if it will only be promoted by certain activists, and the majority of the descendants of the Trakatroukides are indifferent to it. But we owe it to the tradition of our people not to let the recollections of local idioms and their treasures be lost entirely—the more so when they are particularly unusual.
  […]
  Cornelius, seriously, gather whatever you can preserve. From my grandfather’s Andriote Arvanitika, I can only vaguely remember some curse words (because that’s the only time he’d use it), and I am unhappy about it.
• #56 Cornelius: I am not denying my heritage, nor am I acting modernised. And don’t think I want those words and their meaning to be lost permanently from people’s memory. When I was noting them in my notepad I thought with some degree of melancholy that one day there would be noone alive who had heard those words used. But that’s one thing, and grave accents are another. Trakatroukika can’t be an official language or a language of literature.
• #59 Nick Nicholas: True, the idiosyncratic language mixture of one village can’t be an official language. (Of course, Puristic was an idiosyncratic mixture too, but never mind.) But a formal language becomes such through extra-linguistic circumstances. I mean, if we could accept 150 years of Puristic, under different circumstances we could have ended up with an official Trakatroukika.
  So long as we weren’t obsessed with grave accents. [i.e. prioritising continuity with Ancient Greece.] But that’s an extralinguistic factor too.
• #60 Languagelust: balúk (balık) and ríba (риба)! The first Turkish, the second Slavic. You’re giving us these one by one, Cornelius!
• #76 E-Fufutos: I don’t believe that languages, dialects and idioms that have started heading towards oblivion should be preserved in formaldehyde at all costs, either, just so we don’t get called nationalists, or for reasons of folklore. If the speakers themselves are reluctant and it doesn’t sit with them, we should let things develop on their own. You have to like something to want to retain it.
  …
  I also have relatives with some Arvanite roots, and nothing has survived, not even a recollection from a grandmother or a great-grandfather who used to speak Arvanitika proverbs; and I consider that natural.
• #77 Angelos: We should not keep it alive with a respirator and an artificial kidney. But we should preserve it in formaldehyde, like we would piously preserve a mammoth if we found it close to death in Siberia, dying in our hands. Not “just so we don’t get called nationalists, or for reasons of folklore”. But because every manifestation of the human phenomenon par excellence we call language is interesting and worth recording.
  And if Trakatroukika are indeed a “mixture” of Bulgarian, Turkish, Greek and Armenian (and not just a South Slavic dialect which has absorbed a lot of Turkish, Greek and Armenian words), it would be a very rare if not unique linguistic phenomenon. That’s one more reason to record and study it to the extent it is still spoken (if indeed it is at all), even if it’s just ten old ladies.
• #78 Cornelius: I asked my grandmother today and she told me that ókhtse was the word for a big sheep, and a small sheep was called kuzía.