Notes from Cyprus

By: | Post date: 2023-07-08 | Comments: No Comments
Posted in categories: Linguistics, Modern Greek

Just spent a week with my relatives in Cyprus.

I’m not going to make this a course on Cypriot, and I’m not going to explain the technical terminology thoroughly; at least not yet. I’m tired, and I just want to capture the things that struck me about Cypriot after a week of hearing it all around me, even if in a potentially attenuated-for-“Penpushers” form.

(“Penpushers”, καλαμαράες, being speakers of Standard Greek.)

Phonology: my family were certainly doing most of the familiar Cypriot processes: double consonants, dissimilation of fricative + fricative (e.g. vɣ > fk, fx > fk), palatoalveolar allophony. They did not seem to be doing a whole lot of fricative + yod dissimilation (e.g. ðj > θc). x > θ I caught only a couple of times. As I’d been warned by Tsimplakou, lots of dropping of intervocalic ɣ, intermittent for ð (though ɣajðurin > ɣaurin “donkey” and koruðes > korues “girls” was regular), none for v.

Something I didn’t know about beforehand: ʎ > j. So I heard /palja/ [paʎa] rendered as [paja] several times.

Accentuation: the three-syllable limit on stress in phonological words is violated, as it is in Pontic: ˈipen-tu-to. The ascending accent of enclitics, which Standard Greek uses to comply with the limit (o ˈanθropos mu > oˈanθroˈpozmu), is alien to Cypriot, so alien that primary school children, per Tsimplakou, have to learn it sing-song—and come away with the impression that Standard Greek is a sing-song kind of language.

Vocabulary: mostly intelligible, though lots of lurking false friends: pefto is “lie down” as well as “fall down”, vareto is “heavy” instead of “boring”. Occasional stumbling blocks which had to be expslained to me, but they were clearly watching their lexis around me. apoloɣume > apoloume as the English calque “apologise” rather than “speak in my defence” is common, and indeed is used instead of mere “I’m sorry”.

Morphology: as pointed out to me by Mertyris, avoidance of the – contract conjugation in favour of – (e.g. efxaristas), though that is a general trend in Greek dialect, and is characteristic of the vernacular away from learnèd influence. When they do use –, it’s not the Standard Greek –jeme, but the compromise with archaic Greek –jume. So “I’m bored”, an early Cypriot teen website, was : bareomai > varjume > varkume by yod dissimilation.

Syntax: the cleft is so omnipresent, I wasn’t even noticing it. As Tsimplakou pointed out, Cypriots can learn that Standard Greek uses focus dislocation for emphasis instead; what they do not pick up is that Standard Greek does not use clefts at all—it is hard to learn from negative experience.

Deixis: discourse deixis in Standard Greek is done by aftos, the unmarked 3rd person pronoun. In Cypriot, it is done by the overt demonstratives, tutos “this one” and dʒinos “that one”. So the unemphatic Standard causative ɣj afto “for that” ends up as the enthusiastic cleft ˈen pu tuto pu… “it is from this [clause] that…”

Intonation: falling, and particularly for men, choppy. That gets in the way of making sense of the vocative particle re. It is a fine line in Standard Greek whether it is to be taken as insulting or pleading, and the pleading interpretation is based on non-final intonation (or uptalk, when the re is added at the end of a sentence). Choppy falling intonation means a lot of very peremptory sounding re, and as a Penpusher, you do need to consciously remind themselves that they are not being rude at all, however it may come across.


READ: Thavoris, Means of expressing the time of day in Ancient, Mediaeval and Modern Greek

By: | Post date: 2021-11-08 | Comments: No Comments
Posted in categories: Linguistics, Modern Greek

Θαβωρής, Α.Ι. 1959. Τα εκφραστικά μέσα προσδιορισμού του χρόνου του ημερονυκτίου στην αρχαία μεσαιωνική και νέα ελληνική. (Επιστημονική Επετηρίς Φιλοσοφικής Σχολής, Παράρτημα αρ. 2.) PhD Dissertation, University of Thessalonica. Thessalonica: University of Thessalonica.

Antonis Thavoris was Linguistics Professor at the University of Ioannina. His 1959 PhD thesis is on expressions for the time of day throughout Greek; I’m going to mention bits of it from Modern Greek that strike me as being of interest.

  • p. 19. In dialect, if you want to say that someone goes to bed early, you say that he goes to bed at “7, 9, or 10 at midday”—where the hours given are merely conventional.
  • p. 27. Similarly, “7, 9, 12 at midnight” for “very early in the morning”
  • p. 28. The twelve hours of daytime were set by the Babylonians, and known to Herodotus; the extension to a 24-hour day was done by Hipparchus in the 2nd century BC.
  • p. 31. In the Koine, hours were given with ordinals, and started counting from sun-up: “the ninth hour” was 3 pm. (That is the language used in the New Testament, and has been carried over in traditional translations.)
  • p. 32. That ordinal naming continued throughout the Middle Ages; “from the first hour” occurs in the Chronicle of Morea (so 14th century.)
  • p. 32–33: The West started counting hours not at sun-up, but at midnight. So in the early modern period, Greeks started having the option of telling time Western-style or Eastern-style. Since the Ottomans continued Roman and Byzantine practice, Eastern-style was called by Greeks, in Italian, alla turca, and Western-style alla franca. Thavoris gives an example as late as 22 December 1912, from the Athens newspaper Akropolis, when the Greek king visited Kozani, newly added to the Greek state: “at two in the afternoon [Western time, which is what the Athens press used], the herald shouted from the base of the enormous bell tower of St Nicholas: ‘at 11 o’clock alla turca our King is arriving, five o’clock alla franca.’”
  • In the monasteries of Mt Athos, Eastern time is still followed.
  • p. 32. The Modern Greek way of telling time does not use an ordinal: it uses a number + i ora “the time”, e.g. efta i ora “seven the time = seven o’clock.” But the older vernacular pattern (which Thavoris does not date, but he does cite from dialect sources) looks like English military time: “ten hours at night” (stsi deka ores tis nixtos), “hours 11 of the day” (is tas oras 11 tis imeras), “four hours of night” (os tis teseres ores tis niktos), “one hour, the seventh hour according to the older generation” (mia ora, i evðomi ora par arxeoteris).

There’s a deeper story there: ordinals were on the way out of the vernacular, which either made them more transparent (efta “seven” > eftatos instead of evðomos “seventh”), or outright dropped them: contemporary ordinals like tesarakostos ðefteros “42nd” are completely revivals in Puristic. As those old ways of telling time show, Greek wasn’t comfortable with saying “seventh hour”, like Koine and Mediaeval Greek did, and it wasn’t comfortable with saying “hour number seven”, with an abstract number, as the contemporary colloquial language might (ora numero efta): it could only use 7 as a cardinal number, counting hours in the plural.

In fact, Triantafyllidis recorded that the early Greek migrants to New York found themselves in the same discomfort with ordinals, and resolved it in the same, now obsolete fashion. The did not refer to 42nd Street as tesarakosti ðeftera oðos, with a Puristic ordinal, nor as oðos saranta ðio “Street 42”, as the contemporary vernacular might have attempted: they used the expression stus saranta ðio ðromus “at the forty-two roads”.

Evangelos Lolos, remember how we were wondering how the old vernacular would have said something like sistima ɣuintouz “Windows System”, with its current French word order (Système Windows)? The old Greek hours don’t give us the answer, but it does show that that French ordering, which you might see in ora efta “hour seven” or ora miðen“hour zero”, is not how the vernacular used to do it.

I’ll add that the current way of adding minutes to the hour, efta ke ikosi “seven plus twenty = twenty past seven”, efta para ikosi “seven minus twenty = twenty to seven”, is not that old—although Thavoris does not dig further into it. When I was researching how to render contemporary notions into Ancient Greek for a picture dictionary I collaborated on (Textkit Greek and Latin Forums), I found an 1832 naval manual trying to revive the old Koine ordinals, by using “15 minutes of the 10th hour, 45 minutes of the 7th hour, 20 minutes of the 9th hour” (πεντεκαίδεκα λεπτὰ τῆς δεκάτης ὥρας, πέντε καὶ τεσσαράκοντα λεπτὰ τῆς ἑβδόμης ὥρας, εἴκοσι λεπτὰ τῆς ἐνάτης ὥρας).ογδόη+ώρα+και+λεπτά&source=bl&ots=Owe1FwRp8V&sig=49N5Xc-7JyOEJUCIFoeUFxn9qjc&hl=el&sa=X&ved=2ahUKEwjrhIf-l8PfAhUO1hoKHc3DCIEQ6AEwB3oECAQQAQ#v=onepage&q=ογδόη%20ώρα%20και%20λεπτά&f=false

  • p. 39. In one of the more colourful expressions of Modern Greek, sunset is referred to as o iljos vasilevi, “the sun reigns”, presumably because the sun then looks at its most magnificent. But in Oinoe (Ünye) in the Pontus, “the sun reigns” is used in the more predictable meaning of “it is midday, the sun is at the zenith”.
  • p. 46–50. The rising and setting of constellations and stars was used to tell time at night: the Pleiades (pulja), Orion (aletropoði “the Ploughshare”), the Morning Star (avɣerinos) and Evening Star (aposperitis), Andromeda (stavros “the Cross”), the tail of Scorpio (galonomja, “the milking sheep shepherd’s stars”).
  • p. 52. The Ancients told the time by how long their shadow was—Aristophanes refers to seven-foot shadows and ten-foot shadows; p. 53 shepherds still did so in Crete in 1918: if their shadow was eight feet long, it was time to take the milking sheep in for milking.
  • p. 56–58. The Homeric amphilykē “twilight” survives in the Koine lykophōs “twilight”, which was revived in learnèd Modern Greek. Thavoris doesn’t go deep into the etymology, and apparently there is still controversy around it, but it seems to me lyk– is the same word as Latin lux “light”, which also turns up in the epithet of Apollo as lykeios. So amphi-lykē “both-light”, “either side light” could refer to amb-iguous light; the etymology of lyk– forgotten, lyk– was reinterpreted as meaning just “twilight”, and re-elaborated as lyko-phōs “twilight light”. The word is first attested in Aelian, and Aelian already thinks the lyko– refers instead to wolves (lykos), because twilight was when wolves came out; Eustathius of Thessalonica, a millennium later, reports that “vulgar erudition” instead distorts it to ɣlikofos, “sweet light”.
  • p. 99. Times of eating have been used consistently to name times of the day; it’s useful to be reminded not just that Modern apoɣe(v)ma “afternoon” is literally “from-meal”, “after-meal”, but that ɣjoma, an old term for “noon”, is just a vernacular form of ɣe(v)ma “meal”.
  • p. 103. The ancients routinely referred to bedtime as the “first sleep”, peri prōton hypnon; the expression survived into Mediaeval Greek as proθypnion, and in Modern dialect as proto(i)pnin. Thavoris does not seem to make the explicit connection, but as I’ve seen mentioned once by someone on Quora, people who in olden days got to bed at sundown used to wake up around midnight, and then get back to sleep for a second sleep; apoipni“from-sleep” in Symi is glossed as “second, morning sleep”.
  • p. 107. The Roman notion of watches of the night (=guard duties), lasting three hours each, survived into Byzantium; they persisted into the pre-WWII navy, and I’ve reported here that the 2–4 am watch in the contemporary Greek army is called “German” (ɣermaniko), because you have to be a Schwarzenegger to endure it.

ATTENDED: Workshop on The Greek language after Antiquity

By: | Post date: 2021-11-08 | Comments: No Comments
Posted in categories: Linguistics, Modern Greek

Although I did my best to attend the entirety of this workshop, I was pretty exhausted—the humidity on Saturday did not help me stay alert till 5 AM. So rather than drill down into talks (including talks I wish I’d stayed awake for, like Io Manolessou’s), I’m going to limit myself to headlines.


  • Volume 22 of Kriaras’ Dictionary of Early Modern Greek (1100–1669, vernacular) came out in June: it now goes up to τέως. There’s another three volumes left to go, which makes it another decade.
  • The third volume of the abridged version of Kriaras’ dictionary came out in June 2020; the abridgement goes up to volume 20 (σταματώ). The abridgement is in places a reworking of the earlier volumes’ materials, and does need to be looked up independently.
  • The first two volumes of the abridgement (up to παραθήκη, first 14 vols of the full dictionary) were already online, at Επιτομή Λεξικού Κριαρά . The 19 first volumes of the dictionary, up to σι-, have just been put online at Αρχικη . (The same site hosts Katos’ dictionary of Greek slang, Κάτου Γ. )
  • The 22nd volume was dedicated to the warriors of the Greek War of Independence, published as it was on the 200th anniversary of the war. The presentation was dedicated to the memory of Yannis Kazazis, who had assumed directorship of the dictionary after Kriaras, and who has just passed on. I will always hold a candle in the memory of Tasos Karanastasis, who ran the dictionary for two decades, and got little acknowledgement for it (including by Kazazis): RIP: Tassos Karanastassis . But the dictionary is not the property of any one scholar, not Emmanuel Kriaras’, not Kazazis’, not Karanastasis’: its the labour of all those who work on it, including the impossibly enthusiastic Maria Kesoglou, who is now running it.
  • And more than those four K’s of Early Modern Greek lexicography, it’s the patrimony of all of us.


  • Trapp’s Dictionary of Byzantine Greek (800–1200 plus whatever dictionaries have neglected to pick up either side, learnèd) completed work in 2017, and is now available online in its entirety via the TLG (provided you have an account there, which I do not.)
  • The Supplement to Trapp’s Dictionary is now happening under Elisabeth Schiffer: it will be online-only, and it will be published author-by-author, rather than in alphabetic order (although alpha, iota and rho are ready to do).
  • The Supplement will start going online in mid 2022. The full dictionary had 187k entries; some 30k entries are going to be in the Supplement, of which around 24k are to be updates of existing entries, and 6k brand new entries.

Historical Dictionary (“Dialect Dictionary”)

  • My good friend Io Manolessou, who is the linguist I hoped to be when I grew up and then some, is now Acting Director of the Academy of Athens’ Historical Dictionary of Modern Greek, Commonly Spoken and Dialectal (1800+, dialectal and standard Modern Greek). My heartfelt congratulations to her, and I feel all the more guilty that I ended up falling asleep during her presentation on Greek historical dialectology (it was getting to 4 AM here.)
  • After decades of being stuck, the dictionary is now publishing again; the latest volume published, March 2021, is vol. 7 fasc. 1, going up to δόγης.
  • The Academy’s digital repository includes the first five volumes of the dictionary (1933–1989), up to δαχτυλωτός; ΑΑ portal .
  • Historically, there was a divide between academic linguists and the Academy’s linguists. It is only a good thing that that divide is now gone.

Thesaurus Linguae Graecae

  • The TLG is continuing to expand into Early Modern Greek, and will encompass both literary and subliterary texts as it does so.
  • The lemmatiser of the TLG was maintained for a long time (2003–2017) by Nick Nicholas, whose contribution was acknowledged.
  • Since 2017, the director of the project Maria Pantelia has assumed management of the lemmatiser.
  • The lemmatiser’s morphological indexes are being extended on an ongoing basis to deal with Early Modern Greek.
  • Coverage has also been expanded by adding lemmata not currently included in dictionaries (which is a policy change from when Nicholas was maintaining the lemmatiser.) The TLG has provided athesaurista to lexicographic projects in the past, and these athesaurista are available to lexicographers on request.
  • The TLG is eager to initiate collaboration with linguists and lexicographers as experts in Early Modern Greek.

ATTENDED: Workshop on Modern Greek Koineisation

By: | Post date: 2021-11-08 | Comments: 2 Comments
Posted in categories: Linguistics, Modern Greek

So, over the past weekend, I’ve attended two linguistic workshops in Greece. (I’ve already complained about the scheduling conflict on Nick Nicholas’s answer to Why in Australia do people of Greek & Italian origin just say they are Australian, but in the US Greek/Italian American? Being closer to Europe, do they still wish to keep the old world ties in tact? How does that work also with Chinese Australians? .) The workshops were in the evening into the wee hours for me in far distant Australia, and my attention did fade as I went. But the ability to dial in to workshops and conferences, rather than have to get on a plane and take off two weeks and $5k, has been a boon of COVID to your antipodean correspondent, and he appreciates that it is now possible.

Even if that means he doesn’t get to go out to a taverna for chops with the participants afterwards.

Herewith, reactions to the bits of the koineisation workshop I attended and can remember. Because of the scheduling conflict, I wasn’t able to attend as much of the koineisation workshop as I’d have preferred; I only caught the first morning session.


As background: Contemporary Modern Greek as we know it emerged in Athens as the capital of the Modern Greek state; it owes only a couple of words to the native dialect of Athens, which was extinct by the 20th century, and a lot to the dialects that converged in the capital, settled from elsewhere in the Greek-speaking world. That makes it a koine, but a koine with a poorly understood history: we don’t have a lot of written records of how the dialects converged in Athens in the 1840s and 1850s.

We also don’t have a lot of records of Peloponnesian, the dialect group widely held to be the basis of that koine: dialectologists have not bothered to record it historically, precisely because they assumed it was identical to the Modern koine. (And as Nikolaos Pantelidis has been saying for decades, they were wrong.)

We do know that Puristic Greek did contribute to the formation of that koine, as a template to fall back on when dialects disagreed (or even when they didn’t): that’s something Brian Newton worked out in the 70s, in explaining why Standard Greek has gone with the archaic –ome instead of –ume as a verb inflection. (Yes, that is also the inflection used in the Ionian islands; but no, the Ionian islands really did not have as much influence on Standard Greek as literature scholars like to think.)

And because the Greek Language Question was and remains so ideologically loaded, pointing out that Puristic contributed to the Modern koine is still contentious. I have anecdotally heard of conference papers being rejected for suggesting so.

So, what papers do I remember?


I’m going to start with a paper from the time-clashing workshop, on post-antiquity Greece, by Pantelidis himself. Thematically, I think it’s a better fit with this group of papers, and I have already mentioned it.

Pantelidis’ current interest is the survival of the Ancient Greek pronunciation of /y/ as [ʏ, ʉ] in Modern Greek dialect (as opposed to Standard /i/, with a few dialects doing [ju])—notably in the dialect group of Old Athenian, which also includes nearby Aegina and Megara. There are lots of quite clear statements in the 19th century that people in Athens were pronouncing it as <ü>, but they weren’t being made by Greek speakers (who could not hear the sound, because they were Greek speakers); those statements were being made by Germans and French, and they were either ignored by subsequent scholarship, or disbelieved.

Pantelidis concedes that in most places, the old pronunciation has vanished with nobody remembering it was ever otherwise. But he has been digging up more recent recordings of Aegina and Megara, and he’s got the instrumental phonetics to prove that people are still producing [ʏ, ʉ] in 2016—a millennium after a poem in 1030 mocked a priest as coming from a village “where people’s intellect is not better than oxen”, because they were using the new-fangled [i] pronunciation. And he’s been the first to notice it.

Roger Lass once wrote a paper called “When is a language change”, noting that Southern English started dropping its r’s in the 17th century, but a few people were still pronouncing them in London as late as 1870. It is convenient to think of language change as a brief switch-over from form A to form B, especially when looking back to a change that happened centuries past; but that is a convenience. Language change takes a very long time to settle.

I’ll also note that there’s another instance of the cloth ears of Greek dialectologists, that Pantelidis bemoans, in the eccentric dialect Samothrace. That dialect happens to have high central vowels. No ethnic Greek dialectologist has ever noticed them. August Heisenberg (Werner’s dad) didn’t notice them either, when he recorded the dialect of a Samothracian POW. And that’s because high central vowels are not a feature of either Standard Greek, or Standard German. If you’ve got a poor ear, you’d have to be a native speaker of a language like Romanian to notice them.

… Or Aromanian. Which is why the Vlach linguist Nikos Katsanis was the first to record them in Samothracian, in the 1980s.

Pantelidis is indignant about the poor job Greek dialectologists past have done with the data of the Greek language, and concluded by saying that “I’m afraid we’re going to have to redo everything from scratch”. I find it terrifying to hear him say that, because he may well be right.


Eleni Karantzola’s paper was an overview of the five grammars of Modern Greek written in the 16th and 17th century, the time period she specialises in and keeps doing wonderful work on. I’m proud to say that I was already familiar with those grammars (I wanted all the data I could get; unfortunately I was working on syntax and not morphology, so those grammars had nothing much to tell me.) Her conclusion was that those grammars were trying to set up a standard form of the language; they do point out dialectal variation (Romanus’ more than others, but I remember Portius’ doing so too), and they do pick forms as the preferred forms.

She also found that the literary works of the time completely ignore those grammars’ advice, and that’s to be expected: unlike France and Italy and the other countries who were standardising their languages at the same time period, those grammars did not have the authority of the State or of organised writers behind them.

Christos Karvounis’ paper on the formation of Modern written standards in the 18th and 19th century, I am told, was eagerly attended, but Karvounis was a no-show, so can’t speak to that.


Just as Karantzola is working on the 16th-17th century, Peter Mackridge is working on the Phanariot texts of the 18th century. The 18th century is the last frontier for Greek historical linguistics, and has had very little attention paid to it to date; even the texts we have were mostly published in the last two or three decades—and there are plenty of texts still unpublished. That lack of attention has mostly been ideologically motivated: the Phanariots, as loyal Ottomans, were regarded as traitors in the new Greek state, and 19th century language activists dissociated their initiatives from them.

The Phanariot texts are grammatically consistent in their version of the vernacular. (At least, the vernacular of their authorial voice and their Constantinopolitan characters. To the Phanariots, all their islander servants sounded the same, which is why the low-class characters in their literature mix up the dialects of the Ionian islands with those of Crete and the Aegean.) And whatever koine had formed in Constantinople must have contributed to what happened later on in Athens, even if it didn’t contribute much. (The imperfect –usa inflection is the only element of the Modern Standard we can be sure is Constantinopolitan, although Triantafyllidis did notice that the Northern Greek accusative indirect objects used to be fashionable in Athens; now, they are what Athenians mock Salonicans for.)

But much like the other literary “koines” that preceded it—that of Early Modern Greek literature, of Cretan literature, and in a sense of folk song—the Phanariot literary “koine” seems to have been a passive thing, rather than an active thing. They dismissed islander dialect as peasant-talk, but they didn’t have a rigorous norm of what mainland dialect features were in or out.

Rather, they had a vague and varying sense of what dialect features particular to their own region were not used more widely, and sometimes they avoided them, sometimes they didn’t bother. They de-Thracianised their language, but they didn’t de-Thracianise it assiduously. Mackridge has confirmed that they avoid the characteristically Constantinopolitan broad use of pu as a complementiser, which Cavafy and (in the first edition of My Voyage) Psichari did not manage to—although that could just be because it was a later innovation. But some authors use the Thracian subjunctive ðjo “see”, while some avoid it and use the more common form ðo.

Mackridge finds literary Phanariot charming, as an evolutionary dead end of koineisation—what might have been Standard Modern Greek, if history took a different path. People have long said the same about literary Cretan—and, because of Language Question politics, more people have felt they had license to.


Christos Tzitzilis’ talk…

… ok, this would normally be a career-limiting move, but this isn’t my career to begin with.

Tzitzilis is coming from a place of militant demoticism, which has always been Salonica’s preserve. But to refight battles that were settled in 1910 is just embarrassing. And the prescriptivism is no more welcome in linguistics from someone who thinks I should be talking like an unsullied 1900s peasant, than it is from someone who thinks I should be talking like St Paul.

The ire of Tzitzilis was drawn in this instance from the current confusion around the 3rd person aorist passive. Puristic Greek is officially dead as of 1976; unofficially, Puristic morphology is still around, and was reasserted in the public sphere from the 90s on. As a result, there’s a vernacular standard inflection (-θike), there’s an archaic variant which is now showing up with high register verbs in 3sg (-θi), and there’s a wide range of hybrid forms, as speakers are confused as to how archaic their learnèd aorists should be: should they also be doing an internal augment? A second aorist? Should they be levelling the stems in the vernacular direction instead?

You can look at that as a language community member, or indeed a language activist, and find it laughable. I’ve done it with the bizarre revival of feminine –us genitives (Σαπφώ, Σαπφούς). You can think that the country has lost its marbles by reviving those forms. You can agitate for them, and you can agitate against them.

But you’re not doing linguistics when you do that. You’re doing something different.

And to turn around and say that the Greek people is dishonouring its history, by not following the vernacular idealised norm, is as silly it is to say that it is dishonouring its history, by not speaking like St Paul. To go on to say that fully productive inflection paradigms are a fundamental entitlement of speakers (!) — that somehow the 3sg-only archaisms are a violation of human rights—is several steps beyond silly. And to defend it with a teleological argument is just embarrassing. Sure, the Standard vernacular inflection resolved some problems that the earlier inflection had in a Modern context.

That doesn’t mean it was inevitable and optimal: what language change is? The island dialects have been slow to adopt the innovative inflection: what does that make them, brain-damaged? Less mentally agile to see a problem than their mainland cousins? No. It just means teleology is a poor way to explain language change. None of it is inevitable, or just about problem-solving.

Greek has always been subject to learnèd influence. Greek has always had people hesitate on how to inflect learnèd forms. Greece is neither the first nor the last language to have been impacted by diglossia, or indeed by clashing registers. Yes, it’s messy if you’re a neogrammarian. Yes, it violates the simple rules that the mainland dialects followed in 1800. And that just means you’re going to have to rewrite your rules, because that’s how people are now speaking. The Common Greek Language is what Greeks commonly speak. Standard Modern Greek is the language of texts considered to be standard in Modern Greek. It’s not the other way around, however distasteful that might be politically.

That term “Common Modern Greek”, Koinē Neoellēnikē, was coined by George Babiniotis, when he proclaimed that the new standard of 1976 was not just the Demotic of old that Psichari had defended: that it was indelibly marked by the heritage of Puristic. He said so in a polemical paper, “Beyond Puristic and Demotic”, a title consciously patterned after Nietzsche’s “Beyond Good and Evil”.

Yeah, you can say any number of bad things about Babiniotis, from his alleged linguistic tub-thumping in support of the junta, through his eccentric pedantic orthography, to his promotion of antiquarian inflections, and just recently his dismissal of abuse survivors in schools he was on the board of.

… As a dialectologist told me years ago: “As a human being, he is a turd (σκατάς). But linguistically, he was completely right about Demotic.” And he’s more right than Tzitzilis is.

Tzitzilis in fact knew that he was going too far in his decrying of those inflections. He did concede that there are learnèd elements that have been accepted into the Modern Standard, and that there are formal models of language variation that can be used to make sense of it. He cannot but: Psichari is still being mocked today for rejecting those learnèd elements 130 years ago. But, he insists, there are limits! There are features that the people’s linguistic intuition shall never accept!

Well, no. There are no limits. Nobody blinks now at hiatus—the learnèd violation of the universal rule, that /i, e/ before a vowel was reduced to [j]. Everyone accepts that, everyone knows that vjazo means “to rush” and viazo means “to rape”, that aðja means “empty” and aðia means “day off”, even if they are spelled identically. Because people have adjusted their internal grammars.

But that’s been no less chaotic and artificial a change to Greek than the passives Tzitzilis is decrying. It’s actually added an entirely new phoneme to the Greek lanɡuaɡe (/i/ distinct from /j/), a distinction neither Learnèd nor Vernacular Greek made, and which is all about register. itʼs thrown the phonoloɡical analysis of Modern Greek into disarray for decades (because people really did not want to accept that /j/ is a new phoneme.) It’s made for confusion diachronically: Vamvakaris in the 1930s sang “divorce” as five syllables, ði.a.zi.ɣi.o, since at least the 90s songs scan it as four syllables, ðja.zi.ɣi.o, and I don’t doubt there are people out there who pronounce it as three, ðja.zi.ɣjo. (The hiatus of διαζύγιο “divorce”)

That confusion is OK, and the passives confusion is not? That’s mere hair-splitting. And note that people’s linguistic intuition is playing a role in how that archaic passive is being revived, and how it is being integrated into the vernacular: they’re only using the 3sg –θi, and not the 1sg –θin or the 2pl –θite, because the 3sg suffix fits much more readily into the vernacular than the other persons.

Mertyris & Sampanis

The framework Dion Mertyris and Konstantinos Sampanis presented on Standard vs Koine Modern Greek was altogether more satisfactory, and I’m not just saying that because I was the unofficial supervisor of Dion’s thesis. Their framework draws on an unfortunate conflation Greek was doomed to make: the “common” language was how Greeks always referred to its vernacular—”Demotic” was a 19th century term, itself a creation of Puristic Greek. When Babiniotis rejected that the new Standard was Demotic, he fell back on the old word, Koine. But “Common Greek” is not the same thing as “Standard Greek”; and English-language scholarship has been careful to use the latter. Greek-language scholarship has not, and virtually never refers to Standard (Protypē), as opposed to Common (Koinē) Greek.

This allows them to differentiate between ±Standard and ±Koine elements of Modern Greek, without getting as stuck on the Learnèd/Vernacular, Puristic/Demotic dipole. There are elements of the Athenian Koine that are not part of the Standard language: Athens slang, and informal inflections like the southern Greek imperfect –aɣa (displaced in the Standard by Constantinople’s –usa), or the pronominal ending –onon (e.g. alunon “of others”, ðjonon “of two of them”). There are elements which are both Koine and Standard. There are elements which are Standard, but not Koine: features which are not of vernacular origin, but which are accepted in the Standard. And there are elements which are neither Koine nor Standard—Puristic archaisms which have been left out of the current Standard.

As Mertyris and Sampanis concede, the Standard has changed: the Standard of 1976 allowed more learnèd elements than the Standard of 1941. I’d add that the Koine has shifted too, if we define it, not as what is of learnèd origin, but what Greeks’ linguistic intuition does not protest. (Mertyris and Sampanis themselves refer to learnèd elements treated as compatible with the vernacular—as indeed did Tzitzilis.) The hiatus of /i/ vs /j/ is clearly learnèd, is clearly now regarded as +Koine, and just as clearly didn’t used to be. People have developed more of a tolerance for archaic elements than they had two centuries ago. Even if there are still elements they have not accepted.

(Are there limits, after all? Yes there are. But they are contingent, not predetermined. A new passive is as eligible as a new phoneme.)


The last talk I got to attend was Dionysis Goutsos’, on what the corpora of Greek he’s been involved with for the last three decades tell him about the spread of vernacular features in Greek. The curve we expect to see, from Labov’s work on sociolinguistics, is an S-curve: a feature starts infrequent, then accelerates in take-up, then levels off as it becomes universal.

In a context of language-planning, where features are promoted or suppressed in official discourse according to the politics of the day, we expect to see some wiggles in the data that don’t go in one direction, as a monotonic change. But Goutsos’ data, tracking various learnèd shibboleths through the 20th century… was weird. There was a consistent trend of lots of vernacular features in the 1900s-1910s, retreating through to the 1960s, and then coming back up—particularly in the 1970s, once the official rejection of Puristic took place (and even more so in the 1980s, when the Socialists mainstreamed vernacular choices in grammar—something we have seen a backlash to in decades since. The 1980s is when the –eōs singular of Classical –is nouns died out: poleos > polis “of the city”.) Goutsos finds a U-curve.

There were political changes one could attribute to the timeline: the political reaction to early Demotic activism, culminating in the Gospel and Orestiad riots of 1901; the relaxation on Demotic in schools in the 1920s; the Greek Civil War, and the association of Demotic with Communism. But the curve doesn’t seem to fit those dates, and I admit that I cannot understand how the U-curve happened: I would want to have a closer look at the data.

Thankfully, Goutsos did provide a closer look, breaking the corpus down by genre. The broken down results did make much more sense; in particular, movie dialogue and newsreels were level in the degree of their vernacular, reflecting consistent, vernacular-leaning prescription. The most interesting genre, with the most wiggles in frequency, were private letter-writing: private citizens took in learnèd influence, and the ups and downs in prestige of the vernacular—and they adjusted how they wrote decade by decade, without any copyediting to push them consistently one direction or the other, the way that happened in either Puristic, officialese discourse, or conversely Demotic, movie and literary discourse.

Recent Quora translations on Sarantakos’ blog

By: | Post date: 2020-03-06 | Comments: No Comments
Posted in categories: Ancient Greek, Modern Greek, Other Languages, Writing Systems

I have had updated versions of old Quora posts published in Greek on Nikos Sarantakos’ blog

Updated post on “Which formerly Ottoman-occupied peoples understand “s–tir” today?”

By: | Post date: 2019-08-10 | Comments: No Comments
Posted in categories: Linguistics, Modern Greek, Other Languages

I have had an updated version of my old Quora post Which formerly Ottoman-occupied peoples understand “s–tir” today? published in Greek on Nikos Sarantakos’ blog, as Η υστεροφημία του σιχτίρ, “The legacy of sixtir”.

Updated post on “Does the Greek word for watermelon, karpouzi, come from Ancient Greek?”

By: | Post date: 2019-07-31 | Comments: No Comments
Posted in categories: Linguistics, Modern Greek, Other Languages

I have had an updated version of my old Quora post Does the Greek word for watermelon, karpouzi, come from Ancient Greek? published in Greek on Nikos Sarantakos’ blog, as Από πού βγαίνει το καρπούζι;, “Where does karpouzi come from?”.

Updated post on “Why do English-speaking people not prefer to say natrium, silisium, kalium, and use other Latin names of elements instead?”

By: | Post date: 2019-06-22 | Comments: No Comments
Posted in categories: Linguistics, Modern Greek

I have had an updated version of my old Quora posts Why do English-speaking people not prefer to say natrium, silisium, kalium, and use other Latin names of elements instead? and Where do the distinctive Greek names for chemical elements come from? published in Greek on Nikos Sarantakos’ blog, as Τα ελληνόψυχα χημικά στοιχεία, “Greek-souled chemical elements”.

Updated post on “How are Rumi’s poems in Greek?”

By: | Post date: 2019-06-08 | Comments: No Comments
Posted in categories: Linguistics, Literature, Mediaeval Greek

I have had an updated version of my old Quora post How are Rumi’s poems in Greek? published in Greek on Nikos Sarantakos’ blog, as Τα ελληνικά ποιήματα του Τζελαλεντίν Ρουμί, “The Greek poems of Jalal ad-Din Rumi”.

Updated post on “What are some interesting examples of Ancient Greek vernacular?”

By: | Post date: 2019-05-27 | Comments: No Comments
Posted in categories: Ancient Greek, Writing Systems

I have had an updated version of my old Quora post What are some interesting examples of Ancient Greek vernacular? published in Greek on Nikos Sarantakos’ blog, as Ο Κρίμωνας και το παιδί του Βαθυκλή, “Crimon and Bathycles’ son”.

  • Subscribe to Blog via Email

    Join 325 other subscribers

  • October 2023
    M T W T F S S
    « Jul