On the retreat of Polytonic

I’ve been putting off this post because I lost an earlier draft to a crash. The Cloud will come back to bite us yet; but until it does, why can’t I have access to the Cloud on the train? Without having to remember to top up my wireless modem?

So, it started a few weeks back, when I copy-pasted a verse from the Chronicle of Morea in a comment at the magnificent Nikos Sarantakos’ blog. Nikos saw the comment, fully decked out in polytonic Greek, and asked: “Did the manuscript have those diacritics?” Because I thought the answer was obvious, and because I know how Nikos feels about the polytonic, I immediately got my back up. Like any manuscript of anything before the 19th century, Greek script meant polytonic. So of course it had them. And of course it had the wrong diacritics, and ignored iota subscript, because the scribes weren’t going to bother writing the vernacular “properly”; and the editor wrote some paragraphs justifying how he tidied it up, following a post-Classical norm. (It was 1904, you still had to justify not having an enclitic stress on properispomena in the modern language. As in χῶρος σου, not χῶρός σου; that’s how Ancient accentuation works, but not Modern.)

But the Chronicle was written, printed, and digitised in a polytonic scribal tradition, and I didn’t see the point in shaving the squiggles off. After all, noone retypes what they can copy-paste, even if they dislike the source orthography. (There had been a whole comment thread detour chez Nikos on the copy-pasting of the modern crasis σὄλεγα, which had been mis-monotonicised as the unrecognisable σόλεγα. Not that the more correct σό ‘λεγα or even σό’ ‘λεγα are much more helpful.)

“Woah woah woah”, answered my much too patient host, “I wasn’t suggesting that you shave the squiggles off. But I agree with Tasos Kaplanis that there’s no point in printing Early Modern texts in polytonic.”

At which I gulped. Tasos Kaplanis, mine godfather—for ’twas he who first called me Opoudjis—had been working at the monotonic Kriaras dictionary of the Early Modern vernacular at the time, and is now a lecturer at Cyprus U. He has put up on his shortlived Early Modern Greek Etc. blog a defence of monotonic for Early Modern texts, which he’d originally presented at the Oxford Neograeca Medii Aevi conference in 2000. It was long, it was tightly argued, it was lucid, and it was on the counterattack: the traditionalists argue that these texts should stay in the polytonic? Then what’s their argument for that, apart from mere unwillingness to change?

Though I too don’t like the concept of monotonic Early Modern texts, it’s hard to defend intellectually: I will try, but I don’t think I’ll get very far. It’s becoming even harder to defend in practice, because this is a post-polytonic world. This post sketches that world for those who aren’t in it; I’ll stutter my feeble “but… but… but…” to my godfather’s paper in another post.

In Attic Greek, accentuation is surprisingly easy. Antepenults are always acute; whether the ultima is acute or circumflexed is determined by the grammatical suffix, and which inflections are circumflexed through vowel contraction is predictable. The only ambiguity is with α ι υ in the penult, and the penult is the least common place for an accent to go. There’s still some looking up of dictionaries for those ambiguities, and the Byzantines were quite at sea with them: there’s hardly a penult ῖ that they didn’t acute, and often enough vice versa. But the system as a whole can be dealt with, and noone is seriously proposing ridding us of these troublesome squiggles. Teachers of New Testament Greek often drop them in the early parts of their courses, as a practical measure—the squiggles are less important for their audience than for trainee classicists, and by Christ’s time had no phonetic value. But the Koine New Testament is still published in polytonic.

For the modern language, on the other hand, the polytonic is burdensome and artificial, and a whole philology grew up around deciding when to circumflex and when to acute. Because of dialect mixing in the Koine, you can no longer tell whether a final α is meant to be long or short. Because of dropped final ον, there’s no right accentuation for etymologically diminutive neuters—whether μοναχοπαίδιον should be accented as μοναχοπαίδι (just cut the ον), or μοναχοπαῖδι (apply the accentuation rules from scratch). Is my name in Greek and Sarantakos’ Νῖκος or Νίκος? It’s bad enough having to memorise ancient long penults, as the Byzantines found; if you truncate names like Νικόλαος, you’re now accenting in the penult vowels that had never before been exposed to the choice. If the word was borrowed at any time in the last two millenia, any notion of a vowel being long or short is artificial: why would you put a circumflex on a loanword from Turkish? And so on.

Polytonic was an ill-fitting enough garment on Modern Greek, that it got progressively toned down from the rules applicable to Attic. By the 1960s, it had settled on a lightweight form, as practical as could be expected: drop breathings on rhos, drop graves, back away from iota subscripts, when in the slightest doubt go with the acute. Once Puristic Greek was dropped in 1975, with its haphazard dependence on Anything But Vernacular Greek, it was relatively easy in 1982 to legitimise what the press had already been doing in practice in the ’70s, release schoolchildren from their travails, and drop the extra squiggles.

So you’d only write Modern Greek in polytonic now if you wanted to assert continuity with the Greek written tradition, against the 1982 spelling reform. The monotonicists reading this shouldn’t get too smug: you’re writing in Greek script, and with a mostly historical orthography, for the same reason of asserting continuity. These are matters of degree; and just like the shades of gray between Puristic and Colloquial Greek, they are politicised degrees. The vehicle of the traditionalist conservative portion of society used to be Puristic. Now, it’s impossible to get anyone to write in Puristic, without your audience bursting out laughing; so the traditionalist portion (or a small and de facto marginal subset) can only assert its traditionalism through polytonic Demotic. Which still gets laughed at anyway.

Psichari taunted Puristic Greek for its eclecticism, which made it linguistically incoherent. The opponents of polytonic can likewise taunt the proponents with “which polytonic are you going to choose?”—because the ill-fitting garb has gone through several successive tailorings. But precisely because using the polytonic now is a matter of ideology, the polytonicists don’t choose the lightweight polytonic of the 1960s. They don’t want learnable squiggles, they want authentic Hellenistic squiggles: if someone is making the effort to put circumflexes, they will more than likely put graves in as well. (Not breathings on the rho as much.)

I was saddened when I found out that Hestia, the one Puristic holdout of the Athens press in the ’80s, switched to polytonic Demotic in the ’90s: that to me was the real death of Puristic. The fact that polytonic Demotic succeeded a fifteen-year Last Stand of Puristic (and that Hestia was the last newspaper printed with Linotype), though, tells you a lot about the role polytonic Demotic is now assumed to play in Greek society.

The monotonic has been official a generation now. In the 1930s, the accentuation rules were what you did in primary school, entrenched enough to write a rebetiko song on:

—I haven’t finished school.
I haven’t learned that much.
But I do know 1 + 1 is 2,
And that there’s seven vowels.

—So long together, yet you still don’t know
my likes, what makes me tick.
The antepenult’s never circumflexed
when the ultima’s long.

In the 1930s, everyone was supposed to know the song was misleading. Whether they did depended on how seriously they took the rules, because the song certainly didn’t: the penult’s never circumflexed when the ultima’s long; the antepenult’s never circumflexed at all. Kids now get taught that rule still, but in high school, with Ancient Greek (whether optionally or compulsorily depends on the current education minister). People no longer live in a polytonic world.

Doing anything in the Modern Language in polytonic is increasingly less tenable. The people who knew the polytonic rules are getting pensionable, and the diehard online advocates of the polytonic rely on software for their accent rules—much more than it is safe to do. Polytonic keyboards are around, but they aren’t the default, and most people don’t see the point in hunting them down. Polytonic Greek characters are now widely available even in default System fonts, but at least one commenter on Nikos’ blog is complaining about The Reign of Blank Squares, because that’s all Internet Explorer is letting her see. Typing in polytonic was always more work than using a single acute, so even if you’re transcribing a polytonic text, you’re not going to make the extra effort unless you see a point to doing so. Beševliev did not print Lord Krum’s text in monotonic, but I did. Of course, Lord Krum’s POWs chiselled the stone in all caps, so any diacritics were a modern intervention anyway.

Because most people aren’t going to bother getting a polytonic keyboard or using it, you will see snippets of Ancient Greek on blogs and the press, cited in monotonic. As here (possibly a cut and paste, since the post also cites another Ancient passage in polytonic):

Η γαρ ευγένεια τώνδε των ανδρών εκ πλείστου χρόνου παρά πάσιν ανθρώποις ανωμολόγηται. Ου γαρ μόνον εις πατέρ’ αυτοίς και των άνω προγόνων κατ’ ανδρ’ ανενεγκείν εκάστω την φύσιν έστιν, αλλ’ εις όλην κοινή την υπάρχουσαν πατρίδα, ης αυτόχθονες ομολογούνται είναι. Μόνοι γαρ πάντων ανθρώπων, εξ ήσπερ έφυσαν, ταύτην ώκησαν και τοις εξ αυτών παρέδωκαν, ώστε δικαίως αν τις υπολάβοι τους μεν επήλυδας ελθόντας εις τας πόλεις και τούτων πολίτας προσαγορευομένους ομοίους είναι τοις εισποιητοίς των παίδων, τούτους δ’ γνησίους γόνω της πατρίδος πολίτας είναι.

Ἡ γὰρ εὐγένεια τῶνδε τῶν ἀνδρῶν ἐκ πλείστου χρόνου παρὰ πᾶσιν ἀνθρώποις ἀνωμολόγηται. οὐ γὰρ μόνον εἰς πατέρ’ αὐτοῖς καὶ τῶν ἄνω προγόνων κατ’ ἄνδρ’ ἀνενεγκεῖν ἑκάστῳ τὴν φύσιν ἔστιν, ἀλλ’ εἰς ὅλην κοινῇ τὴν ὑπάρχουσαν πατρίδα, ἧς αὐτόχθονες ὁμολογοῦνται εἶναι. μόνοι γὰρ πάντων ἀνθρώπων, ἐξ ἧσπερ ἔφυσαν, ταύτην ᾤκησαν καὶ τοῖς ἐξ αὑτῶν παρέδωκαν, ὥστε δικαίως ἄν τις ὑπολάβοι τοὺς μὲν ἐπήλυδας ἐλθόντας εἰς τὰς πόλεις καὶ τούτων πολίτας προσαγορευομένους ὁμοίους εἶναι τοῖς εἰσποιητοῖς τῶν παίδων, τούτους δὲ γνησίους γόνῳ τῆς πατρίδος πολίτας εἶναι. (Demosthenes Epitaph 4)

With such Puristic Greek as makes it online, you’ll routinely see them in monotonic:

οι Σάξωνες ανήγειρον πάλιν την θρασείαν και ακτένιστον κεφαλήν των και βυθίζοντες την χείρα εις το αίμα ουχί ταυρείων, αλλ’ ανθρωπίνων θυμάτων ώμνυον εις τον Τουΐτονα, τον Ιρμινσούλ και Αρμίνιον, ή ν’ αποσείσωσι τον Καρόλειον ζυγόν, ή διά του αίματος αυτών να φυράρωσι του Άλβιος και Βισούργιδος τας όχθας. Ήλθεν, είδε και ενίκησε κατά το σύνηθες ο άμαχος Αυτοκράτωρ διά της λόγχης εκείνης, ην κατά τους Ευαγγελιστάς εβύθισεν ο Ρωμαίος στρατιώτης εις του Σωτήρος την πλευράν, ο δε αρχάγγελος Μιχαήλ εμφανισθείς καθ’ ύπνους τω Καρόλω εναπέθεσεν επί της κλίνης του, ίνα κατά τους χρονογράφους ανταμείψη αυτόν, διότι και από εψημένου και ωμού κρέατος απέχων την Τεσσαρακοστήν εκοιμάτο μόνος.(Emmanuel Rhoides, Pope Joan, Project Gutenberg copy)

οἱ Σάξωνες ἀνήγειρον πάλιν τὴν θρασεῖαν καὶ ἀκτένιστον κεφαλήν των καὶ βυθίζοντες τὴν χεῖρα εἰς τὸ αἷμα οὐχὶ ταυρείων, ἀλλ’ ἀνθρωπίνων θυμάτων, ὤμνυον εἰς τὸν Τουίτονα, τὸν Ἰρμινσοὺλ καὶ Ἀρμίνιον ἢ ν’ ἀποσείσωσι τὸν καρόλειον ζυγὸν ἢ διὰ τοῦ αἵματος αὐτῶν νὰ φυράσωσι τοῦ Ἄλυος καὶ Βισούργιδος τὰς ὄχθας. Ἦλθεν, εἶδε καὶ ἐνίκησε κατὰ τὸ σύνηθες ὁ ἄμαχος αὐτοκράτωρ διὰ τῆς λόγχης ἐκείνης, ἣν κατὰ τοὺς Εὐαγγελιστὰς ἐβύθισεν ὁ Ῥωμαῖος στρατιώτης εἰς τοῦ Σωτῆρος τὴν πλευράν, ὁ δὲ ἀρχάγγελος Μιχαὴλ ἐμφανισθεὶς καθ’ ὕπνους τῷ Καρόλω ἐναπέθεσεν ἐπὶ τῆς κλίνης του, ἵνα κατὰ τοὺς χρονογράφους ἀνταμείψῃ αὐτόν, διότι καὶ ἀπὸ ἐψημένου καὶ ἀπὸ ὠμοῦ κρέατος ἀπέχων τὴν Τεσσαρακοστὴν ἐκοιμάτο μόνος. (Wikisource copy, with some accent corrections)

I find the monotonic somewhat harder to read, because in Ancient Greek, the diacritics aren’t purely decorative: they do help disambiguate the text. If people are careful with how they render Ancient or Puristic Greek in monotonic, some of those ambiguities can be dealt with: article ο vs. relativiser ό (ὁ, ὅ), or interrogative τίς vs. indefinite τις. Because monotonic Ancient Greek happens on an ad hoc basis, people often aren’t careful enough to make up that kind of stress-based disambiguation. That’s actually one of my main concerns with putting older stages on the language into monotonic, though admittedly it’s not an insurmountable one.

In fact the heavy lifting of disambiguation in Ancient Greek isn’t really done by circumflex vs. acute at all: there are minimal pairs, like ἤρα “he loved” ~ ἦρα “I lifted”, but there aren’t many. The breathings do some more work than the accents (ὀδός “threshold” ~ ὁδός “road”), and because of the way inflection works, the iota subscripts do a lot of work (nominative λύπη dative λύπῃ). Without any of these hints, you can still read Ancient Greek (epigraphers do, after all, and so did Ancient Greeks). But it’s somewhat more haltingly than it needs to be (as I found with the Demosthenes at least): we’re not native speakers of this lingo, we shouldn’t be deprived of hints.

So if you use accents to disambiguate monosyllables in a systematic way, and are not so purist that you refuse to allow iota subscripts in your monotonic, you’ll get a transcription which is workable at a last resort for Ancient Greek, and may be close to acceptable for Puristic. Of course, a monotonic with iota subscripts might seem to defeat the purpose of having monotonic at all; but monotonic was not designed for a language with a productive dative.

So just as there are different flavours of polytonic, of increasing irrelevance to the modern language, there can be different flavours of monotonic, of decreasing incompatibility to the ancient language and the stages in between. There were multiple flavours before 1982, and there are decisions taken in 1982 that people still chafe on, like clitic του or the interjection να. But of course, the point of getting a State-approved monotonic was that there should only be one flavour (and that noone could counter-taunt “which monotonic are you going to use?”): Kriaras had to change the system he was using in his dictionary after 1982, even if he didn’t agree with all the decisions made. Still, he has pretty much tweaked his monotonic to deal with archaic elements the way I’m suggesting. So this is his entry for κυκλεύω, written in 1985:

ῴ (ενν. τῳ ποταμῴ) κλήσις Αμαζονικός, έχων κύκλευμα χρόνου· … τέλος, αρχήν ουκ έχει Βίος Αλ. 5503. [Kriaras modified monotonic]

ω (ενν. τω ποταμώ) κλήσις Αμαζονικός, έχων κύκλευμα χρόνου· … τέλος, αρχήν ουκ έχει Βίος Αλ. 5503. [raw monotonic]

ᾧ (ἐνν. τῷ ποταμῷ) κλῆσις Ἀμαζονικός, ἔχων κύκλευμα χρόνου· … τέλος, ἀρχὴν οὐκ ἔχει Βίος Ἀλ. 5503.[polytonic]

So if people want to monotonicise Early Modern, Puristic, or otherwise macaronic texts, they can—as long as they don’t monotonicise them in exactly the same way they accent the contemporary language, and the way they do monotonicising them is consistent and documented.

They can; should they? After all, there’s more digitised monotonic Emmanuel Rhoides than Apollonius Rhodius out there, and there’s a reason for that. And Early Modern Greek texts are only slowly starting to be published in monotonic, and there’s a reason for that too. It’s not purely a linguistic reason. But that’s for the next post.