Talk:Ancient Greek accent

Recent edits

User:Thanatos666 asked for an explanation of why I objected to his recent set of edits [1]. Well, fair enough, and I don't want to sound impolite, but then... sigh. Where to start?

• insertion of "tonoi" section heading: leaves the intro far too short, and "tonoi" is not really a common term in English
• addition of "(tonos; tonoi in the pl.)": awkward phrasing, disruption of sentence flow, and unnecessary (we don't need to discuss the grammar of the Greek technical terms here)
• addition of disclaimer "(the /e/s are not really two...": unnecessary and awkward editorializing; if you think that representing a long vowel with two short ones is phonologically wrong, then just don't use that representation, don't first use it and then say it's wrong. But as it is, this is actually quite a common assumption in many branches of phonology.
• " the αι diphthong in this case is, or is treated as being, short)": again, unnecessarily awkward phrasing, and unnecessary complication in the example – why not instead use a more straightforward example where the shortness of the last syllable is unambiguous?
• addition of "(likewise)": unclear what it refers to; another piece of confusing pedantry
• example "Μακρὰ συλλαβή": another confusing example, misleading the reader into confusing the point being illustrated with the meaning of the words used to illustrate it (the rule that's being illustrated has nothing to do with "long syllables". An example that's not taken from the topic domain of phonology itself would be much better here.
• addition of " (unless, that is, the supposed exceptions should be considered a case of, for example, hidden, underlying synizesis)" – another piece of unwarranted editorializing and speculation. If the exception is motivated by synezesis, just find a source that says so and cite it. Also, again, needlessly awkward wording
• "In English they are called – names derived from Latin – the ultima ("last"), penult ("almost last"), and antepenult" ... – another piece of tortured wording. You have a habit of sticking parenthetical material somewhere into sentences in ways that really disrupt the natural reading flow in quite extraordinary ways. Same for the next bit: "Their names in Greek – they have, more or less, the same meaning - are respectively..."
• Tables in "length of accented vowels" section: now these are really just plain wrong. The idea is not that there is a rising contour within a short vowel, as your tables seem to indicate, but that the single-mora vowel is simply stationary higher. And the "Alternatively we can use arrows to graphically display and represent rise of tone" is again unnecessary awkward editorializing.

Sorry, I can appreciate you have been working hard on these pages, and I hate to be so critical, but the quality both of the presentation and of the content is really not satisfactory. Fut.Perf. ☼ 14:20, 1 April 2014 (UTC)

• 1: True. It doesn't hurt to be mentioned (spread of knowledge), related word (tonic; or was it as in gin&...) had already been mentioned elsewhere, its not that big of a problem, but anyway OK. Easily solvable.
• 2: a.Rephrase/reword it if you think it's awkward... Or suggest, aske me to do it. Before reverting... b.I could say we can mention such words. I could say we must mention such words. I could say we must do it if, e.g. in the future we're to add references (is something missing or is it my idea??). Refs that prorbably include ancient sources... c.But OK. Let me give in. Let's take this out. At least for now...
• 3: A.Is Athenai written in invisible ink? B. I can use the method, even if I disagree with it, if I qualify/explain it, making the issues clear to the everyday joe/jane that might get to read this, which is exactly what I did. You on the other hand are using it (or protecting/defending stubbornly its use) unqualified and unexplained. I.e. something highly misleading to the non specially trained, knowledgable eye and mind. This is the case no matter what and why I believe about long vowel representation as far. e.g., as the quite a common assumption in many branches of phonology is concerned...
• 4: The text read as if it's weird (these) diphthongs being short (common mantra on..) I rephrased covering both possibilities. I'm being open to different views. Or put in another way I'm playing it safe... hehehehe
• 5: Reference is the /ee/. The added indented passage is very structured, symmetrical; look again... Perhaps you're being a bit too hasty?!?!?
• 6: I liked the interplay. Very much so!! But Ok no problem, chose some other word, if you find this troublesome. Deletion, though, is simply unjustifiable...
• 7: are you sure I'm speculating?!?! PeRhaps you should refresh you memory (provided my guess on your knowledge is correct)? ;-)
  FYI: They way I see it I was probably the one that had added in the past the mentioning of the exceptions to the rule:at the Ancient Greek Phonology article; then there was probably a copy paste around articles... Today, I just bumped into it again and thought that counterbalancing possibilities -not unheard of, I repeat, brush up...;-) it's old stuff...- should be added...
• 8: I like parentheses. :) English is so restrictive in its damn word order. I like freee flow!!! I like to emphasise what I want when I want, without having to rewrite everything... :) But OK no problem: reword them! Or ask for theri rewording. Just do not delete them. A strangely phrased passage, that contains valuable information justifies a rewording/rephrasing, not outright deletion...
• 9: However things may be, all 3 relevant articles read as if everything is about rising and/or falling pitch. I personnally prefer a view of, let's say, high/normal/and-perhaps-lower tone levels and 1 rising/falling tone view, but I was just being open to what the articles you're so keen to protect, defend in essence read... ;-) But Ok we could work on this. Why did you also delete the rising/falling table? Why did you negate the centering/reformating/shaping-up at the old tables. Why, Why, Why...???

Sorry but then I will also have to be blunt: the prior state was at least an order of magnitude worse; an interesting mess, it was and now is again... ;-) I, on my part, appreciate your work in general. I also appreciate that you did write down this; done post-edit-reverts, but things could be worse... What I cannot in anyway appreciate is your edit reverts in these article parts/sections that more or less suck, your unwillingness to improve, work upon my edits, etc...
P.S.Enough for now from me. Edit-reverting -if needed- on my part will come in the evening and onwards; you have got all the time in the world...till then, to get/make constructive ideas and edits/changes... Or to get ready for battle!!! :D Thanatos|talk|contributions 15:27, 1 April 2014 (UTC)

P.P.S. Oops I forgot: Where there were indeed, in fact, admittedly, many problems on my edits, was in the perispomenon article! YET AGAIN they are-were-are easily fixable if one is actually is willing to, open to... Thanatos|talk|contributions 15:33, 1 April 2014 (UTC)

Summary of recessive accent, and whether AG accent is free

It seems to me that the section entitled 'A summary of the ultima rule' is unnecessary, since the rule is explained perfectly clearly in the preceding section. The 'summary' only complicates matters and is harder to understand than the preceding section. I suggest that the person who contributed this section should delete it.

I deleted the word 'free' in the phrase 'free pitch accent' in the first line since it isn't quite accurate; the accent in Greek wasn't entirely free (unlike that of Vedic), but restricted to the last three syllables.Kanjuzi (talk) 19:39, 1 September 2015 (UTC)

Yes, the summary rule is difficult to understand. I included it because it's an example of an attempt to reduce the recessive accent rules to a single definition, or to figure out the underlying motivation for the rules. It's probably not helpful for people learning Greek accent rules for the first time. Maybe it should be moved to a different section, titled "linguistic analysis" or something.

Free is defined in the intro. In the case of Greek, it means that the accent rules allow the accent to be placed in one of several different positions in a word. Ancient Greek accent isn't as free as Vedic, because it's more restricted, but it's more free than Latin, in which words can only be accented in one position for the most part. I think this use of free is generally accepted, but I don't have a source to back this up. — Eru·tuon 18:38, 2 September 2015 (UTC)

I applaud any attempt at simplifying or summarizing a wordy (set of) rule(s)! And I do not exempt the present attempt. If it succeeds, I will be grateful.

I find, however, one problem with the current formulation of the summary: as far as I understand, it does not explain the difference in legitimization between λόγος and *δήμος. In both cases, there is no mora at all between the accented morae, yet only the first word is legitimized by the rule.

Can the summary be adapted in a way such that it really covers all cases?Redav (talk) 15:40, 11 September 2017 (UTC)

Ultima rules

Here it says, among other things, that if the ultima is short and the penult is long and accented, the accent on the penult must be a circumflex. What about the first word of the Odyssey? Seadowns (talk) 11:12, 12 November 2017 (UTC)

That's true. The rule isn't very precisely formulated. Also the idea of "moras" (which usually refers to syllable length not vowel length) would be much better replaced with "long vowel" and "short vowel". Kanjuzi (talk) 13:18, 12 November 2017 (UTC)

It is wrong, not vague! And has been for years, it seems. Seadowns (talk) 15:57, 12 November 2017 (UTC)

Transcriptions

I wonder why, when the transcriptions appear to represent pronunciations rather than phonemes, you have put them in between slashes / / rather than in square brackets [ ] as is normally done. Wouldn't square brackets be better? Kanjuzi (talk) 20:20, 19 September 2015 (UTC)

Incidentally, I notice that you have transcribed σημεῖον as /sɛ͜ɛ.mé͜e.on/ "sign". I don't think this is quite correct, since it appears that ει was pronounced differently before a vowel. See for example Allen (3rd edition) p.72, 83. One piece of evidence is Latin transcriptions such as Medea, Aeneas etc. with 'e' as opposed to Nilus, Pisistratus with 'i' etc. Allen's conclusion is that before a vowel ει was pronounced [eyy] (where he writes 'y' for IPA [j]). Perhaps you could choose a different word to illustrate properispomenon words such as γυναῖκες. Indeed I am wondering why you chose some of the words you did. The Nom.pl. ἀλήθειαι for example doesn't occur anywhere in the whole of Greek literature. Why not choose ἕκαστος, for example, which is a word learners of Greek might actually meet with (it occurs 982 times)? Or if you want a word ending in -αι, how about βούλεται or λέγεται? The latter occurs 1382 times so is very common. Kanjuzi (talk) 17:57, 20 September 2015 (UTC)

The reason for choosing ἀλήθειαι was that it's a noun (thus has different case forms), has a long vowel in the penult, and an -αι as an an ending. However, now that you mention it, I realize it's naturally a bad form, because it's an abstract noun that is never pluralized, like information in English. The plural meaning "true things" would be ἀληθῆ. So another word should replace it. It has to be a plural feminine noun with three syllables, recessive accent, and a long vowel in the penult. Any ideas?

The question of phonemic versus phonetic transcription is hard. Allen doesn't go into that distinction. I would agree that some transcriptions in the article are more phonetic than phonological (like the ones that use the rising and falling pitch symbols), but some people would say you can never transcribe the phonetics of a dead language. (I would disagree, to some extent.) I agree that σημεῖον is probably phonologically /sɛ͜ɛ.mé͜i.on/, but that's just my opinion, not something substantiated by a reliable source that I know of. — Eru·tuon 19:16, 20 September 2015 (UTC)

Well, ἀλήθειαι is sometimes used in the plural (mostly by Epictetus, e.g. οὕτως τυφλὸς ταῖς ἀληθείαις καὶ κωφὸς εἶ; ‘Are you so blind and deaf to the truths?’) but just not in the Nominative, so it might be best to avoid it. You could always use δέσποιναι or βασίλειαι which are found (6x and 3x respectively, as you can check from the ‘Perseus Philologic’ website). /sɛ͜ɛ.mé͜i.on/ is a much better transcription, but I should put it in square brackets: [sɛ͜ɛ.mé͜i.on] Allen says (p.10) 'When indicating particular sounds in a phonetic notation it is customary to use square brackets, e.g. [t^h] to represent the initial sound of English tin; phonemic symbols, on the other hand, are conventionally set between obliques, e.g. /t/ for the phoneme which includes the initial sound of tin and the final sound of hat.' As for the pronunciation [ei] before a vowel, it isn't, I think, a matter of opinion, but the evidence is set out quite clearly by Allen in the pages I cited, so I should go ahead with it. Kanjuzi (talk) 05:02, 21 September 2015 (UTC)

I'm quite aware of the evidence that shows ει before a vowel was pronounced as a diphthong. My point is that Allen doesn't phonemically analyze it as either /ee/ or /ei/. I suppose you're right it has to go between square brackets, but I'm dissatisfied with that, because the whole point of putting the acute accent on the accented mora was to show phonemic analysis.

I was under the impression you were saying ἀλήθεια was never used in the plural, and if so then it would be a form without a meaning and shouldn't be used. (It would be a bad form, ungrammatical, like someone saying "informations" in American English. I remember reading that some other English dialect uses that plural form, but in American English it's incorrect. "Pieces of information" or something like that would be used instead.) But since it is used in the plural, the plural form does have a meaning and it doesn't matter that it happens not to be attested in the nominative. So it's fine to use it as an example. δέσποιναι and βασίλειαι are fine too, though.

You can make changes if you feel you can improve the article. The article is problematic in many ways now that I look back at it, though I do like some of the examples I chose and some of the other work I did. I just don't have the brainpower to spend on improving it right now. — Eru·tuon 06:55, 21 September 2015 (UTC)

Well, what I find about the article is that, although obviously you have put a lot of work into it, is that after a very brief introduction it goes straight into the moraic theory. That is to say, it gives a theoretical explanation of the facts without first explaining the facts themselves in a simple way. I would like, if you don't mind, and if I have time in the next few days, to extend the introduction a bit to put the moraic theory in context. Let us suppose that we are writing it for, say, a person studying New Testament Greek who wants to know a bit more about accents, but doesn't want to be weighed down by theoretical matters to begin with. We should also say a bit about the origins of the accents, and what became of them; and put Ancient Greek in the context of other tonal languages (such as Bantu languages) with the same type of tonal accent. Would you allow me a few days to put together a few notes and add them on the front. If you don't like them, you can always remove them later.

As for the moraic theory itself, I don't think it's that essential. I notice for example that in Probert's (2003) 215-page book on Greek accentuation she devotes just 6 lines to the theory. What I find unsatisfying or over-complex about it is that counting morae doesn't seem to explain why for example ἄνθρωπος and πόλεμος have the same accent, why, for example ῥινόκερως 'rhinoceros' doesn't obey the rules. It is certainly complicated for beginners, for whom perhaps a simple rule such as 'consider the length of the last vowel' would be more satisfactory. But let me write a few lines first and then see what you think of them. 41.78.248.58 (talk) 14:37, 21 September 2015 (UTC) Kanjuzi (talk) 14:40, 21 September 2015 (UTC)

It sounds like you have some good ideas for improvement. It's certainly true that ῥινόκερως completely breaks the accent rules. For some reason I've never seen the word before (in Greek at least). I don't see how the accent placement of ἄνθρωπος and πόλεμος are puzzling (they have the same mora count in the accented syllable and the ultima), but if it really is confusing based on what the article says, then the article's description needs improvement. Fresh eyes like yours are better than mine in this case.

It would be great to describe how accent changed in Koine Greek, but I haven't read any good sources that go into detail on that topic. The most I've read is that the accent changed from pitch to stress, with no description of how exactly that happened. I have my own theory, which hasn't been published and can't be used in the article. Maybe someone else more recent than Allen has come up with a similar theory. I'm not very well-read on this topic. — Eru·tuon 17:47, 21 September 2015 (UTC)

Well, with regard to modern Greek I'm sure there is no need to say more than Allen and Probert do, but presumably it went through a stage where it was both pitch and stress (something like Persian, perhaps). It is not surprising that you have not come across ῥινόκερως, though; according to the excellent Perseus Philologic search engine (do you know it?) it occurs only once in literature, in Pausanias. But χρυσόκερως/χρυσόκερων 'golden-horned', which also violates the rule, occurs in four authors. Kanjuzi (talk) 19:09, 21 September 2015 (UTC)

I'm sort of familiar with Perseus, but I mainly just use the LSJ on there. I can't remember having ever encountered any "horn" compounds. But I haven't read Ancient Greek literature very widely.

I'm pretty sure Ancient Greek accent consisted of both pitch and intensity or stress, and the two were closely related, occurring on the same or neighboring syllables. The change in accent happened when pitch and intensity merged together onto the same syllable. That's the barebones of my theory. I'm not familiar with Persian, so not sure what you mean. — Eru·tuon 21:20, 21 September 2015 (UTC)

Yes, if you listen to a modern language which has an accent or tones similar to ancient Greek (such as some of the Bantu languages of Africa) the high tones do very often sound just like stresses. So when a Malawian says Máchinga (a place name) it sounds more or less the same as when someone says Mánchester. The main difference seems to be that the word Mánchester can of course be reversed and pronounced with a rising tone, which Máchinga cannot. I think it would be the same with ἄνθρωπος. The ancient Greek would always have the first syllable on a high pitch, whereas in modern Greek it can sometimes be high and sometimes low. As for Persian, when you listen to it it sounds at first as if it has a stress-accent - the accented syllables are definitely louder. But when you listen carefully it you find that the stressed syllables are also always higher-pitched than the unstressed ones. I imagine that, as you say, Greek went through a similar stage where the accented syllables were both higher and louder than the unaccented ones. - As for those 'horn' words, I only came across them myself this week. I was about to correct someone's work but found to my surprise that correction wasn't needed! Probert doesn't mention these words in her smaller book. — Preceding unsigned comment added by Kanjuzi (talk • contribs) 04:25, 23 September 2015 (UTC)

'This'

I am intrigued to see that certain people who edit Wikipedia articles like to go through them removing all instances of the word 'this'. Perhaps there is a rule taught in schools in certain countries that the word 'this' is to be avoided in formal English. Another word which the same or similar editors seem to dislike is the word 'which', even when it is used perfectly correctly, as in the present sentence. Kanjuzi (talk) 06:46, 19 February 2016 (UTC)

A General Comment

Thank goodness there are other articles on this subject on the net! As for the moraic theory, I rather wonder whether the derivation is Latin or Greek.

The words boukeros and taurokeros are in the Bacchae, I think. (This is from memory only, and I can't get the Greek letters.)

If it is any consolation, I have read that even Richard Bentley made mistakes with his accents. Seadowns (talk) 13:16, 24 October 2017 (UTC)

I should think anyone would make mistakes in their accents who does not pronounce them. Kanjuzi (talk) 13:55, 24 October 2017 (UTC)

Yes, and those who do pronounce them too. In any case, do we really know how they were pronounced? I have personally known Lobel, Dover and Page, and often heard Eduard Fraenkel and Dodds in lectures, and none of these great scholars, or many others, bothered with this sort of speculative endeavour. Seadowns (talk) 19:22, 24 October 2017 (UTC)

I can't answer "Do we actually know?", but I would say that there is some evidence on the pronunciation of the accent starting in the early Koine period (for instance, musical notation), and that of the many possible questions regarding accent, some have more evidence pertaining to them than others. — Eru·tuon 21:15, 5 November 2017 (UTC)

The difficulties came when one was given 3 hours to translate a page of classic prose or verse, without any books, and one tried to get the accents all right from memory. Even people who try to pronounce them have to know what they are first. Seadowns (talk) 14:50, 4 November 2017 (UTC)

I think what Wikipedia needs is an article that would enable somebody with it and a lexicon to accent an unaccented text, or perhaps to read the proof of a text, ensuring the accents are right. For this, I think the only phonological information you would need would be how to tell long and short syllables. No need for morae. The phonological material now in this article in relation to accents could be written up separately, if desired. Seadowns (talk) 18:24, 12 November 2017 (UTC) Seadowns (talk) 10:56, 13 November 2017 (UTC)

I agree with Seadowns – this article as it stands is a little disappointing and could certainly be made more informative. Too much space is taken up explaining about moras and not enough to show readers where the accent should be placed, supposing they wish to learn to do it, or for that matter the evidence for how scholars think the accent was pronounced. (Surprisingly, the article at present hardly makes any reference to any modern authorities such as Probert, or Devine and Stephens.)

The mora question is actually more complex than it appears: for example, the second syllable of κῆρυξ has two moras metrically, but counts as only one accentually; whereas in Καππάδοξ the final syllable counts as having two moras both accentually and metrically. Isn't it simpler to say that the second vowel of κῆρυξ is short, rather than saying it has one mora and then having to specify what you mean by mora?

Some parts of the article, such as the section on pitch contours, seem rather puzzling, in my view, and make what is rather simple appear very complex. I suggest that these could be simplified and the material explained in words, preferably without the use of IPA symbols.

At any rate, I am shortly going to add a section on evidence for how the accent was pronounced, which at present is very scantily dealt with. This will no doubt make the article seem unbalanced at first but when more material is added with guidelines explaining where to place the accent, it will hopefully become balanced again. Kanjuzi (talk) 11:38, 12 July 2018 (UTC)

@Seadowns: The article has now been made longer and more informative, and your input would be valued if you have any suggestions for improving it. Does it now give the information you require? Kanjuzi (talk) 11:36, 22 July 2018 (UTC)

Comparison with Latin

This is not relevant, and it would improve the article to delete it altogether.

Also, it says that if the penult was light (=short) it was not accented. What about 'fuit", then? Where is the accent? Seadowns (talk) 18:38, 24 October 2017 (UTC) Seadowns (talk) 18:45, 24 October 2017 (UTC)

I corrected the statement. You're quite right, it was wrong. — Eru·tuon 19:07, 24 October 2017 (UTC)

I agree, this part of the article is totally irrelevant. We do not need to know the rules for placing a Latin accent in an article about the Greek accent. I have deleted it. Kanjuzi (talk) 02:54, 25 October 2017 (UTC)

@Kanjuzi: I don't know what you mean by "totally irrelevant", but a comparison of the two is certainly related to the topic of the article, and if done well, it could be useful to language learners. — Eru·tuon 01:39, 5 November 2017 (UTC)

In my opinion it was not done well and the article is much better without it. It would be just as odd, perhaps, in an article about Spanish pronunciation, to find a long paragraph all about the exact rules for pronouncing Portuguese or Italian. The rest of the article has not been done well either; one good place to start improving it would be to begin by removing the absurd (and inaccurate) transcriptions such as /a.lɛ͜ɛ.tʰe͜é.a͜as/. These are absurd (a) because the exact pronunciation of the word is irrelevant to the accent; the pronunciation of the letters changed over the centuries (e.g. between the time of Thucydides and Demosthenes) but the accent remained the same; (b) such transcriptions are completely off-putting to the average reader, who may be for example, a student at school or a person studying New Testament Greek, and who simply wants some brief rules of thumb for placing the accents; (c) the examples chosen, such as the plural of ἀλήθεια, are hardly ever met with and (d) inaccurate because in any case according to Allen ει before a vowel was not pronounced e͜e. Kanjuzi (talk) 05:51, 5 November 2017 (UTC)

I made those transcriptions to demonstrate a typical phonological analysis of the accents in the Classical period, in which the acute indicates accent on the last mora of a vowel, and circumflex on the first mora of a two-mora vowel. Now I realize that's anachronistic; the accent marks were apparently invented in the Koine period, so they are probably somewhat inaccurate for the accent of Classical Attic and more inaccurate for other dialects. Where the inaccuracies lie is hard to determine.

I think some sort of phonological transcription does have a place in the article. It explains the origin of the accents, which only makes sense during the period when Ancient Greek had a vowel length contrast and some form of pitch accent. For the periods in which there was stress accent and no contrast in vowel length, the distinct accent marks have no significance; a tonos would have done just as well. That includes later Koine and Byzantine at least, and maybe New Testament Greek as well. But I agree that it's confusing and misleading to have pseudo-Attic transcriptions all over the article, with no indication as to what period they pertain to. (That's aside from the inaccuracy in the transcription of prevocalic ει that you point to.) It's also completely wrong that all dialects or periods of Ancient Greek had a pitch accent, or the same pitch accent represented by the accent marks. Probably the explanation of the sort of mechanical rules for placing the accent marks should be separated from the phonological significance of the accent marks in the different periods and the explanation of how the accent rules may have made sense phonologically when the accent marks were invented. — Eru·tuon 20:21, 5 November 2017 (UTC)

The Latin stuff can be found elsewhere. Seadowns (talk) 10:35, 5 November 2017 (UTC)

Section on morae

I am going to delete the section on morae, as it is not needed and perhaps even confusing. All the scholars such as Kiparsky and Mouraviev who try to formulate the rules in terms of morae make it seem over-complicated and do not seem successful. For example, Mouraviev (1972) proposes: "The accent in Ionic-Attic Greek orthotonic words could stand on any of the following three places: (A) on the last vocalic mora; (B) on the vocalic mora immediately following the mora defined in (C) below; and (C) on the vocalic mora immediately preceding the syllable containing the penult". I think there are simpler ways of describing it. The position of the accent on the antepenultimate syllable also doesn't depend solely on the vocalic morae of the final syllable, but also on its weight (for example, it is not allowed if the word ends in -ax). So I think nothing will be lost if we delete the paragraph. Kanjuzi (talk) 11:30, 22 July 2018 (UTC)

Lang tags

@Kanjuzi: The words that came up on the spell check of the 2020-05-20 dump for this article which don't look like English words are:

prosōdía, oxeîa, bareîa, perispōménē, padás, padí, yugáṃ, áśvaḥ, śatáṃ, návaḥ, papríka, pápríka, melism, duhitá, ăs, ŏs

-- Beland (talk) 06:19, 4 June 2020 (UTC)

I added some templates, except to the following ones:

papríka, pápríka, melism, ăs, ŏs

--178.1.233.143 (talk) 19:06, 28 August 2022 (UTC)