Latin's Word Accent: Stress or Pitch? (Discussion Moved from True or False Thread)

Bitmap

Civis Illustris
Look here.

This is strictly-speaking true. But I was trying to be nuanced. Syncope is strongly associated with stress-accent languages, to the extent that Adams (Social Variation and the Latin Language, p90) straight out says 'Syncope is the loss of short unaccented vowels in a language with a stress accent.' The fact that there are roughly three common examples in Greek, compared to hundreds of examples in Latin, is a good indication that Latin has a stress-accent.
You could call it hundreds of examples in Latin if you counted the individual instances (I suppose by the same strategy I could increase the count for Greek as well), but if you look at it on a general level, you will see that syncopes in Latin follow a relatively narrow set of rules. Intervocalic -v- can get dropped and there are a few instances of dropping short u (or i) as in tegmen, saeclum or postum.

The words used by the grammarians do indeed imply a pitch accent, and suggest a distinction between acute and circumflex.

Most linguists nowadays, as far as I know, think that this is a misunderstanding based on applying Greek terminology to Latin. The terms the grammarians use are direct translations from the Greek: acūta (ὀξεῖα), accentus (προσῳδία), (circum)flexa (περισπωμένη), gravis (βαρεῖα). This is pretty much what happened in the rest of Latin grammatical terminology, and is paralleled elsewhere when the terminology used to describe one language is transferred to another. cf. pp282-8 of the Latin Grammarians book, which for me at least shows up in the preview.
Well, that assumption would make sense if Quintilian had differentiated between the different kinds of accents by using vel, i.e. by basically saying 'it doesn't matter much, does it?'.
But he doesn't.

As is frequently brought up nowadays, it is very rare for an accent to be wholly pitch-based or wholly stress-based; usually there is an element of both involved - although we can still meaningfully call eg. English a stress accent language. So it's probable that the Latin accent included some element of pitch, which eased the adoption of the Greek terminology, whilst still being a stress accent.
That's a different field of discussion, but I don't see how standard English (both in America and in the UK) or standard German are not wholly stress-based. Some dialects of them may have pitch-elements, but certainly not the standard language.

In any case, your argument is a non sequitur. You claim that there are no languages that fully rely on a pitch OR a stress accent (which I won't even deny), and THEREFORE Latin had to be a stress-timed language?

Has it ever occured to you that if that were true Latin would probably be the only language in the history of the world that wrote poetry (and even employed very strict rules for prose) AGAINST the way it was actually pronounced?

By the time Varro or Quintilian was writing, it is quite probable (from Greek linguistics) that the pitch accent had been mainly lost in Greek, replaced by a stress accent. So if they were to read the account of a Greek grammarian and listen to Greek being spoken by native speakers, then when they came to apply the same terms to describe Latin, they would naturally use them to describe the Latin stress accent much as they heard them used to describe the Greek stress accent.
That argument works on the assumption that both Quintilian and Varro A) lived in a time where ALL dialects of Greek had lost their pitch accent, B) even if A applies, they were completely oblivious to the fact that Greek (or at least some dialects of it) must have had a pitch accent once and C) even if A and B were true, it never occurred to them that any classical Greek poetry would not work under a stress accent.
Their statements make a lot more sense under the hypothesis that Latin simply had a pitch accent.

Btw., I never denied that Latin didn't have any elements of stress. I think I stated that above. My argument was simply that any such stress was far less relevant than people nowadays seem to think.

Good luck pronouncing Latin with a pitch accent...
I've come to accept the fact that I, for the lack of any native speakers of classical Latin, may at best produce a rather barbaric imitation of what it may have sounded like. How does that undermine anything of what I said?

I'm not claiming that Latin was a stress-timed language, I'm claiming that it had a stress accent. I would guess that Latin was probably syllable-timed or mora-timed (cf. Czech).
I'm claiming that it didn't.

So you think that the correlation in the last part of the hexameter in Augustan poetry is a chance coincidence? If so, why is it more true for later poets than in Ennius, and why specifically in the last two feet of the line?
No. I think I've mentioned before (probably in the Every-time thread, so it's rather lost :p) that I was under the impression that it was moving towards a state in which the stress would be more important in Augustean times ... but at the same time, it didn't quite seem to be there, yet.

I don't understand what you mean here. Perhaps restate it in different words?
If you finish an iambic line in a 2-syllale word, the last ictus is bound not to meet any stress-timed pronunciation you'd assume. In most cases, the 5th ictus wouldn't meet that requirement, either.

Why should it, when the influence of Greek culture was so strong? We know very little about 'native' Latin poetry; as Pacifica said, it was suggested at one time that Saturnians were based on accent.
Latin poetry was modelled after Greek poetry to a large extent, but it wasn't a blind copy of it. If Latin poetry hadn't been able to work under the prerequisites of Greek metre (which it duely copied), it would at the very least have been very unlikely to copy it for centuries.

Considering that the stress accent used to be on the first syllable in the prehistoric period, the Classical Latin accent must be 'relatively' recent, though, making this less plausible.
The way Latin was pronounced in archaic times has no bearing on the pronunciation of classical Latin. I don't pronounce German the same way my ancestors who went to invade England did, either.

In the last two feet, yes. But I think Sihler's idea is that Ennius' poetry has a greater tendency to overall agreement between ictus and stress accent, whereas Augustan poetry intentionally avoids coincidence of ictus and stress accent in the first four feet, whilst encouraging it in the last two.

As I said, Sihler is not reliable, and doesn't cite sources. I suspect this might come from the statistical study of Sturtevant (1919). Even Ennius (p379) exhibits some tendency towards agreement in the last two feet.
I think his argument is pretty weak (and, to be frank, I think his observation is wrong). But I also think it doesn't contribute much to this debate (neither on your nor on my behalf).

If you still want me to write in greater detail about why I think his argument is weak, tell me.

The role of stress in iambic shortening is complicated. There's a very short article here which has so much linguistic terminology in it that you'll undoubtably dismiss it without consideration (to be fair, I don't fully understand it, so it may go either way). But there is some relationship: stressed syllables (according to the rule) are not shortened.

Is there any language with a (dominant) pitch accent in which stressed syllables (i.e. syllables receiving a special pitch) are shortened?
 

Godmy

Sīmia Illustris
...language in the history of the world that wrote poetry (and even employed very strict rules for prose) AGAINST the way it was actually pronounced?
I always believed it was so. (that idea didn't give me much of a pause - it does to you? [talking poetry]) Romans were just cultural copy-cats, so in love with the Greek models that they would do anything to apply the same methods. It reminds me of some Roman 'historical linguist' that had a neat theory* that Latin was in fact just a very remote [long-lost] accent of Greek (historically).

Now, I don't know if "the only language", but I frankly don't care so much nor do I think Romans did ;) After all, in the ancient times, only a handful of civilizations had so developed writing culture, therefore it was easy to deviate from all the others [in such a small company].


*hypothesis
 

Godmy

Sīmia Illustris
I mean, I don't even think anyone can take any Roman grammarians seriously. I don't say they were writers of fiction, but I say they need to be heavily interpreted like Pythia* :)



*meaning: they do bear an important testimony and they are used in the modern historical lingistics (Allen uses them, certainly Sihler too), but they are usually useful only after heavy "interpretation".
 

Bitmap

Civis Illustris
I always believed it was so. (that idea didn't give me much of a pause - it does to you?)
Yes.

Romans were just cultural copy-cats, so in love with the Greek models that they would do anything to apply the same methods.

And the kept doing so for centuries? Could we maybe operate under hypotheses that are likely?
 

Godmy

Sīmia Illustris
Why not?? It's the same thing why one civilization chooses to freeze some version of the standard dialect and use it consciously in WRITING ONLY for several centuries to come. Egyptian did that (Egyptians chose themselves "classical Egyptian", Latin did that too...) People are just damn conservative. That's all.
 

Etaoin Shrdlu

μεσσηγυδορποχέστης
Czech has been keeping alive adjectival endings (both in writing and pronunciation) and even a whole long vowel [in pronunciation) artificially for 3-4 centuries now.
I don't understand this, probably because I don't know anything about Czech. I can see how authorities could enshrine an official written version of a language that had little or nothing to do with how it was spoken, but if people are pronouncing the adjectival endings when they talk, how can this be regarded as artificial?
 

Bitmap

Civis Illustris
There have been Latin writers of epic poetry that went against their predecessors in terms of content and style ... like Lucan being pretty much of a counter-Vergil.

You would have thought that the first and most obvious thing to do would – at least if by his time Latin had been a stress-based language – be to create a different metre than Vergil did. But he didn't. That was the very thing he sticked to.

You would also expect Seneca minor, who breaks with all of Cicero's style, to also break with his prose clausulae. But that one element happens to be the very thing he didn't break with.

Why? Because classical Latin never had the stress-based accent that you think it had.
 

Godmy

Sīmia Illustris
I also wanted to add: and we must not forget that Romans had no "universities" of their own (ehm, I mean, the closest thing to a "university" in the ancient world, no matter how far and how different from what has been developed in the west a millennium later), if a man wanted to receive our counterpart of the "ternary education" they still had to go and wanted to go to Greece and they had to deal somehow with things taught there. The Greek 'linguists' (grammarians) would hardly give much nuance for the sake of their less important Roman students. What they said, was sacred (unless there was another teacher who would preach otherwise).

I don't understand this, probably because I don't know anything about Czech. I can see how authorities could enshrine an official written version of a language that had little or nothing to do with how it was spoken, but if people are pronouncing the adjectival endings when they talk, how can this be regarded as artificial?
They are artificial because noone really feels that they are natural, we are forced into them. It's some kind of "highest" register we learn to use in the educational system. But what I mean is, that those vowels and those adjectival endings were natural some time in the 17th century, then they died out, then were resurrected both in writing and pronunciation, but never came back to the "natural pairlance" so to say... I'm not sure if I'm clear enough, ask further questions if needed, I'll get back to it. (the artificial vowel, btw, is "é")
 

Godmy

Sīmia Illustris
There have been Latin writers of epic poetry that went against their predecessors in terms of content and style ... like Lucan being pretty much of a counter-Vergil.

You would have thought that the first and most obvious thing to do would – at least if by his time Latin had been a stress-based language – be to create a different metre than Vergil did. But he didn't. That was the very thing he sticked to.

You would also expect Seneca minor, who breaks with all of Cicero's style, to also break with his prose clausulae. But that one element happens to be the very thing he didn't break with.

Why? Because classical Latin never had the stress-based accent that you think it had.
Maybe, and maybe you give those silver-age writers too much credit in their free thinking. But an interesting discussion!
 

Bitmap

Civis Illustris
I also wanted to add: and we must not forget that Romans had no "universities" of their own (ehm, I mean, the closest thing to a "university" in the ancient world, no matter how far and how different from what has been developed in the west a millennium later), if a man wanted to receive our counterpart of the "ternary education" they still had to go and wanted to go to Greece and they had to deal somehow with things taught there. The Greek 'linguists' (grammarians) would hardly give much nuance for the sake of their less important Roman students. What they said, was sacred (unless there was another teacher who would preach otherwise).

You're making the same argument as Iason made above (although I have to say you're making it in a slightly different way): Roman grammarians were completely obvlious to the way Greek may have worked in the past (or may still have worked in some dialects) and therefore completely mixsattributed Greek terminology to their own language ... to the extent that they made a sharp distinction between different kinds of pitch accents when, in your terms, there was none.
 

Bitmap

Civis Illustris
Btw. ... I have nothing to gain from this dispute ... I'm just pointing out the things that don't make sense to me.
 

Godmy

Sīmia Illustris
You're making the same argument as Iason made above (although I have to say you're making it in a slightly different way): Roman grammarians were completely obvlious to the way Greek may have worked in the past (or may still have worked in some dialects) and therefore completely misattributed Greek terminology to their own language ... to the extent that they made a sharp distinction between different kinds of pitch accents when, in your terms, there was none.
What I meant was that this is how they 'interpreted' it from their Greek classes with a Greek teacher in Greece, the classes they had undergone as students. That is, a typical Roman scholar would study in Greece under one or more teachers (depending on the subjects he was interested in (maybe also the finances)) and, naturally, in Greek. So their classes of "grammar" would apply to the current of version of Greek (more or less) and nobody would really give a damn about Latin during the "educational process", then [after the fact] only some of the 'intellectual rebells' and 'patriots' would go and write long treatises in their mother tongue (Latin) about the things they actually learnt abroad and not for their own language. Why would these Hellenified scholars care too much whether there was some deep linguistic difference between the two languages and that perhaps the words of their teacher weren't a universal sacred law after all? I don't know, I just don't give them [these Romans] as much credit. These things about Latin might have just as well been written by Greeks [but in Latin] and there would have been little difference, as I see it, with the way the high education was conducted

Anyway, the restituted model for all practical purposes chooses the leaning to a stress-based pronunciation, I know of some public exceptions as Evan der(?) Millner/Molendinarius who in his countless (literally hundreds, thousands) public Latin lessons for beginners religiously adheres to a pitch accent (and three accents) in the Latin pronunciation, afaik, but he is rather felt as a black sheep in that regard. (not saying it means anything, I'm just saying how I feel the practical reality is in the pronunciation circles...)

In Vox Graeca, I may recall some passage that said that even Greek might have had some/unknown stress based accent on the top of the pitch one, but probably not too important or just too weak for anybody to care and, in such case, it might not have even coincided with the pitch accent position (like a fourth accent). But certainly, the pitch vs. stress is rather a spectrum than a binary distinction when it comes to languages in general...

I may be losing the thread here (I also haven't read the whole discussion from the beginning), so I'm just loosely giving here thoughts, so in case it could seem to you that I'm constructing some master argument that makes no sense, then... I'm not, I'm just... writing whatever comes to mind on the given topic. So, my apologies in that regard ; P
 

Dantius

Homo Sapiens
Staff member
There have been Latin writers of epic poetry that went against their predecessors in terms of content and style ... like Lucan being pretty much of a counter-Vergil.

You would have thought that the first and most obvious thing to do would – at least if by his time Latin had been a stress-based language – be to create a different metre than Vergil did. But he didn't. That was the very thing he sticked to
I don't think this is much of an argument; to me it seems much more impactful for Lucan to use the meter of epic if he's going to invert its tropes otherwise (just as satire also uses hexameter rather than its own meter).
 

Iáson

Cívis Illústris
I think this discussion merits its own thread by now... something like 'Did Classical Latin have a stress accent?'
Well, yeah, we do have stress (hence why I said "almost") but to a rather limited extent. A dispassionate French sentence will sound roughly evenly unstressed throughout, with only the last syllable before every pause receiving a slight stress. We aren't constantly hammering like in English, so to speak.

Interesting: thank you for disabusing me of that false belief. So French has no lexical stress at all (which I guess was what Bitmap was suggesting for Latin before accepting the notion of a (pitch) accent).

You could call it hundreds of examples in Latin if you counted the individual instances (I suppose by the same strategy I could increase the count for Greek as well), but if you look at it on a general level, you will see that syncopes in Latin follow a relatively narrow set of rules. Intervocalic -v- can get dropped and there are a few instances of dropping short u (or i) as in tegmen, saeclum or postum.
My impression at least is that it is an order of magnitude more common in Latin. Consider suprā, suprēmus vs. superus, īnfrā, valdē, caldus, soldus, surgō... and from the Eunus documents I linked to earlier it was probably much more common in everyday speech: Alxadrīnī, monocpī, prīculō, redtūrum, optumm. What's the common factor in the examples? They never delete an accented syllable.

Well, that assumption would make sense if Quintilian had differentiated between the different kinds of accents by using vel, i.e. by basically saying 'it doesn't matter much, does it?'.
But he doesn't.
I'm not arguing that Quintilian didn't believe that there was a difference between acute and circumflex, so for my purposes it is irrelevant whether vel is used.

That's a different field of discussion, but I don't see how standard English (both in America and in the UK) or standard German are not wholly stress-based. Some dialects of them may have pitch-elements, but certainly not the standard language.
What do you think 'stress' is? Unlike pitch, 'stress' is not definable in itself; it's phonetically a product of other factors, which include duration, volume, and pitch. In standard English, all these are used as cues to whether a syllable is stressed or not (for more on this, see this document, under 'Perceptual Cues to Word Stress in English'). But this does not mean that we can't contrast English stress with Japanese pitch accent.

In any case, your argument is a non sequitur. You claim that there are no languages that fully rely on a pitch OR a stress accent (which I won't even deny), and THEREFORE Latin had to be a stress-timed language?
No. Firstly, I'm expressly not stating that Latin is stress-timed, I'm saying it has a stress accent.
Secondly, my argument is:

Languages with a stress accent often include an element of pitch in stress.
If Latin had a stress accent, word stress may have thus included an element of pitch.
If this is so, Latin speakers listening to a pitch accent language may plausibly have heard the pitch accent as being like their stress accent.
This explains why, if Latin was a stress accent language, it adopted the terminology of the Greek pitch accent to describe its accent.
Consequently, that Latin grammarians adopted the terminology of a pitch accent to describe their own language is not as implausible as it might seem to you.

One might add that Greeks continued to use the same terminology to describe their accent even after it became a stress accent.

Has it ever occured to you that if that were true Latin would probably be the only language in the history of the world that wrote poetry (and even employed very strict rules for prose) AGAINST the way it was actually pronounced?
This argument seems invalid. Why is it working 'AGAINST' the language to use one aspect of its phonetics rather than another to create poetry? That's like saying that alliterative verse, or rhyming verse without a metre, in English would also be 'AGAINST the way it was actually pronounced'. People still write quantitative verse in English (which incidentally I rather like), or syllable-counting verse (haikus and so on).

That argument works on the assumption that both Quintilian and Varro A) lived in a time where ALL dialects of Greek had lost their pitch accent, B) even if A applies, they were completely oblivious to the fact that Greek (or at least some dialects of it) must have had a pitch accent once and C) even if A and B were true, it never occurred to them that any classical Greek poetry would not work under a stress accent.
Their statements make a lot more sense under the hypothesis that Latin simply had a pitch accent.
No, it doesn't. A) Some speakers of Greek could have still had a pitch accent, and the speakers with the stress accent would have undoubtably interpreted this as those speakers just pronouncing the accent in a different way. If the Greek accent was realised both as a pitch accent and a stress accent by different speakers - but described by Greek grammarians in terms of the original pitch accent - it would in fact make it even more likely that the Roman grammarians would adopt the same terms to describe a stress accent.
B) I don't think they would have distinguished 'pitch accent' and 'stress accent'. No ancient grammarian that I know of ever refers to such a difference, nor is there any reason why they should.
C) Why wouldn't it 'work'? The accent is immaterial to Greek quantitative verse. A bigger problem in reading Greek poetry was that the quantities were disappearing - but this didn't stop people 'pretending' that they were still there and continuing to produce quantitative poetry.

Btw., I never denied that Latin didn't have any elements of stress. I think I stated that above. My argument was simply that any such stress was far less relevant than people nowadays seem to think.
'less relevant' ... what does that mean? Was it there or wasn't it? I thought that your original point was that there was no stress accent, and when reading verse the verse ictus was given stress.


I've come to accept the fact that I, for the lack of any native speakers of classical Latin, may at best produce a rather barbaric imitation of what it may have sounded like. How does that undermine anything of what I said?
Nothing, I was just making a sarcastic comment that wasn't part of any argument.

I'm claiming that it didn't.
Good, in that case I have understood you correctly - but why have you then used the phrase 'stress-timed' twice elsewhere in the same response?

No. I think I've mentioned before (probably in the Every-time thread, so it's rather lost :p) that I was under the impression that it was moving towards a state in which the stress would be more important in Augustean times ... but at the same time, it didn't quite seem to be there, yet.
So stress as a factor was moving towards being there but not actually quite being there? That doesn't seem to make sense to me. Surely in (say) Vergil's time there was either a stress accent or a pitch accent. I guess in between the two you could arguably have some sort of intermediate phase in which pitch was the main cue but there was an optional association of intensity or duration etc. ... but I suspect that would also be called a stress accent by a linguist. I guess you could argue that the Latin accent was normally one of stress but that in reading poetry pitch was used to mark out the word accent and intensity to mark the ictus... or vice versa for that matter...

If you finish an iambic line in a 2-syllale word, the last ictus is bound not to meet any stress-timed pronunciation you'd assume. In most cases, the 5th ictus wouldn't meet that requirement, either.
So your point is: 'Plautus and Terence often end an iambic line with a two syllable word; this makes it impossible to have coincidence of ictus and lexical accent; thus they can't have been trying to ensure coincidence of ictus and lexical accent.' ? Surely this doesn't rule out an overall statistical preference for coincidence of ictus and lexical accent.

Latin poetry was modelled after Greek poetry to a large extent, but it wasn't a blind copy of it. If Latin poetry hadn't been able to work under the prerequisites of Greek metre (which it duely copied), it would at the very least have been very unlikely to copy it for centuries.
As above, I don't really see why quantitative metre shouldn't 'work' in a language with a stress accent.

The way Latin was pronounced in archaic times has no bearing on the pronunciation of classical Latin. I don't pronounce German the same way my ancestors who went to invade England did, either.
You miss my point (which isn't directed against your argument). If Saturnian verse is stress based (which nobody really believes any more anyway) and it is truly ancient, it presumably goes back to when Latin had a word-initial stress accent, before it had the Classical accent (whatever that might be): thus it's not a good guide to the nature of the Classical accent anyway. So even if Saturnian verse is stress based, it says nothing about the accent in the Classical period.

I think his argument is pretty weak (and, to be frank, I think his observation is wrong). But I also think it doesn't contribute much to this debate (neither on your nor on my behalf).
If you still want me to write in greater detail about why I think his argument is weak, tell me.
Sihler or Sturtevant (Sihler himself doesn't really make an argument)? The question of whether Latin poets intentionally tried to produce coincidence/clash of word accent and ictus is surely relevant to whether word accent existed, although less relevant to whether it was a stress vs. pitch accent.

Is there any language with a (dominant) pitch accent in which stressed syllables (i.e. syllables receiving a special pitch) are shortened?
I don't know. But I guess you could argue that a pitch accent might also block shortening; my argument was directed against the idea that there was no word accent in Latin, rather than that this was not a pitch accent.
 

Bitmap

Civis Illustris
I think this discussion merits its own thread by now... something like 'Did Classical Latin have a stress accent?'

It probably does.

Thank you for the long response. I'm a bit busy right now, I will have to look at this later.
 

Issacus Divus

H₃rḗǵs h₁n̥dʰéri diwsú
Yes, I was going to say that. It needs to break off.
 

Iáson

Cívis Illústris
In Vox Graeca, I may recall some passage that said that even Greek might have had some/unknown stress based accent on the top of the pitch one, but probably not too important or just too weak for anybody to care and, in such case, it might not have even coincided with the pitch accent position (like a fourth accent).
Yes - Allen makes a lot of this in Accent and Rhythm, and even uses it to try and explain some features of Greek metrical technique. I haven't read the book fully, so I don't want to judge for certain: but I think other scholars have suggested that it might be some sort of psycho-acoustic phenomenon, ie. a structure that's there in the way people listen to metre but has no phonetic realisation. In any case, I'm a bit doubtful over the existence of a phonetic feature that is never described by an ancient grammarian, never affects linguistic change, is deduced entirely from abstract analysis of some metrical tendencies, and mysteriously disappears without a trace when the attested accent shifts to a stress accent.

Incidentally, since starting this argument I've changed my mind a little over the practice of reading by stressing the ictus, which I had assumed was some modern invention designed to make ancient poetry readable in the Middle Ages. In this thesis on 'The Christianisation of Latin Metre', in the section entitled 'The evolution of prosodic terminology in late antiquity' (pp187 on), the interpretation offered is that the practice of stressing the ictus began in late antiquity when the vowel quantity distinctions disappeared/became quality distinctions, and that this can be seen in the grammarians' descriptions of ictus and metre. In that case, stressing the ictus would indeed be an ancient practice, though not original to the composition of Classical Latin poetry (and I'm not particularly more partial to it than before). Obviously I guess Bitmap would disagree with the basic assumptions involved in the interpretation.
 
Top