Prosody and syllable stress

Latin Learning Latin
41

Luci, I haven’t listened to your recording but will do so immediately after writing this.
Why do the accents sometimes change? Because we stress words in different ways to convey different meanings. You could equally ask why the same word in a different place or sentence has no pitch accent. These are pitch accents, remember, and typically used for emphasis in a communicative context, rather than in the predictable context of enunciating single-words only.

Is there ever a grave accent placed upon a stressed syllable? I don’t believe that will be likely.

Look again at Lily’s rules and note that stressed syllables can receive a grave stress: fuèrunt, penultimate long by nature, ultimate long by position (in the category of ‘other cases’). I think this is very revealing, because we do pronounce the word this way. It helps to establish that the rules are descriptive before being prescriptive.

I’d be happy to call stress ictus but what’s wrong with just being rigorous about talking about stress and pitch to distinguish these types of accent. I think that might be better because, there’s a strong possibility that the stressed syllable (normally understood as the accented syllable) is retained in an adverb such as ‘sane’ (the first syllable) while receiving a drop in pitch accent on the final syllable. Although, it would be much more interesting if that were not the case.

42

Luci, I listened to your recording. What a beautiful and brilliant exposition of the materials. I found it very difficult, myself, to pronounce, in a convincing fashion, an ultimate circumflected vowel, without resorting to nailing the ultimate with a stress accent. That’s why I thought I had to prove an ablative-‘a’ had to have a primary stress. I suspected that, ultimately, it would come down to showing the possiblity for separation between the pitch accented syllable and the stressed syllable (the máximè behind maximè) but I couldn’t feel the solution, and you’ve just demonstrated one wonderfully. The documentary evidence is there somewhere and I’ve got piles of works to sift. Brilliant.
Adrianus.

I read your suggestion again and I think that’s a very good idea to talk about ictus and accentus. I thought it might be confusing in English but it isn’t in Latin. I’ll start doing that from now. Hang on, it is confusing. Won’t it be difficult when you talk about the same word from the latin grammarian’s point of view, who uses accentus to identify the ictus syllable (applying the penultimate rule, say), when the speaker might apply an accentus to a different syllable in the same word?

43

Apologies to the moderator for carrying on a dialogue with myself but this information was requested earlier:

Frank Abbott (1907), The Accent in Vulgar and Formal Latin?, Classical Philology,Vol. 2 No.4: 444-460 says the grammarians talk mainly about literature, not speech.

Quintilian (fuller quote more interesting): sed accentus quoque cum rigore quodam, tum similitudine ipsa minus suaves habemus, quia ultima syllaba nec acuta unquam excitatur, nec flexa circumducitur, sed in gravem vel duas graves [vox] cadit semper itaque tando est sermo Graecus latino iucundior, ut nostri poetae, quotiens dulce carmen esse voluerunt, illorum id nominibus exornent. [vel duas graves!!]

Passage from Priscian illustrating how the grave could be thought of like a secondary pitch accent (when a ‘primary-stress’ –if elsewhere–stays put on the median level) : "ubi quod interrogativum paenultimam acuit, ut si dicam úbi est Pamphilus? relativum gravatur, ut Vergilius in I Aeneidos: saevus ubì Aeacidae telo iacit Hector, ubì tot Simois correpta sub undis, quomodo et unde, ut Horatius in II sermonum únde et quo Catius? interrogativum acutum paenultimum habet, relativum gravatur; Vergilius in I genus unde Latinum? [I presume genus undè Latinum]

I promise not to speak unless spoken to.

44

You’re hardly speaking out of turn, amice! I’m delighted to read more. I’m also glad you liked my interpretation.

“fuèrunt” makes a lot of sense; I’ve used this accentuation before too. As for Quintillian and Priscian, what then can we say? Are we wrong after all? Are these non grave pitch accents on the final syllables incorrect by Roman standards? I’m not sure how to explain them in the Middle Ages and in the later periods, as you have demonstrated for us, without the Roman præcedent.

A double grave? Is it possible that’s what I was doing in my “ergô”? and others? a sort of drop in tone, up again to neutral, then down again?

45

Care Luce :

Fascinating reading, well done.

This discussion has raised some tangential questions, some of which have been implied by adrianus’s comments. To wit, some random questions and commentary:

What influence did the study of Greek have upon the interpretation of Latin pronunciation in the 16th and 17th centuries ?

What influence did Frankish, Gothic, Saxon, and other “barbaric” tongues have upon the pronunciation of Latin during the early Middle Ages ? Remember, these people were the dominant societies, not the Romans, and I wonder about reciprocal influences.

Following Pound: The use of interlinear Greek/Latin texts diminished by the end of the 19th century. Why ? You can locate many texts in large libraries that indicate such practice was popular and widespread. At UCLA and at the Regenstein in Chicago I found Latin translations of Greek lyric poets, Homer, Hesiod, the dramatists, and even the Tao Te Ching (!), almost all of which dated from before the 1800s. This point is significant re: my first question.

You or adrianus mentioned that some scholars from the 16th/17th centuries studied in the middle East and Greece, which would support certain contentions about Greek, but I fail to see how it mattered deeply to Latin, except as proof of the influence of Greek studies on the pronunciation of Latin in those centuries.

Pitch is certainly present in any language, but I do not see or hear it as a structural determinant in Latin, nor apparently did any of the classical grammarians. Your excellent examples sounded more to me like an Italian speaking Latin with a classical Greek pronunciation. And given the fact that educated Romans spoke Greek, perhaps there’s more to all this than I can tell. I look forward to more of adrianus’s revelations re: Lilly.

For anyone wanting more knowledge re: the pedagogy of Latin in the 16th/17th centuries, check out Pound’s “Early Translators Of Homer” in his Literary Essays. EP’s focus is primarily the Renaissance translators (Valla, Obsepeo, Divus, et al) but he includes Salel and Samuel Clarke in his view.

The earliest Latin verse appears to be insensitive to quantity. It might be argued that all Latin pronunciation is bogus to some extent, since even the Latin authors themselves borrowed their forms and their prosody from the Greek.

I have another question. How are questions intoned in spoken Latin ? What do the grammarians (from any period) tell us ? This instance is one in which there is a pitch variance in almost all languages.

And yet one more: Lucus, I’ll be home all day today. Would you like to connect via Skype this afternoon ? Again, I’m cantator1.

46

Cantator care,
Responses to some of your questions:

  1. The influence of the study of Greek on the interpretation of Latin pronunciation in the 16-17th centuries relates to the way late-humanist scholars can read and interpret (sometimes between the lines) the terminology and intention of classical and late- grammarians of Latin, who were influenced by Greek language and linguistic legacy (and then make their own interpretations and misinterpretations, based on appeal to ancient authority).
  2. Barbarismi in middle ages --I can’t contribute anything.

except as proof of the influence of Greek studies on the pronunciation of Latin in those centuries

–that is the point and it’s very significant, I think (same as point 1). Note, also, authoritative and prescribed texts are one thing, but actual implimentation and speech practice are others, and there are many complicated currents undermining any study or appreciation based solely on school and university textbooks. It’s also guaranteed that any historical interpretation must compromise on conflicting evidence, or even swallow the conflict unresolved,-- if it wants to accurately reflect the confusion inherent in the past (and the present!).
4. On the pitch accent, the literature on this big late-19th-early-20th century debate provides lots of evidence, however you wish to interpret it. Most scholarly books on the Latin, and specifically classical-Roman, accent address this. The debate is relevant, also, to consideration of historical change in the pronunciation of Latin.
5. Information on asking questions in buried in the previous discussion. There’s a nice example in the earlier Priscian quote. Questioning in Latin frequently involves an acute on either the first word or the last syllable of the last word (as well as frequently involving other possibilities-- ‘quà ?’, say). Lots of great examples. Just one, used most in teaching is “quid faciám?” Another I think is just great is in making lists in a subclause in a statement (not a question), where the final syllable in the list strikes acute (‘as in a questioning manner’) as it meets the last comma marking the end of the clause, just as we often affect in English. That’s from a 15th-century source but noted in a 16th-century edition.

Luci care,
‘the double grave’! That’s exactly why I single out this quotation. It’s possible this allusion bears on the utimate circumflex accent, but a circumflex is clearly NOT a double grave, even if it might be made to stand in for one, for lack of something better (and later grammarians had a symbol more logically suited to the double-grave purpose, as I imagine it). To me, everything about this notation is interesting, but the crucial question is about ablative-‘a’, because the pursuit of it has brought to light all the rest.
I think your Italian take on accenting is justified because you are illustrating revealing possibilities, and that is my take on the materials, too. It’s for the listener to leave room in their imagination for national variations, so we can understand contemporary criticism of pronunciation style (‘too Italian’ in England), which, I think, relates not only to alternative vowel sounds but also to the balance struck between stress and pitch (English Latin more evenly pitched than Italian). Because, even though Lilly’s grammar reveals the influence, I think, of one style, the actual practice (how it was generally taught and how it was actually spoken) leaves room for ‘native-English’-latin pronunciation (an alternative reader response than the author’s intentionality).
Adrianus

47

To Cantator, and to Adrianus if he can make it, I will also be on this afternoon (starting now) and I would love to continue discussing these matters. I just wanted to put out this message before responding in detail.

48

I can’t participate because I’m at a meeting for the next few hours but I would love to again. I’ve never used Skype, so that would be interesting. I would have to overcome nervousness about exposing how limited my Latin actually is. You see, I am a learner and not yet graduated from 16th-century Latin for ten-year-olds. I console myself that the standard must has been exceptionally high in that period, and I have escaped the beatings.
Adrianus

P.S. Cantator, if you read the Priscian quote above you will see it as indisputable proof that he (at least) was talking about a pitch accent, at least regarding asking questions: únde and undè. He’s talking music because he cannot be talking stress or attack or volume, since stress or attack or volume cannot differentiate such cases (únde and undè). In fact, I believe I have a good explanation that I’m developing regarding stress vs pitch. Pitch must dominate in the classroom (and that includes the grammarian’s page) where emphasis is so important, but, outside the classroom, and when upperclass Greek trendiness had gone, it’s more stress than pitch. But, even outside the classroom, pitch must remain important in instances requiring emphasis, disambiguation, rhetoric and colour.

49

Luci, you said

Are these non grave pitch accents on the final syllables incorrect by Roman standards?

Not at all (or not necessarily). A defence has been made already above. The problem is we’re repeating selective quotations from the grammarians, who are often being selective themselves. Better to look at the primary sources in Keil,–that’s the thing to do next.

50

No worries! You’ve certainly researched this topic well enough. We’d love to discuss these things with you.

51

Hu care,
Thanks for the compliment about the research. An inability to speak or write good Latin didn’t stop many in the 16th-early-17th centuries from teaching and lecturing authoritatively on the language. They were criticised, of course, by those (in the tradition of Ramus, Erasmus, Comenius, etc.) who put practice before theory, and who called for educational reform. The latter were the smaller camp, very influential in modern educational theory, but whether they won the debate in the context of language learning…
I hope to look at Keil soon.
Thanks
Adrianus

52

I spoke with Lucus last night, we touched on this subject but I was pressed for time. Here’s how I’m currently considering pitch in Latin:

A question, spoken or written, relies upon some kind of marker to indicate its difference from a normal statement. However, this case does not prove anything beyond a general recognition of the phenomenon of pitch in spoken language. To my knowledge there is no systemic practice of pitch application in the Latin language. There was no such thing in the earliest pre-Greek Latin verse, and it seems to have had no systemized pitch intonation in the vulgus. That pitch variation occurs is a fact of emotive emphasis, i.e. the intensity and actual pitch will vary unsystematically from speaker to speaker.

Perhaps I’m missing an assumption. Have Latin instructors been teaching Latin without an appreciation for intonation ? How can this be done ? You have to really work at it to say almost anything without creating a pitch curve. When I read Latin poetry I look for the significant words and phrases and I emphasize them naturally, allowing pitch variance in natural response to what is being read/said. Is this the kind of practice we’re trying to establish here ? If so, I’m all up on that. :slight_smile: But if we’re trying to codify a practice as a system of predictable intonation I think we’re barking (singing?) up the wrong tree. Of course I could be 'way wrong about this, and I’m eagerly awaiting my Vox Latina so I can compare notes with Lord. I also look forward to further evidence from adrianus on this topic.

An exercise for the curious: Speak a two-syllable word. Now accent one of the syllables without altering the pitch. Now alter the pitch without changing the accent. Now change the accent or pitch without altering the volume. Good luck.

These terms (stress, accent, ictus) all imply difference between the phenomena they name, but it might be better to note that like a musical tone, they are in fact parts of a whole, and like the musical tone you can’t really take it apart without destroying the whole. There’s no such thing as an accent without a tone, no such thing as a tone without an accent. In music we can compose and analyze these things as though they are separable from one another, but they are not so in the reality of the sounding tone.

As always, it’s all just my two pfennigs. :slight_smile:

53

Cantator care,
I know we disagree about pitch as a marker in Latin (I think it was used as such, especially so for language instruction and grammatical illustration, and you disagree about any structural implications). We debate.
I agree with you practically 100% on the last paragraph–no speech without stress, accent, ictus. What I mean is volume, pitch and timing (understood as syllable length punctuated by beat). Is that what you mean, also–sort of, by stress, accent, ictus?
But we diverge on

An exercise for the curious: Speak a two-syllable word. Now accent one of the syllables without altering the pitch. Now alter the pitch without changing the accent. Now change the accent or pitch without altering the volume. Good luck.

If I understand what you’re saying in this quote (and you’re distinguishing accent from pitch, where you are using ‘accent’ to mark the syllable that one intends to be outstanding,–that has the beat or ictus, like morse code), I don’t think that’s truly difficult. It’s most exaggerated natural forms are in song (particularly chant), religious rite, and are brought out also, I think, by quickened natural speech. It characterises what we understand as mechanical-sounding speech, or speech without emotion. I’m not so silly as to think I couldn’t be wrong, and my arguments are probable unsophisticated and a bit naive. It’s actually fun not to know who I’m talking to, and be able to make a fool of myself. But I’m not disagreeing to be perverse. Regarding mechanical-sounding speech, in fact, that’s one of the motivations for the work that I’m doing: computer speech synthesis, in which you can precisely control variables,–but you still can and do affect it naturally (the sort of independent control of variables in speech that you’re talking about in the quotation above).
Adrianus
(I might be quiet for a few days because I have some stuff to do, but I want to keep debating until I’ve got nowhere to go but to humbly backtrack.)

54

I got the Keil volumes (Grammatici Latini) and had to look at them right away. It’s not really what I had planned for learning Latin, but its wonderful. At first sight, it seems that Lilly is just straight Priscian (vols. II and III). Priscian talks quite a bit more about accenting the final syllable, and its unambiguous, because he’s clear about his acutes, graves, circumflexes, macrons and his breves which he defines (no way does a macron = circumflex). I’ll put more direct quotes up (no doubt very selective) asap, but still missing a statement about differentiation and ablative-a (but it might be in there somewhere). If you read between the lines, his statement about accenting ‘unâ’ to differentiate it from a noun would seem to obviate the possibility of ablative-â (understood as primary accenting) to differentiate ablative from nominative.

55

Diomedes, Artis Grammaticae, Liber II, De Accentibus
Accentum legem vel distinguendi vel pronuntiandi ratio vel discernendae ambiguitatis necessitas saepe conturbat.
Accentus acuti nota ita / per obliquum ascendens in dexteram partem. Gravis nota ita \ a summo in obliquam dexteram partem descendens. Circumflexi nota de acuto et gravi figuratur, vel c deorsum spectans ^. Longus linea a sinistra in dexteram partem aequaliter ducta —. Brevis virgula similiter iacens, sed panda et contractior, quasi c sursum spectans ˘. Sed in illis sonos, in his tempora dinosci videmus.
I thought an inverted ^might be distinct from an upwards c ( ˘ ) and maybe part of a double grave. From this quote, it clearly isn’t. More likely a double grave (mentioned above) is simply a double drop in voice register. It has to be distinct from a circumflex. Quintilian isn’t in Keil, so Institutiones Oratoriae is needed.

56

I must add that no-one here seems to be saying that it does, I’m just arguing with myself out loud. As I read Lord’s texts I’m beginning to have greater respect for the old grammarians. They recognized pitch as a structural component (yes, I agree that it’s certainly a component in the structure of speech), and they faced the same problem with it as we do. Its application is largely unorganized, yet it adds significance to the literal meaning of the spoken words.

Adrianus, as someone involved in speech synthesis, I’m sure you’ve looked at pitch contours. The diacritical markings from the grammarians were their attempt at notating pitch contours. Could that double grave simply been an attempt to describe a pitch lower than a normal grave, a “doubled low” ? And how many such markings have been applied to Latin ? So far I count acute, grave, circumflex, and this double grave. Are there others ?

I’m rethinking all this, it’s getting more interesting on a practical plane. I’ve been reading Ovid and Propertius, changing some pronunciation and trying some of the accents suggested from these discussions. It’s resulted (sometimes, and maybe only to my ear) in more natural and more expressive recitation. I’ll post some examples soon.

Btw, I’m reading Ovid’s tale of Baucis and Philemon, it’s just fantastic. His story-telling abilities are awesome, and his use of Latin for poetic effect is just incredible. Pound was right, there are things we get from Ovid not found in the extant Greek poetry, poetic values new to the times (and language).

57

Absolutely, Cantator, the diacritics are grammarians’ pitch-contour markings. That’s what this discussion has always been about, and its been providing the evidence for that. And I think also, Cantator, that a double grave is a double drop in voice register, or you say a ‘doubled low’. I can’t agree. though. that the grammarians were approaching pitch as “largely unorganized”. The opposite. They admired pitch pronunciation in the Greek voice and identified those elements in Latin. Some have said they exaggerated, wanting to make the Latin model fit more exactly the models of the Greek linguists; some disagree about the degree of any distortion by the early grammarians. The organised accenting that the grammarians are describing is for the purpose of pronouncing a word ‘correctly’, or noting departures from the rule. It helps to establish WHAT is said. The second level of accenting, however, is more subjective, and I could understand you referring to that ‘sort’ of pitch as ‘unorganized’. It isn’t unorganised, though; if it were, we couldn’t understand what others mean (we sometimes don’t, but that doesn’t mean it’s unorganised). At that second level, accenting is more emphatic and dramatic, and communicates HOW clauses and sentences are meant. An author (the educator, more frequently) adds them only to certain stressed words to define the overall sentence prosody. Naturally, a stressed word gets stressed on its stressed syllable, but the first meaning of stress here (to stress a word) is significantly different from the second meaning of stress (to stress a syllable). Only some words (those whose stress carries emotive meaning) get word-stressed in this system . You have to be careful, also, because later editions use circumflexes consistently to differentiate meaning and not pitch–circumflex = macron in differentiating cases, which is how it has come to enter our modern thinking on this. Not a mistake, just a different use for the symbol. Important to me is when this subtle shift in meaning took place, and I think it happened in the late-17th century.

There are more speech signs, but they need research. It is very illuminating, isn’t it, when you recite with this approach, both (1) reciting the written word with the emphatic signing, and (2) reciting to take account of the differentiating notation from the later period with the correct pitch interpretation. But the jury’s still out on ablative-‘â’. By the way, my involvement in speech synthesis is not scientific but in educational multimedia applications development, and extending speech accessibility tools to language-learning contexts (Latin, in this case, --for fun, but with practical outcomes).

Adrianus

58

We’re in agreement here, I was indeed referring to your second sort. By “unorganized” i mean that there is no predictive system for this kind of accentuation. I’ve read some descriptions that imply that there is a base tone for a recitation from which the accents indicate a departure, but I don’t agree with that, and that’s where my “unorganized” comes from. There’s certainly a starting tone, but I don’t believe it stays unaltered as a base pitch. Engagement and involvement with the text will prompt changes in the base pitch level. The accents retain their relative influence, and the recitation is far more expressive.

How does that line up with your thinking on the topic so far ?

I guess what I’m trying to discern here is whether the grammarians indicated that Latin words always carried this pitch distinction (i.e. as a matter of grammar) or if they simply observed that pitch is an undeniable aspect of speech and they might as well at least try to codify it along with every other aspect. It seems that the older grammarians are less prescriptive about it, and I’m curious as to why it became a hot topic for the (much) later grammarians.

Btw, thanks for your patience with me. I would have got a better idea of your intentions had I considered the pitch contour analogy earlier. I continue my researches. :slight_smile:

59

Cantator care
I think we’re starting to connect on this. I think ‘unpredictable’ is better than ‘unorganised’.
I think where you’re coming from about expressiveness in recitation applies in the modern period. There’s no doubt that a certain, modern audience would understand the expressiveness you’re talking about. I think it’s very different the further back in time you go, and suspect there are differences in our own time between cultures on this. Stylization can dominate in a recital and be extremely satisfying for the audience, as long as they understand the style (the sort of ‘grammar’) of the performance and likely you do when it’s part of your culture. Even listening to early-20th century recitations will convince you of the strict traditions that applied not so long ago, and you understand that voice timbre (richness) + volume change can be used to very powerful effect instead of pitch, albeit with mighty, descriptive pitch variation at times. (I’m remembering from 20 years ago, listening in the London Sound Archives.) To a modern ear listening to such performances, the average note is incredibly monotonous and sounds distinctly hammy, but they represent the art of the period and the word ‘ham’ should be thrown away because of its unfairness to previous dramatic-training practice. (‘Median tone’ is maybe better than ‘base tone or pitch’, although the recitation is often pitched in base or deep tones ‘to add weight’.) So any evidence for a base tone for recitation, as it applies at a particular time and place, shouldn’t be discounted. (This is separate from natural speech, of course.)

My take (or hypothesis), for what it’s worth, is that the early grammarians did believe Latin words always carried this pitch distinction (as a matter of grammar), and that the proof was in the word’s emphasis. You hear the pitch-distinctions better when you emphasize a word than when you speak it naturally, and that comes about because those distinctions are inherent in the word. Pitch distinction may then be applied as the basis for your rules of pronunciation. We see weaknesses, but that doesn’t matter. I once taught a course in Greek and Medieval philosophy of science and that’s a reasonable argument for the time. (There’s no evidence it was articulated, of course.) Incidentally, you can make that argument even if the Latin accent is actually much more stress than pitch and you, as an early grammarian, clearly hear that (or less clearly hear the pitch element). The fact that you (as an early Latin grammarian) are enamoured with Greek philosophy, encourages you to emphasize (in whatever sense) what is, after all, there: the pitch component, which coincides with the stress component.

Remember, also, that whether it’s predominantly pitch or stress is immaterial to the argument about what syllable gets the accent. But I think the stress-pitch thing is very, very interesting (although I thought there was agreement about ‘a mixture of both’ – not so, and that’s a good thing). :slight_smile:
Adrianus

60

I’m interested in those recordings, are they available on-line ?

I have recordings of Yeats reading his own verse, also some Whitman (very old stuff). They are expressive in the modern taste, very likable to my ear

(‘Median tone’ is maybe better than ‘base tone or pitch’, although the recitation is often pitched in base or deep tones ‘to add weight’.)

Yes, median tone is a better descriptive.

I’m off for the weekend. I’ll try to check in for my daily revelation, but I might not post until Sunday or Monday. Ciao for now.