Textkit Logo

pronun.: cal(l)idus

Here you can discuss all things Latin. Use this board to ask questions about grammar, discuss learning strategies, get help with a difficult passage of Latin, and more.

Re: pronun.: cal(l)idus

Postby Godmy » Wed Mar 19, 2014 1:26 pm

A.A.I wrote:Godmy, I've seen your thread and I have all of the books in the series (that I'm aware of). My learning of Latin only started recently but the desire to do so, and collection of resources in perparation for it, have not.

Great! At least you can train the vowel lengths/stresses this way maybe somewhat more easily...

Your pronunciation seems very polished to me, however, as you're likely aware of yourself, it still has that pedagogical feel.

Well, any attempt to sound more as a "Roman from the street" will inevitably deviate from what we know and we imagine to know... I don't say there are not certain things one can do in his pronunciation to sound even more "Latin" and which I do not do (I know), but there is also something you can call a "safe border line". Anyway, I will see how I will progress with these recordings into the future.

That's probably what you're going for but I've progressed to a stage where the overal phonology and not the individual phonetics are the issue (apart from some work needed on long consonants ;)).

I may be trying partially to encompass both... but let's discuss the particular things in details (well, as you do yourself in the rest of your post :) )

Elison, stress patterns, prosody, etc: It's difficult stuff because we have fewer ways for reconstructing that side of things.

Indeed, but we can do a great deal just knowing with a kind of certainty (which we have) the few facts about prosody we know: that the stress (or the one we most agree on) should be dynamic (= so we may look at Italian/Spanish for some inspiration...well in Italian one has to be aware of the stress lengthening that is absent in Latin, true), that we know where the main word stress is (though we do know how strong can an initial stress be in the short words (e.g.) [I believe that quite weak probably], and that we know with some kind of certainty the vowel phonemes... <- all these things make a big portion of what you call 'a native accent', but of course we still know only the basic outlines (for some fictious Roman educated speaker) and do not know well all the potential allophones or diphones... or are very uncertain about.

About the elisions, nasalizations: I'm still considering it, but ... maybe a two types of recordings would be good to have. One with nearly none nasalization (and fully pronounced "m/n"s) and "kw" and one recording with an attempt to nasalize where it is possible and labialize my 'q' velars... elide what I can elide. Etc.

Anyway what I perform is an example of an "emphatic" speech. I think that lot of those things you observe in Plautus/Terence and in poetry wouldn't be heard in emphatic or careful speech (also probably the speeches in senate) and it wouldn't be **that** much different than what we imitate.

Just working out the individual phonemes must have been a huge task, even with the information we have.

Indeed. If you look, you see that I've been content so far with the vowel inventory of my mother tongue (not because I would think it is better, but because the individual long and short vowel phonemes in my language differ really only in subtle details). But you are right of course... (and I will see what I'm about to do in the future)

Still, it has to come together in some form, in my own head. It's also about not simply allowing English phonetics and orthographic conventions (and my own diaglossic relationship with them) to heavily influence my own Latin.

Indeed. Try to read aloud a lot, it compensates partially for a listening you would have in a real living language (I can say from my own experience)... and well, these recordings can be ultimately also of some help :P At least I think that I'm quite more strict in certain points than H. Oerberg in his original recordings (Like: I don't pronounce 'imperium' with stressed long vowel... the vowel must stay short in all circumstances.. and he's a bit 'lax' about this, breaking the rule :P )

I also don't bother with the 'qu' as a labiovelar stop. Although Vox Latina describes it, it doesn't seem to recommend it.

I may disagree: it seems from V.L. that there is 'almost' no controversy and quite an agreement between the scholars that it should be a labiovelar stop and also all the phonetic transcriptions (on Wiktionary let's say) write it like that. What bothers me however that next to these transcriptions there is always a a kind of audio recording where the speaker always pronounces a sequence of two sounds, like ignoring it. In fact, all the Latin world I know seems to ignore it in the pronunciation.... so I do it too, and it is weird.

But it is true that it is hard to say how would such "k" sound, whether the following vowel would be also affected by the velarization (=rounding) or not... It's just: controversial and right now I don't think I would imitate it sufficiently well.

Also, I don't bother with the different types of L. On the other hand, I do include the final nasal and like to assimilate and elide where I have seen evidence that it occured. I'm still learning more about these processes.

I forgot to mention "L"'s. You are right that one should be more strict about it. I think I'm not (and I should write it therefore to my descriptions), but I also don't think that I always pronounce a clear "L" and that I have some allophones of my own which **might** even be close to what they recommend. But thank you for telling me that I should note this in the description.

You can at least listen to proper long vowels in it ;) (And I do them natively)... of course with the tiny deviation in my vowel inventory, but compare it yourself that it is tiny :)
Last edited by Godmy on Wed Mar 19, 2014 9:26 pm, edited 8 times in total.
Latin IRC chat: http://textkit.com/greek-latin-forum/vi ... =3&t=62017
User avatar
Textkit Member
Posts: 104
Joined: Wed Jun 12, 2013 4:47 pm
Location: Czech Republic

Re: pronun.: cal(l)idus

Postby Qimmik » Wed Mar 19, 2014 3:00 pm

Godmy, if I'm not mistaken, is a native speaker of Czech, a language with distinctive vowel quantity, and should be able to convey vowel length in a more natural and realistic way than those of us whose native language lacks this feature. We tend to exaggerate long vowels, but the distinction is very subtle and is barely perceptible by non-native speakers without substantial training and exposure to native speakers.
Textkit Zealot
Posts: 2090
Joined: Mon Mar 18, 2013 10:15 pm

Re: pronun.: cal(l)idus

Postby Paul Derouda » Wed Mar 19, 2014 7:10 pm

Like somebody said, the "problem" with Italian seems to be that in a minimal pair, a short consonant makes the preceding vowel longer than with the corresponding long consonant (e.g. polo/pollo).

My native language Finnish doesn't have this problem, since in Finnish both consonant length and vowel length are distinctive, even in unstressed syllables. The stress is always on the first syllable, so stress doesn't interfere with the vowel or consonant length (much).

Since this is the sort of thing many people with other linguistic backgrounds have trouble with, I made a little Youtube recording to illustrate the minimal pairs a/a: and k/kk. I read the same words first slowly and distinctly (well, more or less ;) ) and then faster, rather like I would in a real conversation.


The words are: taakka -- takka -- taakkaa -- takkaa -- takaa

(Meanings: burden -- hearth -- burden (partitive) -- hearth (partitive) -- from behind.)
User avatar
Paul Derouda
Global Moderator
Posts: 1938
Joined: Fri Nov 05, 2010 9:39 pm

Re: pronun.: cal(l)idus

Postby A.A.I » Thu Mar 20, 2014 10:14 pm

ooh, further responses!

Just woke up, replying now wouldn't make much sense.

(EDIT: After up fully, I realised how much there was to do... Trying againt today!)
Textkit Neophyte
Posts: 52
Joined: Tue Nov 12, 2013 5:25 am

Re: pronun.: cal(l)idus

Postby A.A.I » Mon Mar 24, 2014 12:55 am

A few days with patchy internet... grr...


Thank you for taking the time to reply to my long post! Sorry for not having learned how to use the forum properly.

- At the moment, I'm feeling pretty confident with vowel length. Working that all in on the sentence level is something I'm experimenting with. Knowing for sure if they are indeed short or long and where the stress is placed can be a trouble.

- I agree. The system I am going for, is for a small group. We'll have a "safe" pronunciation which we will use with others, if we happen to come across any Latin speakers in the wild. haha

- Details are good.

- I am always looking for more information of stress patterns. Some books on developments within the Romance language group are helping me learn some of the changes. Still, so far it's been hard to glean much of use.

- I read almost everything aloud.

- Maybe I'll learn how to do the labiovelar stop for reading poetry. The rounding of the next vowel got me too. Some more research needed... I also wonder about 'gu'

- I seem to have both of the L sounds described but in different contexts. Lucky me, haha

Paul Derouda.

I was thinking about but hadn't mentioned Finnish. Thanks very much for that recording!
Textkit Neophyte
Posts: 52
Joined: Tue Nov 12, 2013 5:25 am

Re: pronun.: cal(l)idus

Postby Godmy » Wed Mar 26, 2014 2:45 am

Qimmik: You are right. In Czech we can have easily 4/5 long vowels next to each other in one word and still we somehow feel them and need them there (an example: splácávává (4 long vowels) or splácávávání (5 long vowels) <- notice that in the former it's the same vowel repeated four times and four times with a double length).
And a transgression not doing that is always noticed... so I understand how important it is to get this right in a foreign language.

Sometimes we have minimal pairs where the length determines the case as it does in Latin in the fourth declension (and we never get this wrong), an example: hradu(castellō/arcī/arce dat.+loc.) vs. hradů(castellōrum/arcum)
(By the way: our "H" is always voiced. You won't see that often in languages unless in allophones and it makes it pretty distinctive from our pure velar fricative we write down as "ch" (chodit-ambulāre vs. hodit-iacere)... it's because it's evolved the original slavic "g"; we later readopted "g" again ind Latin and Greek words)

- Also we have diphthongs and we are quite sensitive not mixing diphthongs with long vowels... an often way in English to convey the French long "e" (as in café) is definitely a diphthong for us instead (and it becomes even more audible when such learners project it to a foreign language).

And we take the concept of vowel length even to deeper 'lengths' than for example the neighbouring Slovak language does, which also has a phonemic vowel length like we do but with one difference stemming from this 'phonotactic' rule in Slovak: they cannot have in a word two or more long vowels next to each other (with a diphthong counting also as a long vowel). So when that happens they usually shorten the other one (and some of them do it noticeably when trying to speak Czech) - we never do that. We pronounce all the long vowels in a word in their full realization.

Still a Czech learner of Latin will naturally also incline to this rule I have described in Slovak to save their energy when pronouncing and to make it more pleasant to their ears -> they will safely pronounce usually only one of the lengths in Latin word: it takes really a diligent and rigorous study even for us not to be seduced by this but to really pronounce the words like ablātīvus or accūsātīvus or nārrābāmus as they should be with all the lengths (+ paying the attention to the word stress).

But once we commit ourselves to do that correctly, it becomes quickly natural.

This is a vowel diagram of Czech:

If you compare it with the graph in Vox Latina in the first page dealing with vowels, you will find out that we take the concept of long vowels in more 'seriously' regarding to vowel qualities: only one of our vowels (as in English) differs distinctively in its vowel quality when lengthened (at least in the standard/official dialect and that is i/ī). In Latin the long vowels (unless in a/ā) usually are more closed (more high) than their short counterparts. On the other hand it makes the little discrepancy between my and Latin's reconstructed vowel inventory (but a tiny discrepancy :) ).

A.A.I.: thank you for response, I think I will answer later :)
Latin IRC chat: http://textkit.com/greek-latin-forum/vi ... =3&t=62017
User avatar
Textkit Member
Posts: 104
Joined: Wed Jun 12, 2013 4:47 pm
Location: Czech Republic


Return to Learning Latin