Let us hear and analyse each other's Latin pronunciation

 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
The rules are simple:

  • Upload a recording of you reading some Latin text (the best will be if the recording is not more than 30 seconds long or at least under 1 minute).
    • Upload it please in an MP3, OGG or WMA format or any format with a high compression, i.e other than WAV or FLAC (see below how) - please do not post here WAV or FLAC files, they are too large.
  • After that the others are welcome to write their objective criticism of what has been posted, to give their feedback. (Typically I might be doing that here, if I'm present, but everybody is invited!)
I must declare here that objectivity can be really attained in this discipline more or less only in the restituted / reconstructed pronunciation, since we have phonetically quite an exact description of the thing we reconstructed (i.e. we do not know exactly the real pronunciation spoken in the given time and place of antiquity, but the model we have reconstructed and agreed on is phonetically described with some precision. That allows us to judge others how well they are able to follow it)
  • State which pronunciation it is (we can always guess, but it will be better, if it is stated):
    • Don't worry if you don't use the restituted pronunciation, but for example the ecclesiastic one or some other. Just make sure we know which one you've just used.
    • At this thread there are some functional links to the book which describes how the restituted pronunciation was reconstructed and advises what is the best way to pronounce each part of it.
  • If you wish, state also what is/are your native language(s): for English you can (but you don't have to) also specify where is your native dialect approximately spoken (USA, UK, Australia...)
  • It will be also helpful if you copy the text along with the recording, if the text is not on the same page on this thread, where your post is.
=============================================================

How to make a simple recording with a freeware application called Audacity:
  1. Download and then install Audacity
  2. After launching the program, hit the red dot button to record audacity-record.JPG
  3. Read the text to your microphone and then hit the 'square' stop button audacity-stop.JPG
  4. In the left top part of the screen click on File and then click on Export
    1. Write the name of the file in the Window that appeared (you can include your nickname there and the name of the text), as file format choose WAV and hit SAVE (unless you can export it in MP3 directly, which not everybody can do in default)
    2. Please do not post the file, until it is converted into MP3
  1. To make an MP3 file from the WAV file you have created, download and open this tool for Windows
  2. Once opened, a little window appears. Drag & Drop your newly created WAV file from the folder it is saved in into the little window. An MP3 will be created instantly in the same place where your original file is.
  3. Make sure that you upload the small MP3, not the large WAV file (it takes too much space and longer time for us to download)
=============================================================


- I will re-post my reading of the beginning of G.I.Caesar's Dē Bellō Gallicō. If you feel like it, try to read it too and me with others will then tell you what we think about it.
- I'm also posting one of the old texts we read in the former thread.
- I'm also posting a recording of the first few sentences from H.Oerberg Familia Rōmāna & Rōma Aeterna - try it, if you feel like it :)
- I'm also posting the beginning of the Latin translation of Harry Potter, Harrius Potter
- I'm also posting the first few lines of Ovid's Metamorphoses

- The text contains macrons. If there is a macron above a vowel, it means a long vowel, it there is not, then the vowel must not be -not even accidentally- read long or somehow prolonged inadvertently.
- The bold syllables in the text are those that receive stress/accent.

The rule where to put an accent in a word is as such:
If the second to last syllable from the end is heavy, it will receive stress (accent), if it is not heavy, then the third one will receive stress (whether it is heavy or light) if such syllable exists. Heavy syllable means that the syllable contains a long vowel or the vowel is followed by more than 1 consonant (typically two consonants) (there are exceptions to this rule I won't state). And if it is followed by at least two consonants, it doesn't matter anymore whether the vowel is long or short (and sometimes we don't even know it - sometimes we do), it is simply heavy. The stress/accent itself does not prolong the vowel in the stressed syllable - the vowel stays short when accented in the case, when it is short per sē (i.e. when we don't think it's long).
The text (Caesar, De Bello Galico):

Gallia est omnis dīsa in partēs trēs, quārum ūnam incolunt Belgae, aliam Aquīnī, tertiam quī iprum linguā Celtae, nostrā Gallī appellantur. Hī omnēs linguā, īnstitīs, gibus inter sē differunt. Gallōs ab Aquīnīs Garumna flūmen, ā Belgīs trona et quana vidit. rum omnium fortissimī sunt Belgae, proptereā quod ā culatque hūmānite prōvinciae longissimē absunt, minique ad eōs mercārēs saepe commeant atque ea quae ad effēminandōs animōs pertinent important, proxique sunt Gernīs, quī trāns Rhēnum incolunt, quibuscum continenter bellum gerunt.

Another text (Barba gladiator...):

Deus Inter Hominēs
1. Barba, gladiārum clārissimus, fortissimus, et optimus sum.
2. Anteā capvus eram; nunc rō in gladiāriō dō mē exerceō.
3. “Crās eris in populī Rōoculīs”, inquit lanista meus.
4. “In naumachiīs prope Tiberim aut mare pugbis.
5. Factio tua vincet et populum dēlecbitis
6. Deus es et glōriōsissima ē fornīs tibi erit !“

Another text (Familia Rōmāna):
IMPERIUM RŌNUM
ma in Italiā est. Italia in Eupā est. Graecia in Eupā est. Italia et Graecia in Eupā sunt. Hisnia quoque in Eupā est. Hisnia et Italia et Graecia in Eupā sunt. Aegyptus in Eupā nōn est, Aegyptus in Āfricā est. Gallia nōn in Āfricā est, Gallia est in Eupā. Syria nōn est in Eupā, sed in Asiā. Arabia quoque in Asiā est. Syria et Arabia in Asiā sunt. Gernia nōn in Asiā, sed in Eupā est. Britannia quoque in Eupā est. Gernia et Britannia sunt in Eupā.

Another text (Rōma Aeterna):
Urbs ma in Tiberis flūminis sita est vīginlia passuum ā marī. Hōc loflūmen facile trāntur, et collēs propinquī bene mūpossunt. Moenia Rōna anqua septem collēs ve montēs complectuntur, quōrum haec sunt mina: Patium, Capilium, Avennus, Caelius, Ēsquiliae, Vīmilis, Quirīlis. Ā colle Quirīlī et ā monte Capitōlīnō
usque ad Tiberim flūmen campus Mārtius patet.
Another text (Harrius Potter):
Dominus et Domina Dursley, quī vībant in aedibus Gestātiōnis Ligustrum numerō quattuor sigtīs, nōn sine superbiā dībant sē ratiōne ordiriā vīvenūneque sē paenire ilus ratiōnis. In orbe terrum vix crēdās quemquam esse minus deditum bus novīs et arnīs, quod ineptiās lēs omnō sperbant.

Another text (Ovidius, Metamorphoses):

In nova fert animus / mūtās cere formās
corpora; , coeptīs / (nam vōstāstis et illās)
adspīte meīs / prīqu(e) ab ogine mun
ad mea perpetuum / dēcite tempora carmen!
Ante mar(e) et terrās / et quod tegit omnia caelum
ūnus erat / nārae vultus in orbe,
quemre chaos: / rudis *ingestaque lēs
nec quicquam nisi pondus iners / *congestaqu(e) eōdem
nōn bene jūnctārum / discordia mina rum.
*Late edit: In fact more correctly ingestaque and congestaqu(e). But fortunately my reading doesn't rely on the stress.
(details about this text and its reading: It is a dactylic hexameter. The slash / symbol is a proposed caesura (the pause) by me. The underlined syllable is a beginning of another foot = the ictus, but is given in this reading (which is also what is usually recommended) no special phonetic prominence (i.e. is not read as an additional stress / or the only stress) or given some special emphasis - it serves just a graphical purpose here. Otherwise the syllables in bold are those that are stressed/accented (no bearing on the meter). If a vowel on the end of a word is in brackets (), it is elided by the following vowel and thus not read. )


My native language is Czech (here or here).

My pronunciation is the restituted one with these exceptions:
- my ē and my ō are done simply as prolongations of [ɛ] and [ɔ], that is ː] and [ɔː]. Ideally they should be done a bit higher in the mouth. *Its advantages over other imprecise ways is that they are not mistaken for i/ī or u/ū, they are clearly recognizable and yet they are long & not-diphthongized.
- I don't usually nasalize the -m endings in prose, but read them fully. It seems as an often practice.
- my qu is read mostly as Italians and most of the restituted pronunciation speakers read them, as [kw] and not as a true labiovelar [kʷ].

- it serves just a graphical purpose here
 

Attachments

Iohannes Aurum

Technicus Auxiliarius

  • Technicus Auxiliarius

Location:
Torontum, Ontario, Canada
It has been stickied.
 

Coffea est bona

New Member

Thank you Godmy for your very exhaustive critique. It was extremely helpful to me. Listening to my own recording, I can now hear what you were talking about, gratias tibi ago! And to answer Alatius's question concerning my r's, yes that is how I learned to do them! Di immortales! I have been found out! :)
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
Thank you Godmy for your very exhaustive critique. It was extremely helpful to me. Listening to my own recording, I can now hear what you were talking about, gratias tibi ago! And to answer Alatius's question concerning my r's, yes that is how I learned to do them! Di immortales! I have been found out! :)
I'm glad you found it helpful... it took me a bit longer than I'd anticipated :)
 
 

Matthaeus

Vemortuicida strenuus

  • Civis Illustris

  • Patronus

Location:
Varsovia
Heus Godmy! Quin nos tota alicuis veteris auctoris carmina integra sive capitula legamus, quo saepius linguam Latinam audiamus? Exempli gratia, aliquis nostrum unum De bello gallico Caesaris librum sive nonnulla capitula legat, alius carmen Catulli, ego autem Senecae quendam epistulam et cetera ... Ego audiam tuam Latinitatem tu autem meam! Senatus haec intellegit! :)
 

Ignis Umbra

Ignis Aeternus

  • Civis Illustris

  • Patronus

Location:
USA
Alicuius secunda in sententia tibi scribendum fuit, Matthaee. :)
 

Pacifica

grammaticissima

  • Aedilis

Location:
Belgium
Alicuius secunda in sententia tibi scribendum fuit, Matthaee. :)
Necnon "quandam epistulam". Sed credo amico nostro animum aliquantum incensum itaque digitos tantum lapsos esse. ;)
 

Ignis Umbra

Ignis Aeternus

  • Civis Illustris

  • Patronus

Location:
USA
Ignosco illi. :p
 

Alatius

Civis Illustris

  • Civis Illustris

Location:
Gothoburgi
I must declare here that objectivity can be really attained in this discipline more or less only in the restituted / reconstructed pronunciation, since we have phonetically quite an exact description of the thing we reconstructed (i.e. we do not know exactly the real pronunciation spoken in the given time and place of antiquity, but the model we have reconstructed and agreed on is phonetically described with some precision. That allows us to judge others how well they are able to follow it)
It may also perhaps be useful to consider to what degree each speaker aspires to follow the model. As you say, we know (or at least think we know) quite a lot about the classical pronunciation, but at the same time I think everyone has some mental notion about what details are more important than others, what is fundamentally important and what is unnecessary pedantry. This individual scale of measurement of perfection is of course individual, and influenced by the practices of surrounding Latinists, by pragmatic considerations, by the phonology of the native language, and so on.

For example (strictly hypothetical), when listening to a philologist who has no real interest in reading Latin aloud to begin with, it would seem to me somewhat petulant to openly criticize the pronunciation, only because it does not correspond to my personal ambition. (Not that said philologist would come to this forum thread to begin with, of course, but for the sake of an example.) But at the same time, there are those fundamentals of the language that I expect any Latinist worth the name to be aware of, such as the rule of accentuation: if someone (such as the above mentioned philologist or a student) consistently says servorum and arborum alike, I would correct them, but I wouldn't go into the finer details of vowel qualities (at least as long as the five fundamental vowels are reasonably distinguished).

I say this as an introduction to my response to this explanation of yours:
- my ē and my ō are done simply as prolongations of [ɛ] and [ɔ], that is ː] and [ɔː]. Ideally they should be done a bit higher in the mouth. Most of you will find this way the most feasible to recreate.
- I don't usually nasalize the -m endings in prose, but read them fully. That's an often practice.
- my qu is read mostly as Italians and most of the restituted pronunciation speakers read them, as [kw] and not as a true labiovelar [kʷ].
I am sympathetic with you on your points about -m and qu: in practice you rarely hear nasalized vowels even in the pronunciations of the more famous Latin speakers (or should I say declamators), and neither is the distinction between [kw] and [kʷ] something that most speakers are consciously aware of: in other words, it is quite common to draw the line at that point, so to speak. But, from my point of view, given your manifest deep interest in the finer details of pronunciation, I find your careless attitude to [ɛː] and [ɔː] somewhat surprising, as I perceive your pronunciation of them as an unwanted accent. First, whether they are more "feasible to recreate" depends completely on the native language, and I don't think you can generalize in that way ([e:] and [o:] are hardly uncommon sounds in the languages of Europe, not to speak of the world.) Second, while you acknowledge that it isn't ideal, I find it very strange that you seemingly actively promote it as something to be emulated! (Perhaps I'm reading too much into your comment, but that was my spontaneous reaction.)

If you want some general critique on your pronunciation, apart from the vowel qualities, there is not much negative to say. :) But there might be one general aspect that you may consider, namely your rhythm, specifically how you sometimes pronounce syllables long by position a bit too quickly (at least in my ears). This tendency was more evident in your previous very careful example (where you were occupied with nasalizing and labialization), but there are traces of it in the clip posted here as well. (NOTA(TE) BENE: I use macron and breve here to signify syllable length.) Most of the time, you say them exactly how I would (īntēr sē dīfferūnt for example), but occasionally they feel rushed, as if you are a bit anxious to proceed, and afraid of resting at them: tĕrtiam, hī ŏmnēs, cŭltū, ĭmpōrtānt.
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
Heus Godmy! Quin nos tota alicuis veteris auctoris carmina integra sive capitula legamus, quo saepius linguam Latinam audiamus?...
Nōn malum prōpositum est - at nōnne melius haec in fīlō quōdam aliō facere erit? Hīc enim omnia, quae lecta erint, phōneticē phōnologicēque perscrūtātūrī sumus jūdicātūrīque. Hōc fīlum magis quam litterārium, phōneticum est adque artem phōneticam spectāns :) Nōn autem pessimē dīcis, amīce... creā fīlum tāle, ubi integra opera legantur! :)
Hīc potius brevia legī mālō, quae facile perscūtēmur, sī rem quam dīcō tenēs...

It may also perhaps be useful to consider to what degree each speaker aspires to follow the model. ... what is fundamentally important and what is unnecessary pedantry. This individual scale of measurement of perfection is of course individual, and influenced by the practices of surrounding Latinists, by pragmatic considerations, by the phonology of the native language, and so on. .... it would seem to me somewhat petulant to openly criticize the pronunciation, only because it does not correspond to my personal ambition.
I would like this thread to be for those who are interested in attaining some kind of phonetic or phonological perfection in following the current reconstructed model, which we have currently some phonetical description of. (Of course others pronunciations are welcome too)

I openly state this intention in the premise of this thread, so I expect the people who are not comfortable with this to avoid this place. So, all those situations, which you stated, where you think it would not be appropriate to dissect and criticize things that will be considered as an unnecessary details and some kind malevolence from the side of the one who gives the feedback, are openly not considered an issue in this thread as long as the feedback is based on a sound argumentation. I just have to recommend people who will not be comfortable with this kind of approach, to not participate.

I think that we miss a place where an open discussion about each other's pronunciation not based on personal preferences but based on what we currently think is right, is conducted. This should be the place. This thread will or should be the same as the other pronunciation thread in the last 4-5 pages, the only thing that changes is that here it is sanctioned by the rules of this thread: feedbacks are invited.

First, whether they are more "feasible to recreate" depends completely on the native language, and I don't think you can generalize in that way ([e:] and [o:] are hardly uncommon sounds in the languages of Europe, not to speak of the world.) Second, while you acknowledge that it isn't ideal, I find it very strange that you seemingly actively promote it as something to be emulated! (Perhaps I'm reading too much into your comment, but that was my spontaneous reaction.)
Perhaps I didn't express myself very well and I should rephrase a bit what I wrote, use the time I still have to edit the message of the first post. I thought about that already, I think it is problematic the way I phrased it. Good point. The somewhat 'recommendations' should not be there, I will think of removing it...

However this description of my pronunciation only concerns the pronunciation in the recording I posted. That doesn't mean I will speak like that always. The point is that the uploader of the recording specifies his/her own pronunciation of THAT recording. I did that. I might post a recording next time where I'll use let's say a central-European medieval pronunciation and I will add an altogether different description to that.

About ē and ō:
The ideal way of pronouncing (long) ē and ō as opposed to (short) e and o is 1) to change the vowel quality slightly (to change how the vowel sounds like, where it is created in the mouth) 2) prolong it

There are many erroneous or imprecise ways how to do that (I didn't do it precisely either and I acknowledged that openly). I'll try to enlist a few of them and say something about each to explain why I even thought that one of them might be better than the others:
  • if one diphthongizes the short vowel in order to pronounce e.g. ē, they will change the vowel quality of the overall result in 50%. The other 50% of the resulting sound is adding a strange element that is unexpected. In that case both elements are not prolonged and only the whole unit can be understood as longer relatively to a short monophthong.
  • if one just changes the vowel quality without prolonging it, they risk other people mistaking the resulting ē for i/ī and ō for u/ū (though it should not be the case for a trained phonetician, if it is articulated well)
  • if you don't change the vowel quality but prolong the vowel (my way), you don't risk mistaking the vowel for another and in the same time you produce a long monophthong & not a diphthong.
This is how I think of the way I chose for that file as of "the better one" from the "imprecise ways of articulating these two long vowels". I leave the source vowel and prolong it. But as I said before, Alatius, it only specifies that one recording - I might do that differently one day. The idea was to specify your own recording you have uploaded.

If you're asking me why I'm sticking right now to this imprecise way, it is surely because that's how we naturally do the distinction (along with minimal pairs distinction) in my own language, at least some phoneticians think so. Other phoneticians, Alatius, however think that even as I pronounce it as a Czech, I change the vowel quality of the long vowels and in fact making them slightly more closed, which would also get me close to what is prescribed usually for the restituted pronunciation.
Nevertheless I'm not yet sure that I can articulate ē and ō in such way to consciously distinguish the vowel quality and not to create a confusion with i/ī and u/ū (meaning: I can do it, I know how - I understand the theory, but I think I will not do that yet publicly in my recordings. You can see more in the recordings and discussion I had with Mettius, if you remember).

This is why I stated it as an exception, because I think it is not what the pronunciation model prescribes :) I was honest about it.

If you want some general critique on your pronunciation, apart from the vowel qualities, there is not much negative to say. :)
Thank you :) I appreciate that...

But there might be one general aspect that you may consider, namely your rhythm, specifically how you sometimes pronounce syllables long by position a bit too quickly (at least in my ears).
Hm, but, as far as I know - and correct me please, if I'm wrong, there is no real rule about pronouncing the syllables long by position in a different speed or with a different prominence. There are some non-objective recommendations in several books dealing with poetry and it feels that they are rather some crutches certain people might create for themselves to feel the rhythm better when reading poetry :)

That's at least my opinion, my understanding of the issue. But I don't know about any specific rule that would state that syllables long by position should receive any phonetical prominence than the syllables not long by position.
I know that there are (text)books about Latin poetry which recommend that, but as I say: I'm not sure yet whether these recommendations are not just crutches to facilitate the hearing of the meter. If there is a real rule we have derived e.g. in the reconstruction process, I will be happy to learn about it.

(Another quite notorious case of crutches in poetry is for example to produce the ictus as stress and to eliminate the natural stresses completely.... )

This tendency was more evident in your previous very careful example (where you were occupied with nasalizing and labialization), but there are traces of it in the clip posted here as well. (NOTA(TE) BENE: I use macron and breve here to signify syllable length.) Most of the time, you say them exactly how I would (īntēr sē dīfferūnt for example), but occasionally they feel rushed, as if you are a bit anxious to proceed, and afraid of resting at them: tĕrtiam, hī ŏmnēs, cŭltū, ĭmpōrtānt.
I see. As I say: I would be interested whether this is a real rule or just a recommendation of some philologists that deal with metrical reading. And I think that could create a long discussion on its own.

But thank you again for your feedback, Alatius, I appreciate that.
 

Alatius

Civis Illustris

  • Civis Illustris

Location:
Gothoburgi
Perhaps I didn't express myself very well and I should rephrase a bit what I wrote, use the time I still have to edit the message of the first post. I thought about that already, I think it is problematic the way I phrased it. Good point. The somewhat 'recommendations' should not be there, I will think of removing it...
Very well, then we agree.

Hm, but, as far as I know - and correct me please, if I'm wrong, there is no real rule about pronouncing the syllables long by position in a different speed or with a different prominence.
The thing is that I don't really ask of you to do anything else than what you already do in the vast majority of your syllables. Syllables long by position are those that are closed, i.e. those that end with a consonant, as you are well aware from Allen, pp. 89-92. I agree with you that you shouldn't have to consciously pronounce the closed syllables in a different speed or give them a different prominence: they will become long by themselves. The problem, however, as I see it, is that in a few cases, syllables that should be long by position are not truly closed in your clip. The best example is probably omnēs at 20s into the clip. It should be divided into syllables as om-nēs, but the way you say it there, at least I perceive it more like o-mnēs. Analysing the word with a spectrogram reveals that the individual phonemes are realized more or less with the following lengths (in milliseconds):
o: 80
m: 100
n: 30
ē: 350
s: 190
Thus, you spend roughly half the time on om as you do on ē alone! I think that is too large a difference between two supposedly "long" syllables.

That's at least my opinion, my understanding of the issue. But I don't know about any specific rule that would state that syllables long by position should receive any phonetical prominence than the syllables not long by position.
Well, there obviously must have existed some real, tangible difference in the nature of long and short (or heavy and light) syllables, otherwise "long by position" would be a rather vacuous concept, and we couldn't have the rule governing the place of the stress depend on it. And all sources that I'm aware of, not the least Allen, agree that this difference is an actual difference in duration. The time ratio I measured in omnēs would supposedly be more correct when applied to a word with a muta cum liquida, such as ăprōs, to take a random example.
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
Thank you, Alatius, for the explanation and the measurement, it is most interesting. The thing is whether this is an issue or not. The "o" was short so much as it would be relatively short to that ē in the same word in a living language (e.g. Czech) that has minimal long-short vowel pairs for all its vowels (If I express myself clearly) and the long vowels in the same word must be always "relatively" long to the length of the short vowels in that word (but not unnaturally long). Then there is "ē" which is more than twice longer than "o", however Allen is not so precise as you are, with his recommendations about "roughly twice as long" so in practice this (ratio 1:3, almost 1:4 with a short vowel under 100ms which could be argued to be a limit length where a vowel still can be perceived) would be still roughly in the tolerance - also you would expect the ratio to go down almost to 1:2 as the short vowel would become longer and more emphatic - more perceivable than 80ms (though it would be better to to compare e and ē or o and ō in such a language).

So then the "m" would have to be longer and that's.... a hypothesis. Also what you say next that there "obviously must have existed some real tangible difference in the nature of long and short syllables" doesn't have to mean that the syllables long by position with a short vowel were comparable to open long syllables. That's a hypothesis of some philologists and also recommendations of many books dealing with Latin poetry where the author honestly wants the student to "hear" it at all costs. But it may also be that a long syllable by position with a short vowel will just be "longer" than a short syllable but not necessarily the same length as a long open syllable.
So again I'm interested if it is just a convenient hypothesis of those sources (convenient because then it seems a non-issue to hear/feel the ancient meter while still reading it normally as a prose without the medieval practice of accenting the ictuses) or whether they have also some other evidence that leads them to that conclusion.

But thank you very much for this discussion, it was most informative for me and I may try to change it - maybe rather in poetry than prose though (not because I necessarily believe it should be so, but because it is convenient, at least while reading poetry, where one wants to feel, as you say, the "rhythm") :)


==============================================================

I'm posting also another recording of the older text from the other thread. Who wants, can try to read it too! :) And who has read it already (as me in the other thread) can try to read it again and do it better now!

Deus Inter Hominēs
1. Barba, gladiārum clārissimus, fortissimus, et optimus sum.
2. Anteā capvus eram; nunc rō in gladiāriō dō mē exerceō.
3. “Crās eris in populī Rōoculīs”, inquit lanista meus.
4. “In naumachiīs prope Tiberim aut mare pugbis.
5. Factio tua vincet et populum dēlecbitis
6. Deus es et glōriōsissima ē fornīs tibi erit !“
 

Alatius

Civis Illustris

  • Civis Illustris

Location:
Gothoburgi
Thank you, Alatius, for the explanation and the measurement, it is most interesting. The thing is whether this is an issue or not.
You're welcome Godmy. :) Whether this is an issue or not is probably something you have to decide for yourself. I found the irregular rhythm distracting, but how a Roman would perceive it is anyone's guess.
The "o" was short so much as it would be relatively short to that ē in the same word in a living language (e.g. Czech) that has minimal long-short vowel pairs for all its vowels (If I express myself clearly) and the long vowels in the same word must be always "relatively" long to the length of the short vowels in that word (but not unnaturally long).
I agree with last part of this, but the first part is simply not true. In this particular instance of omnēs, the ē was pronounced more than four times longer than the o, which is well above what is normally used to make a contrast. From a quick search I was unable to find data for Czech but I did find information about the durations of Finnish vowels:
By extracting data from the first diagram on that page, I estimate that, in unstressed syllables, a short Finnish vowel is on an average held for 45 ms and a long vowel 140 ms, i.e. 3.1 times as long. In stressed syllables, the short is 75 and the long 175 ms, i.e. 2.3 times as long. Thus, stressed vowels are generally longer (no surprise there), and if we compare a stressed short vowel and an unstressed long (as in omnēs), the Finnish speaker would say the ē 1.87 times as long as the o.
Then there is "ē" which is more than twice longer than "o", however Allen is not so precise as you are, with his recommendations about "roughly twice as long"...
(Again, your ē is more than four times as long.) As I haven't said anything about exactly how much longer in my opinion a short vowel should be compared to a long, I can't understand how you have come to the conclusion that I am more precise than Allen. "Roughly twice as long" sounds like a good rule of thumb to me. Naturally we would see a huge variation if we could analyse authentic Roman Latin: I would not be surprised to find 1:3 ratios occasionally, nor 1:1.5 ratios.
...so in practice this (ratio 1:3, almost 1:4 with a short vowel under 100ms which could be argued to be a limit length where a vowel still can be perceived) would be still roughly in the tolerance...
(A vowel is surely perceivable well below 100ms.) Allen proposes 1:2, but if 1:4 is "still roughly in the tolerance", what about 1:5, 1:6 etc.? By what standard do you draw the line?
...also you would expect the ratio to go down almost to 1:2 as the short vowel would become longer and more emphatic
Absolutely agree! Apart from the m, my complaint is partly that your o here is not emphatic enough, and partly that the ē is too long.
So then the "m" would have to be longer and that's.... a hypothesis.
I argue that it would be automatically, if the syllable was truly closed by it, just as your n is fairly long (and rightly so in my opinion) when you said "inter".
Also what you say next that there "obviously must have existed some real tangible difference in the nature of long and short syllables" doesn't have to mean that the syllables long by position with a short vowel were comparable to open long syllables. That's a hypothesis of some philologists and also recommendations of many books dealing with Latin poetry where the author honestly wants the student to "hear" it at all costs. But it may also be that a long syllable by position with a short vowel will just be "longer" than a short syllable but not necessarily the same length as a long open syllable.
It was partly this I had in mind when I mentioned ăprōs as an example: instead of comparing syllables long by position with syllables long by nature, let's compare closed syllables and open syllables, both with short vowels. When you pronounce that word, do you really manage to make the first half even shorter than what you did in omnēs? (Feel free to think up another example with other vowels, if you wish.)

For the record, I don't find any issues with the rhythm in your gladiator clip. (Maybe the r in fortissimus could have been slightly longer, but that's it.)
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
Thank you for your answer. Lots of your points were mentioned somewhere later in the message (where I e.g. reflected on the 1:4 - on which you then reacted few times as if I didn't acknowledge it, but that's just how it appeared. I partially explained it by the short vowel being just 80ms - being extremely short and the long vowel not being comparatively extremely long-short. About 80ms - what I said is that such vowel in a casual speech, not isolated, inside a word may be extremely short and you wouldn't expect this to go down. If listened to isolated, it would surely be perceivable. I just wanted to say as a last thing that we would still have to explain why in maybe good 50% in poetry the muta cum liquida rule is ignored: my hypothesis - because the closed syllable wasn't especially long just because it was close as to be comparable with an open long syllable - unless the poet really tried while declamating (emulating the Greeks?) and so there wasn't a difference how it was syllabified, but that's again just another hypothesis and especially such to contrast it with the ones you mentioned. Anyway thank you again for your response, maybe we can continue the discussion in a different thread, so the potential uploaders are not confused what this one is about (since it is still quite new and the only recordings here are from me).

Anyway if you or somebody else here are interested in uploading a recording, I will be glad :)
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
No problem :) It was a feedback... the only thing is that I want more people to participate, not to feel as outsiders while the thread is so new; -) To bring it here alive...

I'm posting another text which people can try to read (or imitate), along with a recording :thumb-up:
Urbs ma in Tiberis flūminis sita est vīginlia passuum ā marī. Hōc loflūmen facile trāntur, et collēs propinquī bene mūpossunt. Moenia Rōna anqua septem collēs ve montēs complectuntur, quōrum haec sunt mina: Patium, Capilium, Avennus, Caelius, Ēsquiliae, Vīmilis, Quirīlis. Ā colle Quirīlī et ā monte Capitōlīnō usque ad Tiberim flūmen campus Mārtius patet.

Also a recording of a part of Ovid's Metamorphoses for somebody to try :)
In nova fert animus / mūtās cere formās
corpora; , coeptīs / (nam vōstāstis et illās)
adspīte meīs / prīqu(e) ab ogine mun
ad mea perpetuum / dēcite tempora carmen!
Ante mar(e) et terrās / et quod tegit omnia caelum
ūnus erat / nārae vultus in orbe,
quemre chaos: / rudis ingestaque lēs
nec quicquam nisi pondus iners / congestaqu(e) eōdem
nōn bene jūnctārum / discordia mina rum.
(details about this text and its reading: It is a dactylic hexameter. The slash / symbol is a proposed caesura (the pause) by me. The underlined syllable is a beginning of another foot = the ictus, but is given in this reading (which is also what is usually recommended) no special phonetic prominence (i.e. is not read as an additional stress / or the only stress) or given some special emphasis - it serves just a graphical purpose here. Otherwise the syllables in bold are those that are stressed/accented (no bearing on the meter). If a vowel on the end of a word is in brackets (), it is elided by the following vowel and thus not read. )
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
You may also like to try this one, since it is a beginning of the famous textbook (Familia Rōmāna of Lingua Latīna Per Sē Illūstrāta, H. Oerberg) and also a new recording of mine:

IMPERIUM RŌNUM
ma in Italiā est. Italia in Eupā est. Graecia in Eupā est. Italia et Graecia in Eupā sunt. Hisnia quoque in Eupā est. Hisnia et Italia et Graecia in Eupā sunt.
Aegyptus in Eupā nōn est, Aegyptus in Āfricā est. Gallia nōn in Āfricā est, Gallia est in Eupā. Syria nōn est in Eupā, sed in Asiā. Arabia quoque in Asiā est. Syria et Arabia in Asiā sunt. Gernia nōn in Asiā, sed in Eupā est. Britannia quoque in Eupā est. Gernia et Britannia sunt in Eupā.
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
And the last recording for now for somebody else to try to read (or imitate) and send a recording of his own, this one is the beginning of the Latin translation of Harry Potter, Harrius Potter :)

Dominus et Domina Dursley, quī vībant in aedibus Gestātiōnis Ligustrum numerō quattuor sigtīs, nōn sine superbiā dībant sē ratiōne ordiriā vīvenūneque sē paenire ilus ratiōnis. In orbe terrum vix crēdās quemquam esse minus deditum bus novīs et arnīs, quod ineptiās lēs omnō sperbant
 

Coffea est bona

New Member

Salvete Omnes!

Let me know what you all think. This one was fun to read, but I don't think I'm pronouncing "EU" in Europa correctly though. This is my restored classical Latin pronunciation. Very very American English:)
 

Attachments

Top