Talk:Latin phonology and orthography/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Chevrons?

May I protest at the rash of angle brackets/chevrons that have recently disfigured this article. They're ugly, quite unnecessary, and meaningless to most readers. --rossb (talk) 15:00, 1 November 2009 (UTC)

  Sorry, I don't know what you're referring to...can you give us an example?
  William J. 'Bill' McCalpin (talk) 20:11, 1 November 2009 (UTC)

If you look at "Consonants" for example you'll see beneath the table: "‹c› and ‹k› both represent /k/" with the chevrons round the c and the k. And many more examples --rossb (talk) 20:18, 1 November 2009 (UTC)

They're used at orthography related articles like List of Latin digraphs, Spanish orthography, and Trigraph (orthography). They might be "ugly" to you, but AFAIK, that's how you do it I've also seen <angle brackets like these>. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 02:31, 2 November 2009 (UTC)

I can see both points of view. The chevrons mark a grapheme (I think it is an allograph), just as slashes mark a phoneme and brackets mark a phonological representation. Most people don't know any phonology, either. There is an issue of consistency. If we don't use grapheme markers why should we use the others? And yet this can hardly be a professional type article without some phonology. It's a tough question. I appreciate your chiming in, Ross. Let me try this. We already agreed not use small caps as that is very confusing. How about if we put a few footnotes in there to explain our usages and term? This might be done in two autogenerated sections, sources and footnotes, as I've seen done in and have added to some articles (a minority). I have not yet heard the Wikicops complain about that. If they don't like something they usually say so in regulative language. Let me see if I can find the code again. However, the question is by no means closed and certainly will spill over into other articles. Just how linguistic shall we get in linguistics articles? We don't want to mystify the public.Dave (talk) 05:26, 2 November 2009 (UTC)

PS. What a mess. I tried caps on the first table note. I better not go further until we get this resolved. The chevrons do not fit the caps so we have either to go over to genuine leather angle brackets, abandon the caps, or abandon the angles. For sure the public is not going to know that all we mean by the angles is that K alternates with C for /k/. Maybe we should wait for more comments. By the way much of those unreferenced notes are wrong and have to go. I suppose I could work on that. I'm staring at at least a dozen totally unreferenced Latin articles full of linguistics jargon that contains many errors. One thing at a time I guess.Dave (talk) 05:45, 2 November 2009 (UTC)

I don't think the analogy wish slashes and brackets is valid.They're needed to signal that the characters in question are being used in a special way, as symbols to indicate the pronunciation rather than as ordinary letters: without them, the readers would be seriously confused (although I do wonder whether we explain the usage sufficiently in articles). But surely graphemes (including allographs) are just examples of letters of the alphabet, and don't need this extra marking off, which I still think is very ugly and detracts from the readability of the article. In other articles I've seen italics used in a similar context, and maybe this would work here. --rossb (talk) 07:57, 2 November 2009 (UTC)

In adding the chevrons, I noticed sentences like: In classical verse, the letter Z always counted as two consonants. The chevrons negate the need to say "the letter" in this context. Similarly, we use slashes and brackets not just to mark off pronunciations but also to eliminate the need to say "the labial element of the phoneme kʷ".

I honestly don't see how the chevrons make the article less readable. I can concede that it's often neutral (as slashes and brackets can be), but there are times when they can clarify or help reinforce that we're talking about spelling and not a word, such as in the first footnote under the consonant chart: "Misunderstanding of this convention has led to the erroneous spelling ‹caius›". Thus we make a distinction between words (in italics) and orthographic items (in chevrons). — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 08:23, 2 November 2009 (UTC)

We might want to add a footnote explaining what the brackets do. But whenever we deal with orthographic variants, we can potentially lose the reader if we don't dab. kwami (talk) 21:18, 2 November 2009 (UTC)

Should we perhaps create a template like we do warning readers about IPA usage? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 01:48, 3 November 2009 (UTC)

I'm quite unconvinced that "<Z>" is an improvement over "the letter Z". We should we writing articles for the general reader not the specialist. The slashes and brackets for pronunciation are necessary, but the chevrons are not, and I strongly feel that they have no place in this article. --rossb (talk) 07:35, 3 November 2009 (UTC)

If the general reader can understand slashes and brackets, I think they can get chevrons, so including them isn't an appeal to specialists. Do you say that slashes and brackets are necessary because conventions require them or because they're a lot more likely to disambiguate than the chevrons? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:53, 3 November 2009 (UTC)

If the general reader,particularly a reader who's seen other Wikipedia articles, or has used dictionaries, sees slashes and brackets, hey are quite likely to realise that it has to do with pronunciation, although they may not grasp the difference between phonemic and phonetic. The point is that by surrounding the letters with these characters we're marking them out as no longer letters of the alphabet in the ordinary way but as having a special function. So when they see chevrons they're quite likely to get confused and think that they've also got something to do with pronunciation. We're not here writing for the specialist but for the general reader, and I would strongly urge that the chevrons reduce the readability of this article rather than improving it, and should therefore be removed. --rossb (talk) 07:05, 4 November 2009 (UTC)

I see what you're saying, but I don't agree that the chevrons are all that confusing. So far, you're the only one who'se brought it up and I assume that you yourself weren't confused but rather you believe a significant enough portion of our readership could be. Am I correct? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 07:32, 4 November 2009 (UTC)

I'm concerned that the chevrons will just make the article off-putting to general readers. Those readers who have seen articles with slashes and brackets are quite likely to think that the chevrons are some further variation of pronunciation. I think I've already said that the slashes and brackets are a necessary evil: we need them to distinguish between letters of the alphabet and phonemic/phonemic symbols, which just happen, perhaps unfortunately, to use some of the same glyphs. But in the case of the chevrons they're really just saying that the letters of the alphabet are being used as letters of the alphabet. They're totally unnecessary, and make the article less readable, without as far as I can see giving any benefit. -- This last contribution was by me: I'd forgotten to log in. --rossb (talk) 19:27, 9 November 2009 (UTC)

Well, we've both given our opinions on the matter and it doesn't look like we can solve it by butting heads. I can bring it up at the language reference desk. Should this be brought up somewhere else? It seems like an issue that would affect many more Wikipedia articles. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:29, 10 November 2009 (UTC)

As a follow-up, this is from a conversation at User talk:Kwamikagami#Angle brackets:

They're in the maths section of Unicode. I've just been checking Google Books, and many texts use either <...> or guillemots as an approximation, even though they call them "angle brackets". However, I did found this,[1] which uses true angle brackets for literal transliterations of Syriac, and closer to home this,[2] which makes the [x], /x/, <x> distinction. (Also here, here, here, etc.) So it seems pretty clear that when sources say "angle brackets" are used for this, they actually do mean "angle brackets", even if not all printers stock them.

So the convention is ubiquitous, even if it isn't universal. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:33, 23 November 2009 (UTC)

Correction to my comment: there's a second set at U+2329/A, which are the non-math angle brackets. (They were deprecated and replaced for math use because of equivalence with CJK punctuation.) kwami (talk) 02:49, 24 November 2009 (UTC)

The aspirated plosives

Th, Ph and Ch were not exclusive to Greek loanwords. Not in the classical period anyway. They appear in a number of words of non-greek origin such as lachrima, pulcher, triumphus, bracchium, Carthago, sepulchrum and others. I've gone and fixed this, with a note on these sounds' peculiar status in Latin phonology. Szfski (talk) 11:18, 30 October 2009 (UTC)

I am not sure if one is entitled to light-heartedly describe all of these words as "native Latin words"; bracchium is from βραχίων; triumphus from θρίαμβος via Etruscan, and lachrima from δάκρυμα (all three of them were borrowed into preclassical/old Latin). Plus, from a strictly historical/diachronic viewpoint, even the rest of the words might not be considered native, since they don't have an IE etymology. I do understand what you are saying and I agree with your proposal about a disambiguation, but I find the phrasing you used a bit ambiguous. --Omnipaedista (talk) 12:32, 30 October 2009 (UTC)

That's largely irrelevant. By the classical period, there's no reason to think that most Romans would have seen lachrima or triumphus as foreign words anymore than native English speakers think of words like froth, move, chair, skirt, candle, they or sky as Norse or French words. Even if that weren't the case, Latin-speakers clearly didn't aspirate these words simply to mimic their counterparts in Greek, since if they had done so, they wouldn't have added an aspirated ch where the original Greek word had no aspiration (i.e. as if the Greek cognate were δάχρυμα) or refrained from rounding front vowels where they are rounded in Greek (as if the Greek form were were δάκριμα.) This is an entirely different (and, yes, native) phonological process from the conscious imitation of Greek sounds in learned words like pyramidum and Philtrum.

Moreover, the fact that a word cannot be found to have any discernible ancestor in a proto-language doesn't necessarily make it a borrowing or in any meaningful way "non-native." (c.f. American English words like wacko, okay, howdy, bogus, clobber, malarkey, gizmo, bonkers and doodle.) More to the point, the fact of having been borrowed into some earlier stage of a language in no way prevents that word from being subject to the language's native phonological rules in later periods. Diachronic evidence is irrelevant when evaluating what was, essentially, a synchronic phenomenon specific to the classical period.

In any event, even if all the above weren't true, your argument still wouldn't hold up since sepulchrum *does* have a PIE etymology: *sep- "to honor (one's dead)" "to handle skillfully" (c.f. Sanskrit saparyati,' Greek ἑπειν)

In sum, I don't really see your point. Szfski (talk) 13:42, 30 October 2009 (UTC)

I don't disagree with any of the above (and of course you are right about sepelire and ἕπειν); I didn't defend an opposite opinion. All I was trying to say (but obviously didn't express it adequately --sorry for being needlessly arduous) is this: the term "native" by itself might cause ambiguity in the minds of readers who are unaware of the difference between the synchronic and the diachronic levels of description (but know about the etymologies of these words), and thus it might be desirable to add an additional characterization to the adjective (as it appears in the text); for example: e.g.: "native Latin words" --> "synchronically native words" (unless of course, the consensus is that such a change would be überpedantic, and therefore unnecessary). --Omnipaedista (talk) 16:21, 30 October 2009 (UTC)

hmm you may be right. Let me see if I can phrase it better. Szfski (talk) 22:30, 1 November 2009 (UTC)

I don't now how relevant is this to "aspirated plosives", but out of your examples, "lacrima", "pulcer" and "sepulcrum" are better written without h. Abuse of aspirates is affectionate for classical times (see Quintilian's remarks on that), and later such writing gets more common because "h" in later times gets silent. Mamurra (talk) 16:35, 20 November 2009 (UTC)

I take it you mean "an affectation in Classical time";) garik (talk) 16:42, 20 November 2009 (UTC)

This is a descriptive, rather than prescriptive, article. Therefore notions of "abuse" of sounds or which sound is "better" than another are irrelevant to the topic at hand. In any event, the use of aspirated stops in native Latin words of non-greek origin was not a purely graphical phenomenon, nor was it an affectation. It represents an actual change in popular Roman speech habits, for which there exists a fair amount of evidence. For example, such words were often transliterated in Greek in ways that suggest an aspirate pronunciation (e.g. πουλχερ, ανχορα, σεπουλχρον). Some grammarians describe this as a phenomenon of popular speech, such as Cicero (Or. 160) when he says "usum loquendi populo concessi" to describe his acceptance of the fact that stops in non-greek words can be aspirated. Moreover, this is an entirely different phenomenon from the hypercorrection described in Catullus 84, since this aspiration occurred regularly even in vulgar Latin words borrowed into Celtic languages, as evidenced by Welsh hefys and Breton hiviz from Vulgar Latin *chamisia (had the original borrowed word been the more standard camisia, we would expect *cefys to occur in Welsh, just as we have cadwyn and ceffyl from vulgar Latin catena and caballus.)

The silencing of H did not apply to instances of H in digraphs representing aspirates. "Ph" became later /f/ in all environments (c.f. french Triomphe, Spanish Triunfo.) Likewise, a uvular, glottal or velar fricative very often occurs as a reflex of Ch in vulgar Latin words when borrowed into languages that had them.

For sources see Archivum Linguisticum, X (158), 110 ff. along with Allen, W.S. Vox Latina p 26-27. For data on Celtic see Archaeologia cambrensis Vol IV, Series IV, pp 359. Szfski (talk) 19:15, 20 November 2009 (UTC)

I think what Mamurra may be referring to in part is that the article says this allophony resulted in "standard [written] forms" such as pulcher, lachrima, gracchus, triumphus. While this article is descriptive, "standard" by definition includes prescription, so the question is whether the allophony was prescribed and whether the spelling indicating the allophony was considered "correct" or whether it was just a common mispelling. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 05:16, 21 November 2009 (UTC)

That depends on the word in question, and the particular grammarian you're looking at. Cicero, for example, came to accept Pulcher, Carthago, Triumphus and Cethegus as "proper" but wouldn't stand for Otho, Chorona or Sepulchrum in either speech or writing. Of these, I believe Marius Victorinus only mentions Pulcher as acceptable. The grammarians in general show a fair amount of disagreement in the matter, but it seems fair to say that the general practice was to represent the allophony graphically in certain words such as Carthago and triumphus to such a degree that they became the standard forms, at least in the inscriptions and papyri from the 1st century B.C. onward. Allen speculates that this may ultimately be due to a residual Etruscan substratum. Szfski (talk) 06:44, 21 November 2009 (UTC)

Ah, that's pretty much how I interpreted it. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 15:55, 21 November 2009 (UTC)

@garik: you're right, I wanted to say something more like "pretentious" (sorry, English isn't my native language and sometimes I make horrible errors). In any case, I see no etymological (nor any other) justification for "h" at least in lacrima (< gr. dakry) and sepulcrum (< sepelio > *sepulculum or such, no aspirate). I am also in doubt about "pulcer". If the spelling with "h" is purely ortographical thing, does it belong to phonetics? Mamurra (talk) 14:09, 24 November 2009 (UTC)

Mamurra, I can't shake the feeling that you didn't read my most recent post here. At least, not very carefully. As I said, the aspiration of consonants such as the /kʰ/ of lachrima does not require any justification, etymological or otherwise. It was a purely synchronic process that, unlike the learned aspiration of words like cithara or theatrum, did not depend on the word's etymology. Moroever it was *not* a purely orthographical phenomenon, as Cicero (Or. 160) and others explicitly state, and modern scholars such as W.S. Allen confirm. Szfski (talk) 16:59, 24 November 2009 (UTC)

That's possible that I don't understand your point. But you at least seem mistaken with the reference to Cic. Or. 160: Cicero explicitly states there that "lacrima" is commonly pronounced without aspiration, and he also is doing that: "usum loquendi populo concessi" (as to pronunciation of pulcher, Cethegus, triumphus, Carthago) "Orcivios TAMEN et Matones, Otones, Caepiones, sepulcra, coronas, lacrimas dicimus, quia per aurium iudicium licet". So I admit I don't understand, why lacrima or sepulcrum underwent inclusion into a discussion of aspirated plosives... Mamurra (talk) 12:59, 25 November 2009 (UTC)

Pronunciation Problems

The author of this page misinterpretation several statements from Vox Latina:

1) The assimilation of n to [m] takes place only in the preposition in before a word starting with labial. see p28 ¶1

2) n before f represents a labio-dental nasal [ɱ] p29 ¶2

3) I see no evidence for the pronunciation of qu as [k^ɥ] before front vowels on p17. —Preceding unsigned comment added by 134.74.184.7

Take a more careful look at Vox Latina.

The assimilation of n to m is indeed described as occurring with the preposition in, however it does not say anywhere that this is the only environment where such assimilation occurred. Moreover, on page 31, he cites inscriptional evidence and explicitly quotes Cicero and Velius longus for examples of /m/->/n/ in other environments.

On Page 29 Allen states that the same considerations apply to f as to s for preceding n, and uses this as an explanation for inscriptional forms such as cofeci, iferos for confeci, inferos. He then goes on to state that the labio-dental pronunciation existed as an artificial restoration of a previously dropped n.

On page 17, the penultimate paragraph cites Priscian (K. ii, 7) as stating that the "u element of qu when followed by a front vowel has a special quality like Greek υ (i.e. like the initial sound of French huit as contrasted with oui). In a footnote he confirms this for the classical period with inscriptional Greek spellings such as Ακυλιος and Κυιντιλιος for Aquilius and Quintilius, and κυι for qui as against κοα for qua.

Happy re-reading! Szfski (talk) 19:55, 5 November 2009 (UTC)

On point 1 /m/ -> /n/ in not the same as /n/ -> /m/ and to quote the silence of an author as proof seems like a bad case of reading between the lines. 64.131.187.175 (talk) 21:44, 3 December 2009 (UTC)

Division of syllables

Currently, there is a discussion on Wiktionary, concerning the division of syllables, on whether a preposition can form a syllable with an initial vowel of a verb in compounds or whether the separate parts of compounds are kept distinct. It would be helpful if anyone had any more information or views on this issue. For the discussion on Wiktionary, see [3] further down. Caladon (talk) 09:13, 13 December 2009 (UTC)

Pronunciation of short o

The article says that short o is pronounced [ɔ] - is there any good source for this? Allen seems to prefer [ɒ], which I would have thought is more likely. --rossb (talk) 12:45, 26 December 2009 (UTC)

Why is it more likely? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 22:46, 26 December 2009 (UTC)

Sonus medius?

Shouldn't this article mention sonus medius? See the article about Claudian letters, as well as this mention in Italian wikipedia. Jec (talk) 21:56, 25 April 2010 (UTC)

It already did, though it didn't name it. Fixed. ― ___A._di_M. (formerly Army1987) 22:39, 26 April 2010 (UTC)

Nasal vowels

Is there a source that says nasal vowels were phonemic? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 06:28, 10 May 2010 (UTC)

Not directly that I know of. But Wells says,

"The reason that a vowel plus m was subject to elision was that the spelling m here did not stand for any actual nasal consonant but just for nasalization of the vowel."

He doesn't describe it as allophony, and as behaving as any other vowel. — kwami (talk) 06:31, 10 May 2010 (UTC)

They contrast with non-nasal vowels as in rosa vs rosam, but I dunno whether they contrast with vowel-plus-/m/. ― ___A._di_M. (formerly Army1987) 07:43, 10 May 2010 (UTC)

I think we need a little more than that to assert that they were phonemes in the article. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 08:14, 10 May 2010 (UTC)

Even if there's no contrast with Vm, one could argue that they're phonemically simple vowels if they behave as simple vowels. — kwami (talk) 08:18, 10 May 2010 (UTC)

Right but our standard is if somebody has argued it, not if somebody could argue it. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 17:46, 10 May 2010 (UTC)

There seems to be disagreement as to whether they were phonemic or not. Part of the problem is that Latin orthography was not designed for nasal vowels (presumably because they had not yet developed at the time the Etruscan alphabet was adopted), though there are verbal descriptions of them and various ad hoc transcriptions, such as writing only half of the M or N, or marking the vowel with a tilde (this in Classical times). Rodney Sampson (Nasal vowel evolution in Romance, 1999:42-43, 49-50) has this to say:

It is generally assumed that in Classical Latin there were no nasal vowel phonemes [...] However, there are grounds for believing that strongly nasal vowels arose at various stages in the history of Latin including the Golden Age period and that phonemically nasal vowels may even have become established in some varieties of Latin. This is indicated by the fact that certain phonological changes have operated which seem to imply heightened vowel nasality. [...] Many scholars have concluded that after syllable-coda nasal consonants were deleted long nasal vowels appeared in Latin [...]. However, [...] few have wished to accept that these nasal vowels would have given rise to surface contrasts with oral vowels and hence would have been phonemic, as in DĒSUM 'I am lacking' vs. DĒNSUM 'dense' [...] ROTA 'wheel (nom. sg.)' vs. ROTAM 'wheel (acc. sg.)' etc.). Maniet [..] indeed explicitly denies phonemic status to the nasal vowels deriving from word-final vowel + M sequences, describing them instead as combinatory variants with demarcative function. Nonetheless, the evidence for surface contrasts between nasal and oral vowels is strong and it suggests that at certain periods and in certain styles of speech nasal vowel phonemes may well have arisen. [Foot note:] A similar view is adopted by Safarewicz who recognizes distinctively nasal vowels in Latin, which he claims survived down to the secord century AD [...]

So it would seem there is a significant minority of scholars who do argue for their phonemicity. — kwami (talk) 18:38, 10 May 2010 (UTC)

Nice find. The info on the scholarly disagreement should be reflected in the article. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 05:53, 11 May 2010 (UTC)

Meta-clarification request

"Z /z/{{#tag:ref|/z/ was at first represented by ‹S› or ‹SS› in Hellenistic Greek loanwords (e.g. sena from ζωνη). In the around the second and first centuries BCE zeta (‹Ζ›) was adopted to represent /z/. Based on Italian Greek, where ‹ζ› still represented /dz/, di- and de- before a vowel in Latin was represented Z standing for /dz/: zeta for diaeta. Thereafter Z was either /z/ or /dz/.

What clarification? It seems pretty clear to me. <Z> can be either /z/ or /dz/ depending on the scribe or the location. Did I miss something? I'm temporarily removing this tag until I can see what you mean or what you need or think you need.Dave (talk) 13:22, 28 June 2010 (UTC)

I didn't tag it, but the second to last sentence doesn't make sense. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 17:05, 28 June 2010 (UTC)

Consonant table design

I thought this table and its notes were a bit hard to read. The table was too dense, the notes were too wide. It was not clear that they are table notes. The C number might be part of the phoneme, for all the public knows. And yet, the encyclopedic information is all there. We would not want to omit it. Therefore this seemed to be a problem in layout and design. How can we present the table and its notes in a clearer fashion? If we had some of the manual production tools that are out there the job would be much easier. But what we have is WP markup language, which is not bad. So, I tried my hand at a new layout. We want the notes in the table so the reader will identify them with the table. We want to explain the C numbers. We don't want a section with a little tiny head and a great body, like a certain character in Beetlejuice, so we want to open the table out. Also, opening it out makes it less dense and easier to read. If you disagree with the analysis or like the other design better speak up. Try your hand at your own design, but be careful not to leave it messed up. If you just want to change the width there is now a width parameter in the table to do that, at the top. I tried 100% but the page looked a bit crowded to me so I went for 80%. We could make it less and put some sort of graphic right. Also by using the "reflist" template instead of "references" we could make the note font smaller. I tried that but really some of the characters became too small to read. So, what's your input, if any?Dave (talk) 23:39, 27 June 2010 (UTC)

I don't like the change. The columns are much too wide and I don't see this as an improvement (the table is no denser than others if its kind at other phonology articles). Rather than have the notes be part of the table as you have done, we could have a header like the one at Spanish phonology. Something like "phonetic notes."

It doesn't seem to me that the notation is too confusing, but instead of having C1, C2, etc, we could have letters or roman numerals (a, b, c; i, ii, iii). — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 23:49, 27 June 2010 (UTC)

Thanks for your input (input, Stephanie, input!). Spanish phonology is a good-looking arrangement. The table is centered. The table notes are a bit different; the auto-generated note system is not there. It seems less cluttered. If we keep the autogeneration system then actually I like either the letters or the small Roman numerals. In summary these alternatives have been suggested by ye: 1) close-up centered table without internal notes 2) A separate subsection for the notes 3) letters or Roman numerals instead of C numbers 4) no autogeneration of notes. These are good suggestions. Does anyone have more? No point in holding your peace. We don't bite. When we get a good modicum of suggestions then we can vote on these specific features. Anyone else throwing his hat in the ring? Now's the time for input. Once we take a vote (there are three of us currently) one of us can change it (probably me). Incidentally I see the discussions we used to have on the phonetic symbols have all led to a resolution. I'm glad of that. Apparently we have a linguistics article here rather than a digest. Fine with me. The public has to come up once in a while; we can't have all non-technical material. Now let's make it into a good one; however, I would not like to wait quite so long as previously. Still, we need some time for input to develop.Dave (talk) 12:11, 28 June 2010 (UTC)

Well, I only pointed out the Spanish phonology article to make suggestion 2. I don't have a problem with internal notes. I think it can work either way. If we decide to keep them, having the autogeneration is a nice feature. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 17:11, 28 June 2010 (UTC)

Vowel Question

In the vowel chart the short vowels are given as [i], [e], [u], [o], but in the second bullet point it says that they should be [ɪ], [ɛ], [ʊ], [ɔ]. Which one is correct? 64.131.187.175 (talk) 14:36, 27 November 2009 (UTC)

When vowels appear between /slashes/, it means they represent phonemes. In this case, the phonemic representation for short vowels is slightly different from their actual phonetic realization. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 17:23, 27 November 2009 (UTC)

I am really sorry about my lack of linguistic skills, but can you explain this in difference in more detail please. If I understand correctly a Latin speaker would of perceived a [ɔ] as a short o and a [oː] as a long o. Is this correct? 64.131.187.175 (talk) 20:50, 27 November 2009 (UTC)

Yes, you are correct. However, we've chosen to represent the short o as /o/ when we're using the more abstract phonemic representation for a number of reasons. We discussed the matter a couple of years ago here. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 21:32, 27 November 2009 (UTC)

The problem is that the reasons given are in that discussion are not supported by the sources. I quote from the Vox Latina (p 47): "There appears to have been no great difference in the quality between long and short a, but in the case of the close* and mid* vowels (i and u, e and o) the long appear to have been appreciably closer than the short."

Aeusoes's statement that the letters are distinguished only by length is thus incorrect, and the statement that sources use the same symbol for both was successfully countered in that discussion. There are not several reasons.

While phonemic IPA transcriptions do not necessarily match the phonetic transcriptions, there has to be a reason why. If we all agree that the more open symbols are more correct phonetically, and the literature that actually uses IPA agrees, there is no reason to continue to confuse people with the two different notations.

I'm having to right now clarify this for someone who has been misled by this article. — trlkly 10:19, 14 December 2010 (UTC)

Did I say they were distinguished only by length? My argument is that the phonemic representation with both long and short vowels being transcribed the same is consistent with the literature's representation, not with the phonetic accuracy. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 13:32, 14 December 2010 (UTC)

What IPA symbol should be used for the open vowel?

The IPA vowel chart as shown in Vowel#Articulation shows the open (front / central / back) unrounded vowels as [a / ä / ɑ]. Latin spelling and pronunciation#Vowels shows a central set of vowels using /a/. A recent edit to the table "Classical Latin alphabet" in Latin alphabet#Origins has changed the vowel in one of the four pronunciation entries using [a] to [ɑ]. Which IPA symbol should be used? —Coroboy (talk) 04:48, 14 May 2011 (UTC)

As far as I can tell, ‹a› is the correct symbol, even if it is central. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 19:10, 14 May 2011 (UTC)

Small caps

I replaced full-sized caps in the phoneme tables to small caps. Is this what's intended?

Also, Latin in the article is rendered inconsistently: sometimes in small caps and sometimes in ordinary lettering. Should all Latin be converted to small caps, or are there cases where ordinary lettering is to be used instead? — Eru·tuon 00:06, 29 December 2011 (UTC)

Actually, you did not turn them into small caps. The template used this way: {{sc||A}} does not do anything with the input "A": it just passes it through. Since is is a capital, the output is a full capital (no smallcaps). Good example caps and small caps: A -- a.

Next: I have edited out all uses of {{sc}}. First because it is misunderstood (e.g. in you edit). Second because I made the page single style: all letters in examples &tc are full capitals (now). Third because most uses were idle, pass through. Fourth because the {{sc}} template is to be merged into the regular {{smallcaps}}, so exotic usage is to be cancelled.

If you want to use small caps though (I do not object!), I suggest this: use them everywhere in the article to keep consistent style, and use like {{smallcaps|a}}. Important note: to get smallcaps, the input must be lowercase. Uppercase is not changed: {{smallcaps|Aa}} → Aa. So the first one is still a regular capital, unaffected! Another bad effect is that when copy-pasting these smallcaps from the page, one gets lowercase letters in the new place.

If I can help you, just drop a talk. -DePiep (talk) 13:24, 24 February 2012 (UTC)

Yes, he did turn them all into small caps. The behaviour of the template has changed. I reverted your changes. Please sub the template correctly, rather than just deleting the formatting. It's much easier to do it that way, automatically, than to go through it all manually again. — kwami (talk) 13:35, 24 February 2012 (UTC)

You could have read my last line. You also reverted the minor other manual edits. And note: now the page style is mixed again. The subst you point to should have been taken care of when changing the template. -DePiep (talk) 13:49, 24 February 2012 (UTC)

Oh, and on December 29, 2011, when User:Erutuon edited, the second parameter was not in smallcaps: [4]. It was explicitly made regular uppercase. Now, importantly, since you paid no attention to the page style inconsistence, and did not communicate a question, I will revert into my edits. -DePiep (talk) 13:59, 24 February 2012 (UTC)

And then there is this: For better accessibility, Latin quotations should never be set in all caps or small caps, even when such use might seem anachronistic., in MOS:Ety. If only one could Talk. -DePiep (talk) 15:48, 24 February 2012 (UTC)

Your latest revert is not helpful either, and its es is not helpful either. Why don't you just connect in a conversation? -DePiep (talk) 03:01, 25 February 2012 (UTC)

As in other articles, Latin words were displayed in small caps. This is good formatting style. Rather than replacing them with full caps, just so you can get rid of the template, it would be better to leave them as-is and place a cleanup tag. It's a lot easier to search for templated text and reformat it than it is to do a manual search after the templating has been deleted.

Never saw that MOS topic. I agree with it in general – a Latin quotation in an otherwise English text should be punctuated as in English – but this article is specifically about Latin. (Note the MOS topic is "Foreign terms", and in this case we're not dealing with foreign terms, but with the orthography of the language itself.) Latin formatting and punctuation is therefore appropriate. — kwami (talk) 03:24, 25 February 2012 (UTC)

That should take care of it. — kwami (talk) 04:31, 25 February 2012 (UTC)

R

Slight confusion: Article http://en.wikipedia.org/wiki/Latin#Phonology has "r if beginning a syllable = /ɾ/ (as in Spanish pero); r if finishing a syllable and rr = /r/ (as in Spanish perro)" while this article (http://en.wikipedia.org/wiki/Latin_spelling_and_pronunciation#Consonants) says "8.^ The Latin rhotic was either an alveolar trill [r], like Spanish or Italian ⟨rr⟩, or maybe an alveolar flap [ɾ], with a tap of the tongue against the upper gums, as in Spanish ⟨r⟩.[11]"

Also, when it is alveolar, why is it listed in column "Dental"?

As a side note: The IPA bracket symbols do not show correctly for me in IE, Firefox or Chrome. This is annoying. --93.220.56.18 (talk) 20:42, 5 March 2012 (UTC)

Roman ecclesiastical Latin pronunciation required in the Catholic liturgy?

I added calls for citations to the following claims:

Pius X issued a Motu Proprio in 1903 making the Roman pronunciation the standard for all liturgical actions in the Church meaning that any Catholic who celebrates a liturgy with others present be it the Mass, a baptism, or the Liturgia Horarum, then they are to use this pronunciation. The ecclesiastical pronunciation has since that time been the required pronunciation for any Catholic performing an action of the Church...:

I assume that this is referring to the instruction Tra le Sollecitudine (1903), issued motu proprio, by Pius X (Italian original; English translation). However, this document says nothing regarding how Latin is to be pronounced or spoken, only that it alone was to be used in solemn liturgical functions, and that the vernacular was not to be used for the variable or common (propers and ordinary) texts of the Mass or Divine Office (nn. 7-9).

However, in a letter from Pope Pius X to Louis Ernest Dubois, Archbishop of Bourges (later Cardinal Archbishop of Paris), dated 10 July 1912, he expressed his "great satisfaction that since the promulgation of Our MOTU PROPRIO of November 22, 1903, on Sacred Music, great zeal has been displayed in the different dioceses of France to make the pronunciation of the Latin language approximate more closely to that used in Rome" (Michael de Angelis, The Correct Pronunciation of Latin According to Roman Usage, ed. Nicola A. Montani, St. Gregory Guild, 1937, p. 4). He further writes, "We desire that the movement of return to the Roman pronunciation of Latin should continue with the same zeal and consoling success which has marked its progress hitherto; and ... We hope that under your direction and that of the other members of the episcopate this reform may be propagated in all the dioceses of France" (ibid.)

Pope Pius XI, in a letter to the same prelate (now Cardinal Archbishop of Paris), dated 30 Nov. 1928, writes: "We also esteem very greatly your plan of urging all who come under your jurisdiction to pronounce Latin more romano [in the Roman manner]. Not content like Our predecessors of happy memory, Pius X and Benedict XV, simply to approve this pronunciation of Latin, We, Ourselves express the keenest desire that all bishops of every nation shall endeavor to adopt it when carrying out the liturgical ceremonies" (de Angelis, p. 5) De Angelis also includes other letters from Pietro Gasparri, Cardinal Secretary of State, conveying Benedict XV's esteem for similar initiatives by a French abbot and Spanish abbot (p. 6). To Dom Marcet, he expresses that those who followed the abbot's "initiative have given proof of a filial and enlightened respect to the desires of the Holy Father" (p. 6). Notably, he (and Pius XI in the above quotation) did not speak of obedience to commands, but filial respect for desires.

Additionally, I am unaware of any official dictates (liturgical laws, instructions, or decrees), at least from the Holy See, which require that when the current ordinary form of the liturgy is celebrated in Latin, the Roman/Italianate pronunciation be used. Echevalier (talk) 22:41, 13 April 2012 (UTC)

Stress placement and vowel reduction

The article currently doesn't say a lot about stress, just how it relates to the weight of syllables from the end of the word. But the article does mention vowel reduction, albeit in passing. If I'm not mistaken, vowel reduction happened within the Old Latin period, and some very old inscriptions retain the old unreduced vowels. There are also some sources that state that the reduction was conditioned by the earlier word-initial stress placement (as in Celtic and Germanic) and that the shift to the classical stress pattern happened after the reduction (in the same way that the conditioning factor for Verner's law was erased by stress shift). So I wonder if some more information could be included in the article on this earlier word-initial stress, about how it affected vowels when reduction occurred, and how stress then changed into its placement in classical Latin? CodeCat (talk) 15:32, 14 November 2012 (UTC)

Changes by aeusoes1

My friend! Those are a lot of changes to hit us with all at once. You seem to have some knowledge of linguistics. Some, I say. I'm surprised you are in this Latin article because Latin is not in your repertory. I can say that because you are asking for references on things that are standard to Latinists and asking for clarifications of things that are clear to Latinists. Nevertheless an experienced editor with a linguistics background is not to be brushed off. I am sure we might benefit by some professionalism. First, I note that you have gone ahead and done this without discussion. Some of your tags call for discussion. I will in fact ask you for discussion. But, everything in its time. Here are some questions, which anyone may take up and answer. I say questions, I mean questions; that is, they are not rhetorical but are a request for data so that I may assess what you say. This is obviously going to be a long slow process but you have thrown your hat in the ring and I think we should go through it. Otherwise it will be on Wikipedia for a long time as a bad article until the sysadm's have to start calling for experts. I can put little bits of time on a regular basis.Dave (talk) 22:38, 31 October 2009 (UTC) PS I see from your user page you are begging off until December 28. And yet you have enough time for major changes to the article. Perhaps you may spare us a little time so that I may understand you. Otherwise I will have to decide these questions in my favor, and that would be a shame, as I like to understand people if I can.Dave (talk) 22:47, 31 October 2009 (UTC)

There's very little substance that I actually altered. There was an "other orthographic

Sometimes little things are big things when it comes to readability, so I see them as more "substantial". Right now substantial to me means amount of work required to check accuracy and format. Your unsubstantial is my substantial, but no conflict. Analogous.Dave (talk) 23:50, 31 October 2009 (UTC)

notes" section that ended up, for the most part, covering ground that was already discussed. The OGN about C and K sort of contradicted the footnote on it so I merged them and put in a cite request on the matter of dispute. There were a few places where I altered the wording,

I don't see the dispute. C replaced K. What is disputable or unclear about that? see next question when I get it typed. I think you must mean something else.

often to clarify that letters aren't pronounced but rather represent pronunciations. Otherwise, all I did was use single chevrons (‹›) whenever discussing graphemes or letters, make correct distinctions between phonemes (in /slashes/) and phones (in [brackets]).

I have to agree with the chevrons and those are xplained in another article. The public will not know why the chevrons but they can find out with a click and a read.Dave (talk) 23:50, 31 October 2009 (UTC)

It seems like I've done a lot more because I also eliminated spaces between bulletpoints, which messed up the way the track changes sees it. I did change the statement that length was

Well, I have not been through to where I was yet. I had not yet got beyond vowels (and I'm already starting to get bored with it). So, I'm not comparing what you did with the way it was, only checking, as it now seems, what YOU say. I do believe you are wrong there about the phonemic rather than morphemic. Length is often a just a phoneme but in this case it also is a morpheme. If you contrast those two words the only difference is the length of that one vowel and that changes the meaning, so it is a morpheme, and length there is morphemic, not vandalism. I'm the supposed "vandal" by the way.Dave (talk) 23:50, 31 October 2009 (UTC)

often morphemic to that it was often phonemic, which I'm was pretty sure was simply undoing unnoticed vandalism. Believe me, when I make substantive edits, I use sourcing.

Also, by the way, this article is geared towards a mass audience, not one of Latinists. So the two calls for sourcing I put were more of a request to improve the article rather than questioning the accuracy of the statements. Who knows, perhaps upon finding sourcing we'll find a better and more accurate way to word it. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]}

I recognize your edits as serious but your view is considerably ironic. I would have thought this article is one for linguists and you moved even further in that direction whereas the suppose Latinist articles are much easier to read and understand, in my view. In any case I am all for clarification so if you want a ref I can put it in there - but - no insult intended - you are not recognizing the refs that are there! On with the show.Dave (talk) 23:50, 31 October 2009 (UTC)

My point with C and K is that the two notes told two slightly different stories. I suppose it can seem obvious that words with kappa changed to have C; if Allen (2003) can be said to back that up, then we need only to move the citation to after that sentence.

Regarding length, if it can be morphemic then there needs to be an example to demonstrate that there is a "length" morpheme akin to other morphemes. Whether that's true, it is also true (and the example shows this) that length is phonemic for both consonants and vowels. What you describe as a morphemic contrast ("the only difference is the length of that one vowel and that changes the meaning") is actually a phonemic contrast, not a morphemic one.

At some point this article will be reviewed for good article status. If the references aren't clear enough, they'll point it out. I tend to go overboard with references when I edit (see Russian phonology, Catalan phonology, and Spanish phonology for some examples). — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 00:21, 1 November 2009 (UTC)

The small caps question

Do you have a reason for going over to small caps? I ask that because some of your small caps are smaller than lower case; for example, c. Roman square letters are fairly large graphemes. They did not have any lower case and the size of the square letters depended on the context. Would it not be clearer to use caps? Also, you left the caps in the consonant table. Are we doing this for consistency according to some standard or are we trying make things easy to read?What is your thinking here?Dave (talk) 22:38, 31 October 2009 (UTC)

In my edit today, I made some inconsistent usage of {{smallcaps}} to mark orthography (both in talking about graphemes and in examples), figuring if we decide to go against the smallcaps template it's easy to remove. I agree that when the Romans used block lettering, we should as well. I was unfamiliar with the smallcaps template before seeing it on this page, and I'm not sure which is more proper: {{smallcaps|text}} or {{smallcaps|TEXT}}, but I agree we should be consistent. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 23:10, 31 October 2009 (UTC)

If it OK with you then I think I will go back to Roman square letters - our caps- for those.Dave (talk) 23:21, 31 October 2009 (UTC)

Does the use of small caps for Latin represent the norm in Classical Studios, or is it basically a Wikipedia affectation? If the latter, it should obviously go. 96.50.105.102 (talk) 21:46, 10 February 2013 (UTC)

Replacement of K by C

Here is the statement I commented out:

"Words from Greek with kappa (‹Κ›) came to be represented with ‹c› instead."

Here is the preceding statement:

"However, in classical times, ‹k› had been replaced by ‹c›, except in a very small number of words."

At first I thought they were saying the same thing, but now I see the difference. The first statement would certainly need a ref if anyone had said it. It looks to me as though you made one up and asked for a reference on it. Off the top of my head I would say, this is in no way necessary and gives the wrong implication. C generally replaced K from any source, not because it was Greek. The whole alphabet was modified Greek. The Romans did not pick on Greek words for the replacement, they all were replaced. There is no point in that statement and it should come out. Where did you get it? If you got it from below, let's just take it out; it is wrong.Dave (talk) 00:05, 1 November 2009 (UTC)

Yes, I got it from another part of the article. Perhaps someone was under the impression that, at some point, Greek loanwords with /k/ got the k spelling and Latin otherwise used C. If you say it's wrong, then we shall strike it down. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 04:50, 12 November 2009 (UTC)

He/she's right. C is normative in Latin for /k/ regardless of etymology, even where Latin uses special characters to otherwise mark the word's Greek origins (c.f. Cyclops, cyclus, centaurus, Theocritus from Κύκλωψ, κύκλος, κένταυρος, Θεόκριτος.) On the other hand, the rare list of words which did use an optional K grapheme includes many forms (such as Kalendæ) which have no Greek etymology. The letter C is in fact ultimately derived from the Greek Gamma Γ.Szfski (talk) 15:12, 12 November 2009 (UTC)

This sentence in Consonant Table Note 1

⟨x⟩ represented the consonant cluster /ks/, where in Old Latin it had often been used for /ks/, which could be spelled ⟨ks⟩, ⟨cs⟩ or ⟨xs⟩.

says "It changed from /ks/ to /ks/". On the assumption that this is a cut-and-paste error, I'm taking out any mention of phonological change:

⟨x⟩ represented the consonant cluster /ks/, which in Old Latin which could be spelled ⟨ks⟩, ⟨cs⟩ or ⟨xs⟩.

(So was ⟨x⟩ used by itself in Old Latin?)

--Thnidu (talk) 18:09, 9 October 2012 (UTC)

Isn't there a scholarly dispute about the pronunciation of V

But isn't this actually disputed.

I know there is a camp that insists that V is pronounced w, based on a ridiculous theory about a "wine". The Germanic tribes which were the least influenced by the Romans would have a w sound from Scandinavian languages or Polish which has a special letter for slash L or and there is Runic letter for W sound. The languages closests to Latin,pronounce the V as v or sometimes b and the Church which has had the longest continuous usage of Latin says v as /v/ not /w/. On google scholar there is a book from 1880s Three pronounciations of Latin which discuss the dispute (it seems as if it is American (protestant scholars) and English who came up with v as /w/ while the continental Europe was doing the v as /v/ at least back then.

There must be some better scholarship on this. Take for example a greeting around Innsbruck Austria "servus" pronounce with v sound, purported it is a remnant of Roman influence with the v as v.68.55.60.111 (talk) 14:44, 13 September 2012 (UTC)

There is no serious dispute. It is well recognised that Latin V originally represented /w/ (and /u/). The /v/ pronunciation (along with other variants, such as in Spanish) developed later. garik (talk) 17:55, 13 September 2012 (UTC)

what is your source for no serious dispute,

Consider

%T The Pronunciation of Greek and Latin

%A Sturtevant, E.H.

%U http://books.google.com/books?id=PD8QAAAAYAAJ

%D 1920

%I University of Chicago Press

And I've read elswhere that "well before the fall of Rome (476 A. D.), the pronunciation had shifted to something much like that of modern Italian" what I am suggesting is the possibility that the whole /w/ pronunciation was bad late 19th century linguistics and more modern research is casting doubt on theory. The /w/ sound came from the North rather that this theory which leads to presumption that the /w/ sound was introduced to runic, scandinavian and polish by Latin. Why would they make up their own letter for the sound when adopting Latin alphabet if purportedly v already held that spot.

In the Netherlands, the pronunciation system used for Latin is quite similar to the the restored Classical version, but the letter v is pronounced as [v] and not as [w].

A call for some really good scholarship rather that 19th century "folk" linguistics masquerading as actual linguistics. 68.55.60.111 (talk) 12:00, 14 September 2012 (UTC)

Frances E. Lord. — The Roman Pronunciation of Latin; Why we use it and How to use it. Ginn and Company. 1894 This supports [vf] not [w]. I think it is pretty clear that the whole [w] theory is 19th century "invention" which spread and got adopted by mostly protestant American "scholars". 68.55.60.111 (talk) 12:35, 14 September 2012 (UTC)

Title Latin Or The Empire of a Sign: From the Sixteenth to the Twentieth Centuries

Author Françoise Waquet

Translated by John Howe

Edition reprint

Publisher Verso, 2003

ISBN 1859844022, 9781859844021

Describes the history of the ongoing debate between 1870 and 1960, thereafter nobody cared about Latin so it wasn't as if the debate was really settled. A lot of pseudo science in the 19th century. 68.55.60.111 (talk) 12:41, 14 September 2012 (UTC)

BTW, Polish "special letter for slash L" was ɫ even 100 years ago, so projecting w to proto-Slavic times is ridiculous. — Preceding unsigned comment added by 83.23.254.95 (talk • contribs)

I'm not aware that anyone disputes that the sound had shifted before the Fall of Rome. The point is that the Latin reflex of PIE *w was [w], which later changed. There's minor debate as to when precisely the change had been completed, but no serious scholar contends that the Latin letter v always represented [v], or that Caesar didn't say something like [ˈweːni ˈwiːdi ˈwiːki]. garik (talk) 15:23, 28 October 2012 (UTC) altered by garik (talk) 15:32, 28 October 2012 (UTC)

Bear in mind, incidentally, that this article concerns the reconstructed phonology of Classical Latin, a particular dialect spoken (probably by a minority of Latin speakers) up to roughly the third century. This would have differed from the phonology of Vulgar Latin from the same period. The modern Romance languages are descended primarily from Vulgar Latin, and other European languages influenced by Latin were influenced by both Classical and Vulgar Latin to different degrees. There was also, of course, dialectal variation within Vulgar Latin, meaning that some sound changes would have occurred later in some places than in others (or not at all). Now let's put an end to this fruitless discussion. garik (talk) 15:32, 28 October 2012 (UTC)

Umm... 68.55.60.111, I'm a Catholic and I'm fine with the classical pronunciation, it's not just protestants. In fact, I even learned it at a Catholic school. I'm not saying Ecclesiastical Pronunciation is inferior, cause it's not and you can use it if you want to. Personally, I use the Classical pronunciation when it comes to secular stuff and the Ecclesiastical when it comes to Churchly stuff. The [w] pronunciation was used in Classical times, there's evidence; and you can still use [v] if you feel like it, but you'd be using Ecclesiastical. 76.18.160.47 (talk) 20:43, 16 March 2013 (UTC)

"German Latin" pronunciation

The article currently does not address at all the significant differences in pronunciations for choral works in Latin done by Germans (e.g., Bach's Mass in B Minor) as opposed to the pronunciations for ecclesiastical Latin--how most (non-German) choral works in Latin are pronounced.Weyandt (talk) 15:13, 14 June 2013 (UTC)

More on <i> and <ae>

I suppose a great confusion is added for Anglophones when they read that <i> in relation to vowels turns into j, as they probably assume that it's the same sound as in Jones (probably which might be approximated rather by a particular mode of <c> in classical Latin?). However, in both Italian and German pronounciation of Latin, the sound is the same as in English you, i. e. the palatal approximant [j]. It might help to clear up some confusions to mention that in the article.

Second, <ae>. In both Italian, as well as Italian and German pronounciation of Latin and ecclesiastical Latin, <ae> is pronounced as the open-mid front unrounded vowel [ɛ:], which resembles German umlaut <ä> and English (particularly broad AE) <a> as in hand. Now, it occurs to me that the confusion here is due to the traditional association with the spelling <ai>, which happens to refer to the Greek diphtong <αι> that in the classical era was transliterated into Latin as <ae>, whereas before it had been represented in Latin writing as <ai>. This transition from <ai> to <ae> spelling in the same words, for instance, gives rise to the common misconception that classical caesar was somehow pronounced like modern High German Kaiser (as well as confuse people about the classical Greek pronounciation of <ai>).

My educated guess here is based upon the (sourced? see a few sections above here on the talkpage) opinion that has been voiced before above that Greek <αι> represents an [ɛɚ̯] diphtong as in AE maid, as that's exactly what <αι> happens to be pronounced as in modern Greek as well (see more reasons to assume such a pronounciation also for classical Greek in the next paragraph below). Now, how did the spelling change from <ai> to <ae> come about in Latin transliterations of Greek words from the republican to the classical era? Indeed, Old Latin spelling was still largely influenced by Etruscan, as it was via Etruscan transfer that the Greek alphabet evolved into the Latin one (see Old Italic script).

Thus, I posit that [ɛ:] and [ɛɚ̯], albeit not identical, are close enough to each other that early Romans, still influenced by Etruscan writing, may have used Latin <ai> to both transliterate Greek <αι>, *AND* represent the native Latin [ɛ:] sound that is equivalent to a.) the modern Italian pronounciation of <ae>, b.) the ecclesiastical Latin pronounciation of <ae>, c.) the AE broad <a> as in AE hand, d.) and finally, the German umlaut of <ä> which is used for German pronounciation of Latin <ae>. Then, in the classical period Roman writers left their Greek and Etruscan legacy spelling rules behind to now represent their native Latin [ɛ:] by the genuine Latin spelling of <ae>, *AND* at the same time also simplify transliteration of Greek <αι> into <ae>, for [ɛ:], albeit not entirely identical, was their most approximate native sound to [ɛɚ̯]. --79.193.55.150 (talk) 01:08, 11 August 2011 (UTC)

"[ɛɚ̯] diphtong as in AE maid " ?!
IPA ‹ɚ› is a rhotacized schwa, as in American English "butter". I'm not sure what sound you're trying to talk about, but it sure ain't that one. (BTW, it's ‹diphthong›, < Gk. δίϕθογγος.)

--Thnidu (talk) 17:28, 9 October 2012 (UTC)

That is <. -DePiep (talk) 21:53, 9 October 2012 (UTC)

The IP is wrong about the classical pronunciation of ae and Caesar: ae was indeed pronounced [ae] in Classical Latin, much like English eye, and Caesar was pronounced [kaesar], much like German Kaiser. This would at least be the educated urban pronunciation. In rural speech, the monophthongisation to [ɛː] is already found in the Old Latin period, in the 2nd century BC, probably under Umbrian influence, although it is only by the 1st century AD that the monophthongal pronunciation is apparently fully established in popular speech (as attested by the Pompejan inscriptions). Proto-Romance also must have had the monophthongal pronunciation, but it was certainly considered nonstandard (rustic or provincial) by the urban elites. --Florian Blaschke (talk) 14:19, 5 January 2014 (UTC)

@Florian Blaschke:, if you can find sources for this, could you correct the article? It currently says ae was /aj/, which seems a bit unlikely to me as well. CodeCat (talk) 00:02, 14 February 2014 (UTC)

My source for this is Meiser's Historische Laut- und Formenlehre des Lateinischen, but the article does not substantially contradict his account. The pronunciation in the Archaic/Old Latin period was /aj/, in the Classical period /ae/ (with lowered second element) and in the post-Classical period it was a monophthong /ɛː/. --Florian Blaschke (talk) 00:12, 14 February 2014 (UTC)

By the way, I suspect that the IP intended to write [ɛɪ̯̯] or [ɛe̯̯], considering the reference to American English maid. However, the IP is wrong not only about Latin, but also about Classical as well as Modern Greek – the diphthong is not used in, or reconstructed for, any of these languages, except possibly in certain modern pronunciation traditions. --Florian Blaschke (talk) 00:25, 14 February 2014 (UTC)

Some criticism...

1. The article assigns /r/ a value [ɾ] or [r]. This seems acceptable for /r-/ in initial position (just the situation for /r/ in Spanish, Italian, Catalan and South Occitan). But the transcription in one example is [arma] for arma, but in the cosa /-r/ in no Romance language is [-r] but [-ɾ]. It is likely that the single /r/ in Latin has the same allophonic distribution that these modern Romance languages: initial [r-/ɾ-], medial [-ɾ-] and final [-ɾ]. The realization of double /rr/ is another difficult point.
2. About diphthongs ae and oe the article proposes [aj/ai̯] and [oj/oi̯], but phonetically [ae̯] and [oe̯] are perfectly possible. Given the spelling it is probable that the pronunciation was this latter pronunciation. Remember that previous /oi, ou/ dihpthongs evolved to /ū/ (*oinos > ūnus 'one', *lou(k)sna > lūna 'moon'), this suggest that /oe/ and the previous /*oi/ have had different pronunciations. --Davius (talk) 23:49, 4 March 2014 (UTC)

anglicised pronunciation of <ae>

Article currently says: "Latin words in common use in English are generally fully assimilated into the English sound system, with little to mark them as foreign, for example, cranium, saliva. Other words have a stronger Latin feel to them, usually because of spelling features such as the digraphs 〈ae〉 and 〈oe〉 (occasionally written as ligatures: 〈æ〉 and 〈œ〉, respectively), which both denote /iː/ in English. In the Oxford style, 〈ae〉 represents /eɪ/, in formulae, for example"

Surely this isn't true! I don't know that there is a single reference that is "the Oxford style", but I checked my copy of the Oxford English Dictionary, and it says "/i:/ in all positions". Of course, there is a growing tendency to pronounce <ae> as /eɪ/ thanks to confusion with loans from other languages -- notice, for instance, that /eɪ/ is far more common in vertebrae than in formulae, which must be due to the visual similarity with brae -- but thankfully this is not yet standard. 2.25.115.179 (talk) 22:14, 18 March 2014 (UTC)

church latin and i in hiatus

Someone has been applying Italian pronunciation too literally in thinking that i in hiatus is /j/. It is always sung as a separate syllable, and I venture this is clear evidence of its proper pronunciation as such. Benwing (talk) 21:10, 16 April 2014 (UTC)

Also, I question whether the alternation between [ɛ] and [e] as given here is real. I have read that common medieval pronunciation always had [ɛ]. In any case the usage isn't consistent in the pronunciation given, e.g. for gentium which probably should have [ɛ]. Benwing (talk) 21:12, 16 April 2014 (UTC)

Take a look at this link [5] which verifies both of these statements as well as the fact that the /ts/ in a word like gratia is not lengthened (/ˈgra-tsi-a/). Benwing (talk) 21:20, 16 April 2014 (UTC)

You left the /ts/'s long. (I'm not going to fix, I don't know anything about this.)

Your ref is a bit suspicious, though. It's been dumbed down, which allows the possibility that they've left out distinctions in the interest of accessibility. — kwami (talk) 02:52, 17 April 2014 (UTC)

Evidence for Latin Vowel-Sounds?

Right now the article states that the reconstructions of Latin vowel-sounds are based on Romance vowel-sounds. I understand that this is one important source, but afaik Latin transcriptions of Greek works, and Greek, Gothic, Syriac, etc. transcriptions of Latin words are together another important source. 71.191.58.196 (talk) 04:24, 14 October 2014 (UTC)

simple questions

Wondering why the ae in trōiae [ˈtroːj.jă.e] is not a diphthong, and why Gāius [ˈɡaː.i.us] is three syllables. Details like this aren't covered that I see. kwami (talk) 09:21, 27 June 2009 (UTC)

Where do you get the information that "Troiae" is three syllables? I'm surprised by that. I've included information about "Gaius", although I haven't explained "why", but I guess the answer is just "because" :P LjL (talk) 13:40, 27 June 2009 (UTC)

The transcription in this article has three syllables for trōiae. Perhaps it's just an error? I'll adjust the transcription; please correct me if I get it wrong. (Someone noted the same problem below.)

As for gāius, I just wanted to make sure that it wasn't an error too. Given that reicit is [ˈrejjikit], might gāius be something like [ɡaːjjijjus]? kwami (talk) 18:23, 27 June 2009 (UTC)

Where in the article does it show Troiae with three syllables? In the example IPA transcription from the Aeneid it doesn't divide anything in syllables, and I didn't see it anywhere else...

As for Gaius, I don't have sources, but I think there is no reason to even remotely suspect it might be [ɡaːjjijjus]; it's simply that the "i" is being used as a plain vowel, but normal Latin orthography (without "j") has no way to distinguish these two usages. With "reicit" one is simply assuming it's equivalent to "rejicit" because of the etymology, re+ieci. I think I stumbled upon a source for that today but I passed it by without noting it down.

LjL (talk) 20:53, 27 June 2009 (UTC)

Trōiae was transcribed [ˈtroːjjăe], which is three syllables because it has three vowels: [oː], [ă], and [e]. kwami (talk) 21:52, 27 June 2009 (UTC)

That was simply a mistake. No one is seriously arguing that the ending "-ae" should be disyllabic. --Alatius (talk) 10:58, 30 August 2009 (UTC)

This whole discussion puzzles me, as neither Troiae nor Gaius are three syllables - in Ecclesiastical or Medieval Latin. In both cases, the "i" has become a glide like the consonantal "y" in English. Would it be appropriate to mention this in the section that shows "Gaius" as three syllables ("Ga -i-us")? Otherwise, those of us who grew up speaking Church Latin might wonder what this was all about. —Preceding unsigned comment added by Mccalpin (talk • contribs) 20:09, 11 August 2009 (UTC)

For that matter, I don't think the trisyllabic pronunciation Gā-i-us is very common amongst speakers who use the reconstructed classical pronunciation either, either because they are not aware of the original pronunciation or because they don't bother with such minute pedantry. But that section doesn't deal with the pronunciation used today. --Alatius (talk) 10:58, 30 August 2009 (UTC)

Not sure about Gāius, but the very similar māior and māius, and also cuius, huius etc., actually have a geminate /jj/ – resulting in a heavy syllable, which is what the macron awkwardly and misleadingly tries to indicate. Etymologically, /jj/ in these words goes back to */gj/ and */sj/ respectively. --Florian Blaschke (talk) 00:33, 14 February 2014 (UTC)

I'm surprised no one has answered the second question yet. In the Lewis and Short entry on Gāĭus it explains that the word is trisyllabic in several poems, including Catullus 10. Here, I understand (though I have never read Catullus and had to quick look up hendecasyllable) that the meter requires the word to contain a long vowel followed by two short vowels. (Take note that the linked text is a scansion, and the macrons and breves in it mark the syllable weight, not vowel length.) Therefore, Gaius as pronounced by Romans must have been Gāĭŭs /gaː.i.us/, not Gajus /gaj.jus/, and it doesn't rhyme with maius /maj.jus/. — Eru·tuon 23:58, 11 December 2014 (UTC)

diphthongs

Please look at [this]. --Espoo (talk) 12:21, 16 December 2014 (UTC)

Nasal + f sequences

The article currently states that Latin turned sequences of vowel + nasal consonant into nasal vowels before fricatives. These then became long non-nasal vowels in the development to Vulgar Latin. But while there are plenty of cases with -s-, what about -f-? Most sources seem to write īnfāns with the length marks suggesting nasalisation, but the development > Old French enfes seems to suggest that this sequence was not affected the same way. Instead of becoming a long vowel, the development is as a short vowel with the usual lowering, and the nasal consonant is preserved. So I wonder, were these sequences nasalised in the same way in earlier Latin? CodeCat (talk) 14:55, 22 November 2014 (UTC)

Sidney Allen describes the situation with vowels + ns, nf on pp. 28-30 (the page on the letter n) and pp. 65-66 (on vowel length). It seems he indicates the vowels were pronounced in various different ways from Old Latin to the Vulgar Latin period: as long and nasalized, simply long, and as short vowels with the nasal consonant intact (likely from analogy with other forms). He suggests that the latter case was true for the Vulgar Latin form of the words infans and insignem that developed into French enfes and enseigne. This was also true for consilium > conseil, but not for constare > coûter. — Eru·tuon 02:49, 22 December 2014 (UTC)

I think this is a matter of analogical restoration. The problem is that in native/inherited Latin words, f only occurs word-initially, so all examples of mf/nf are due to prefixation and thus susceptible to analogical restoration. Are there any cases in the Romance languages where the nasal indeed disappears before f? CodeCat (talk) 22:30, 22 December 2014 (UTC)

Allen doesn't list any. I searched on Wiktionary in categories of Old French words and found nothing; I searched Spanish and came across cofrade, but I don't know if this was derived from a form in Vulgar Latin or a new coining. Allen does list forms like cofeci, iferos used in Latin inscriptions, but these are not carried over into French (confire, enfers). — Eru·tuon 22:53, 22 December 2014 (UTC)

Template for epigraphic Latin style

I've created {{Script/Latin}} to specify fonts that include the tall ꟾ. At the moment this template only includes font-family styling. It would make sense to create a Latin-language template including both this font-family specification and smallcaps style, and replace {{smallcaps|text}} throughout this article with that template. Perhaps it would be more appropriate to use {{lang}} or a related template for this, though. What do others think? — Eru·tuon 21:58, 10 December 2014 (UTC)

I've removed {{Script/Latin}} because I realized it refers to the Latin alphabet in general, rather than the alphabet used for the Latin language. So, another template is needed to fulfill the same purpose. — Eru·tuon 04:15, 11 December 2014 (UTC)

I created {{Latin-epigr}} to replace both the {{smallcaps}} and {{unicode}} templates encircling Latin text. I've replaced {{smallcaps}} and {{Unicode}} in the Classical Latin example with this new template. The i longa character looks strange, because it's twice as high as the surrounding characters. This is because (I assume) it is unaffected by the CSS font-variant: small-caps; property:

• qvꟾ·prꟾmvs·abórꟾs

But this is probably no different from what editors with proper font-selecting software are already experiencing, so if no one objects, I will replace all the instances of {{smallcaps}} and {{Unicode}} encircling Latin text throughout the article with {{Latin-epigr}}. — Eru·tuon 01:20, 12 December 2014 (UTC)

A solution to the problem with i longa: use CSS text-transform: caps; and font-size: 73%;. This gives the same effect as font-variant: small-caps;, except it applies to i longa as well as the other letters. — Eru·tuon 03:04, 24 December 2014 (UTC)

Final vowel + n

The article currently says that final vowel + nasal sequences were nasalized. Allen says that final vowel + m sequences are nasalized, but he does not mention anything about final vowel + n sequences. The reference currently given is Clackson's chapter on Latin in Ancient Languages of Europe, which I don't have access to; can anyone confirm whether this source indeed says that -Vn sequences are nasalized in addition to -Vm sequences?

-Vn is rarer in Latin than in Greek. If I recall right, the main examples of it are third-declension neuters with stems in -n, like nomen, and various un-assimilated forms in Greek loanwords. — Eru·tuon 03:00, 2 January 2015 (UTC)

This is what Clackson has to say: "In Classical Latin there was also a series of nasalized vowels ... which were restricted in occurrence to (i) word-final position, where in the standard orthography they are written im, em, am, om, um; or (ii) before a sequence of nasal + continuant." Thus, no mention of nasalization of vowels before final -n, and rightly so: the most obvious evidence for -im, -em etc. representing nasalized vowels is the fact that they can be elided in poetry; however, no such thing ever happens with final -n, which not only occurs in third-declension neuters, but also in the common word non. —Alatius (talk) 07:37, 2 January 2015 (UTC)

Thanks! That description agrees with Allen. Strange — someone must've misread Clackson, or modified the text without reading the source. I'll correct it. — Eru·tuon 08:05, 2 January 2015 (UTC)

non is a single-syllable word, though, so it's not surprising that its -n survives in Romance. Compare rem, which retains the nasal in several Romance languages as well. Are there any other final -n words that may have survived into Romance? I know there's the -men nouns, but as they are neuters they would have likely been reformed as masculines at some point, giving -minem > -mene > (medial contraction) -mne. CodeCat (talk) 15:23, 2 January 2015 (UTC)

Quality of nasalized vowels

There is one question that Sidney Allen does not address: the quality of nasalized vowels. For instance, was -um pronounced with the quality of long u or short u? The reflex of -um in the Romance languages was -o, like that of short u. On the other hand, en in mensem became e in Italian mese, the same reflex as long e. So perhaps close nasalized vowels were pronounced like the corresponding short vowels (that is, near-close) and mid nasalized vowels were pronounced close-mid (so that the two were similar, like short u and long o). On the other hand, insulam became Italian isola, not esola, reflecting the quality of long i. Perhaps stressed nasalized vowels were pronounced with the quality of long vowels, and unstressed with the quality of short vowels, or perhaps unstressed ones eventually became short, merging with short vowels in Vulgar Latin. Anyone know anything about this? — Eru·tuon 18:50, 24 December 2014 (UTC)

Not all Romance languages have -o, some have or had -u which can be distinguished from -o (from -ō). See also Romance languages#Unstressed vowels. CodeCat (talk) 20:50, 24 December 2014 (UTC)

There's also Asturian which has a distinction between masculine -u and "neuter" (for mass nouns) -o, but it's not entirely clear how that relates to Latin. I think that the -u ending may continue -um, while -o continues -ud (neuter of the pronouns and articles). CodeCat (talk) 17:12, 25 December 2014 (UTC)

Interesting. So the evidence from the Romance languages on the quality of the nasal vowels is pretty ambiguous. I'm not sure what to think then. — Eru·tuon 03:18, 2 January 2015 (UTC)

I don't know if it's all that ambiguous. It may be that most Romance languages simply underwent a change of final -u to -o at some point. This could be checked by looking for other possible sources of final -u in Romance, which would have to originate from Latin -ū (possibly with a consonant following). If those also show up as -o, then it's most likely a general sound change. At the very least, it seems that the 4th declension plural -ūs may be a candidate. CodeCat (talk) 15:27, 2 January 2015 (UTC)

Z, di, and dz

Allen indicates that Koine Greek zeta was pronounced /z/ or /zz/ while loanwords with this sound were being borrowed into Latin, and that this is what ⟨z⟩ represented in Classical Latin. However, the article seems to indicate that Edgar Sturtevant's Pronunciation of Greek and Latin says that zeta was represented with di in Latin, indicating that an affricate /dz/ was used for zeta in Italian Greek. I don't have access to the book; could someone who has access to it confirm that this is what Sturtevant says, and which period of Latin he is referring to — whether he disagrees with Allen about the pronunciation of zeta during the Classical Latin period? The article on Z seems to indicate that the alternation between z and di occurred in Late or Vulgar Latin, and if so, this should be noted in the article. — Eru·tuon 21:55, 2 January 2015 (UTC)

I found Sturtevant online, and he seems to indicate on pp. 115-117 that zeta was usually pronounced /z/ in Italian Greek, but /dz/ must have occurred as well, given certain spellings of Greek words (I assume from both the Old Latin and Classical periods). He quotes a passage by Velius Longus that says that the pronunciation /zd/ was used in Doric, but that in Latin the sound was not a consonant cluster — it ended with the same sound with which it began. — Eru·tuon 23:24, 2 January 2015 (UTC)

Actually, I misread the source. The evidence for /dz/ is based on spellings with, on the one hand, ⟨z⟩ for ⟨dz⟩ from earlier ⟨di⟩ or ⟨d⟩ before ⟨e⟩, and ⟨di⟩ or ⟨dz⟩ for earlier ⟨z⟩. One spelling, ⟨diaeta⟩, seems to be Late or Vulgar Latin, or at least a colloquial form where ⟨ae⟩ was pronounced [ɛ]. I guess the argument is that the pronunciation /dz/ must have remained in Greek till Classical Latin or later for these spellings to occur. — Eru·tuon 00:04, 3 January 2015 (UTC)

I remember reading somewhere on Wikipedia that, in Late Vulgar Latin, when the affrication of /tj/ and /dj/ started to set in, the spelling -z- was sometimes used to represent former -di-. This still remains in many Romance languages of course, but if my memory is correct, there were some spelling errors of that kind in the imperial period or early post-imperial period already. CodeCat (talk) 00:54, 3 January 2015 (UTC)

Yeah, that's discussed in Z. Nothing on it in Vulgar Latin, which is a disappointing omission. — Eru·tuon 07:02, 3 January 2015 (UTC)

Etruscan's influence on Latin pronunciation

Go to History of Latin § Phonological influence from Etruscan if you know of a source that discusses Etruscan's influence on Latin, or on the possibility that Latin developed vowel reduction by language contact with Etruscan. — Eru·tuon 05:05, 11 February 2015 (UTC)

You mean talk:History of Latin § Phonological influence from Etruscan, of course. — Sebastian 02:33, 18 March 2015 (UTC)

Oops. Quite right. Thanks. — Eru·tuon 02:43, 18 March 2015 (UTC)

Small caps, once again

I posted in WikiProject Latin on small caps in this article. If you've got an opinion on whether they should be used or not, head over there. — Eru·tuon 04:41, 4 January 2015 (UTC)

I replaced {{sm}} and {{unicode}} throughout the article with {{sqc}}. I described this template in Wikipedia talk:WikiProject Latin § Small caps:

The template {{Sqc}} stands for "square capitals", as in Roman square capitals, the letterforms used in Roman inscriptions. It was originally {{Latin-epigr}}, but I thought an abbreviated title would be best, if this template is to be used for almost all Latin text in Latin spelling and pronunciation. Currently the inline CSS of the template specifies fonts containing i longa, and converts text to uppercase and makes it 73% height, which seems to be x-height in some fonts at least. This has to be used as a replacement for the font-variant: smallcaps; property, which does not appropriately convert i longa to small caps. I've also added lang="la" xml:lang="la".

If there are any objections to my adding this template, please voice them here.

If you don't want Latin text in this article to display in small caps, you can add some code to your common.css:

:lang(la) { text-transform: none !important; }

The !important at the end of the CSS attribute is necessary, in order to overrule the inline CSS used in {{sqc}}. To add other CSS attributes, you must put !important after each. — Eru·tuon 20:45, 18 January 2015 (UTC)

Also, input requested: Do you know of fonts that have i longa (ꟾ)? Add them to the inline CSS in {{sqc}} if so. — Eru·tuon 20:51, 18 January 2015 (UTC)

• At least these fonts has i longa (U+A7FE):
  • DejaVu Sans
  • DejaVu Serif
  • Everson Mono
  • FreeSerif
  • unifont

--Lüboslóv Yęzýkin (talk) 23:04, 19 January 2015 (UTC)

As discussed on the WikiProject Latin talk page some time ago (see the link at the top of this section), we need to change Roman square capitals with I, V, apices, and i longa to regular modern Latin orthography with i and u and macrons. Classical-era orthography should be compared with modern orthography in a table. This is a big change, and may be easiest done with AWB. And maybe the table should be created first. — Eru·tuon 07:59, 21 March 2015 (UTC)

I just read the discussion you refer to and it seems the consensus there is to keep the small caps. — Preceding unsigned comment added by 82.139.82.82 (talk) 03:10, 17 July 2015 (UTC)

Table of orthography

I've proposed splitting Help:IPA for Latin into two articles, one on Classical and one on Ecclesiastical pronunciation. To allow the split, we need a table of orthography here. I'll create the table below. — Eru·tuon 20:48, 23 March 2015 (UTC)

Letter	Environment	International Phonetic Alphabet		Examples	English approximation
		Class.	Eccl.
b	in most cases	b
	before s or t	p
c	in most cases	k		[] Error: {{Lang}}: no text (help)
	before e, ae, oe	k	tʃ	[] Error: {{Lang}}: no text (help)
d		d		[] Error: {{Lang}}: no text (help)

I believe you might have swapped the Classical en Ecclesiastic pronunciation. The reason I think this is that to the best of my knowledge the c still was k everywhere in Classical Latin, and the shift to tʃ in certain positions occurred in Vulgar Latin and Romance, and I thought the Ecclesiastic Latin pronunciation derived from Romance, particularly Italian. — Preceding unsigned comment added by 82.139.82.82 (talk) 03:05, 17 July 2015 (UTC)

You're quite right. I fixed it. — Eru·tuon 17:55, 19 July 2015 (UTC)