Eoiní Mhaidhc Ó Súilleabháin read all the phonemes of Cork Irish for me recently, and they are at http://www.corkirish.com/wordpress/pronunciationintonation . I would like to discuss some aspects of Muskerry pronunciation and the transcription of it in Brian Ó Cuív's The Irish of West Muskerry.

1. Vowels - there is little indication of change here, apart from the diphthong /ia/ tending to unify with the diphthong /iə/ in younger speakers (including Eoiní). So the word "éan" was traditionally /ian/, but has become /iən/, rhyming fully with "dian" now.

2. The issue in the vowels is the diphthong /ou/, which has unified with /au/ in the speech of many younger speakers. So, whereas the word "leabhar" is /Lʹaur/ in the Connemara, it is meant to be /lʹour/ in Muskerry, but often/usually sounds like /lʹaur/. I am not sure it has fully unified: I think "togha" is still /tou/ and "Ultach" is still "oulhəx" or "oultəx".

3. Consonants - I think bilabial f /ɸ/ and v /β/ have definitely gone. Eoiní had a clear labiodental pronunciation, ie /f/ and /v/ (or /w/ in certain cases).

4. /j/ - this is straighforward - like an English "y" in most cases, but I have been trying to learn how to pronounce slender dh/gh brefore r, l and n. I don't think these are exactly like an English "y" and so the transcription /j/ is a little basic. It has been suggested the sound is realised in these circumstances as /ʝ/.

However, I think the IPA is weak in its representation of palatal consonants - we have previously discussed with Lughaidh's help the use of symbols such as /c/ and /ɟ/ to represent slender c and g in Irish, despite the fact that these symbols are also used to represent the quite different sounds ty and gy in Hungarian: it seems the problem is that the IPA does not distinguish between front-palatal symbols (as would be required for Hungarian) and back-palatal symbols as would be required for Irish.

The problem is that the use of /ʝ/ for slender dh/gh before l, n and r needs to be qualified: it is definitely a back-palatal sound. You could say that as /g/ is to /ɣ/, so /gʹ/ [or /ɟ/ as it can be written] is to /ʝ/. Now, I know Lughaidh will have my guts for garters for saying this, but this /ʝ/ before l, n and r sounds very similar to the French r when the latter is realised as a uvular approximant and not a uvular fricative. Maybe Lughaidh can comment on these symbols? The Muskerry slender dh/gh before l, n and r is not a y-sound, but is produced in exactly the same place as the broad dh/bh: maybe /j/ and /ʝ/ should be abandoned for this sound in favour of /ɣʲ/, which gets it exactly.

Try saying: mo dhriotháir, an ghrian, sa ghleann - the audio file at http://www.corkirish.com/pronunciation/j.mp3 should clarify what I am on about. The words Eoiní is saying there are: a Dhia, ana-dheas, deá-ghníomh, do ghléas, sa ghleann, mo ghreim, an ghrian, mo dhriotháir.

5. Similarly, /xʹ/ is a little problematic. It can be realised as /ç/, but like all palatal symbols, the IPA is unclear here. /ç/ can be used for the h+y sound for the English word "human", but that is a front-palatal sound, and the Irish /xʹ/ is a back-palatal sound. That is why the word "chím" sounds very lightly lenited, almost like "cím". Put it this way, it is nearer to "cím" than to "sím", whereas the sound in "human" is closer to "shooman" than to "kuman", and that is why the IPA symbol for this Irish sound is ambiguous to English speakers. Maybe /xʹ/ and /ç/ should be abandoned for this sound in favour of /xʲ/, which is what it is, as it is produced in the same place as the broad ch.

Listen to http://www.corkirish.com/pronunciation/slender%20ch.mp3 , where Eoiní is saying: chím, a Sheáin, le chéile, sa chliabhán, ana-chneasta, do chreid sé, sa chré.

First comment, looks like you mix up phonetics and phonology. You write all your transcriptions between slashes ; phonetic realisations of phonemes have to be written between square brackets.

For instance:

quote:

Similarly, /xʹ/ is a little problematic. It can be realised as /ç/,

should be :

"Similarly, /xʹ/ is a little problematic. It can be realised as [ç],"

otherwise it doesn't mean anything, a phoneme isn't realised as another phoneme...

quote:

Now, I know Lughaidh will have my guts for garters for saying this, but this /ʝ/ before l, n and r sounds very similar to the French r when the latter is realised as a uvular approximant and not a uvular fricative.

Actually I know how [ʝ] sounds like and it doesn't sound at all like any French r... I'd say you misheard either French r's or ʝ or you heard someone who didn't pronounce them properly.
[ʝ] is the voiced counterpart of [ç]. They are pronounced much further forth than the French r's.
I can record these sounds for you if you like, you'll see the difference.

1. Both diphthongs /ia/ and /iə/ appear to have been preserved in Corca Dhuibhne. See Ó Sé.

2. Ó Sé gives only /ou/ with two allophones.

3. In Corca Dhuibhne both bilabial and labiodental /f/ and /f´/ and /v/ and /v´/ are to be found. The labiodental is in all probability a relatively new phenomenon.

4. I agree with you that Ó Cuív’s use of /j/ in all situations is misleading. According to Ó Sé, [ʝ], a voiced fricative produced at the hard palate occurs before /i:/, /e:/ or a sonant. The half-vowel [j] normally occurring before /a/, /a:/, /o/, /o:/ and sometimes before other vowels apart from /i:/. (§ 9)

I presume by front-palatals you are referring to articulation with the alveolar ridge, or front teeth, and back-palatal using the soft palate at the back of the mouth?

5. According to Ó Sé, /x/ is produced at the soft palate, /x´/ at the hard palate, so not produced at the same place in the mouth.

Unfortunately, I can’t listen to the sound files on your blog as in work such practices are automatically blocked and my internet link at home has been acting up for a while now. I know that Corca Dhuibhne and Muskerry are two different places but I imagine the pronunciation in both places is not a million miles away from the other.

A good idea would be to check out Raidió na Gaeltachta's archives for interviews with older speakers. I know An Saol ó Dheas used to have a feature whereby they broadcast interviews with various speakers made twenty or thirty years ago. Have a look at their website for more info.

I could be wrong in my theory of the pronunciation of slender dh before r, but to understand what I am trying to say, people could listen to a 6 minutes MP3 of me explaining it at http://www.corkirish.com/audio/musings.mp3

You have a good ear for sounds! I think that you are right and it could be, judging by your recordings, that ch, gh + slender liquids l, r give rise to palatalised velar fricatives rather than palatal fricatives. Another possibility could be that your subject has plain velar fricatives in these clusters, despite the fact that the following l,r are slender.

The first thing that comes to my mind is the shift in palatalisation in clusters, similar to the one I'm hinting at in your case, from the older Moscow pronunciation to the modern one. For example, дверь used to be pronounced with a /d'v'/ cluster word-initially, note that both consonants are palatalised; nowadays however it is standard to say it with /dv'/ (only the fricative is palatalised) although you still here the former version on the Moscow underground :)

To David:

"The Muskerry slender dh/gh before l, n and r is not a y-sound, but is produced in exactly the same place as the broad dh/bh: maybe /j/ and /ʝ/ should be abandoned for this sound in favour of /ɣʲ/, which gets it exactly."

- Exactly, there is definite friction, and yes /ɣʲ/ gets it exactly. It is possible though that the sound may have slightly less friction with some speakers. Try and say "ghreim" and reduce the friction to a minimum. It's just about possible to use a /j/ with little discernible friction. I agree with Lughaidh. The French r is a different animal I think, produced much further back. French r's are uvular if I'm not mistaken? A French style r must have occured in the Irish of parts of west Waterford as one can definitely hear it in the English of places like Annestown and Bunmahon on the south Waterford coast, west of Tramore and east of Dungarvan.

To be fair to Ó Cuív, though, he does call the second allophone he mentions, the one before consonants as a palato-velar fricative, which seems an accurate description.

Peter, it is difficult to get slow enough recordings to really examine it in detail, but I think "mo dhriothár" is such a vital phrase that I need to be able to say it in Irish! yes, you are right, the speaker may have had plain velar fricatives - as long as people in the Gaeltacht can understand you, they don't pore over your pronunciation, baochas le Dia!!

Пётр, я учился русскому языку в университете, но это было много лет назад - я был в Москве в 1994 году, и мне сильно понравилось Московское Метро имени Ленина! Я прожил 3 месяца в Воронеже, тоже в 1994 г. Да, хорошо помню как говорили на метро "Осторожно, двери закрываются!"