I still don’t have full access to Ó Cuív’s
Irish of West Muskerry, but I’ve got
Gaeilge Chorca Dhuibhne by Diarmuid Ó Sé, and that’s what he writes about a-like vowels (thankfully he does use full IPA to describe the sounds as accurately as possible):
Gaeilge Chorca Dhuibhne, Na gutaí gearra, § 13, p. 21 wrote:
a Guta íseal, liopaí neodrach. Nuair is consain chaola amháin a bhíonn in aice leis bíonn sé timpeall ar Ghuta Cairdineálta 4 [a], nó beagán níos airde, m.sh. g´ar´ɪd´ gairid, at´ ait, f´a feadh. Nuair a bhíonn sé idir consan caol agus consan leathan (pé acu ord) bíonn sé beagán siar [a̠] ó Ghuta Cairdineálta 4, m.sh. f´ar fear, ʃas seas, kat´ɪ caite, gan´ɪv´ gainimh, sdal´ stail. Bíonn sé níos faide siar fós [ä] i ndiaidh consan leathan liopach nó l, m.sh. bal´ɪ baile, mad´ən maidean, far´ɪ faire, lat´ loit. Tá rogha idir cáilíocht thosaigh agus cáilíocht chúil nuair a bhíonn sé idir consan caol agus x i bhfocail mar ʃaxd seacht, t´axd teacht (ach cáilíocht thosaigh in k´r´ax creach de bharr an r´). Nuair is consain leathana amháin a bhíonn in aice leis bíonn sé ina ghuta íseal idir Guta Cairdineálta 4 agus Guta Cairdineálta 5. Cáilíocht leath slí eatarthu nó mar sin a bhíonn ag go leor cainteoirí i bhfocail mar mak mac, abɪr´ abair, da dath, ach tá cáilíocht [ɑ̈], timpeall an triú cuid den tslí chun tosaigh ar Ghuta Cairdineálta 5, an-choitianta chomh maith. Nuair a chuirtear faid leis an nguta seo is féidir leis a bheith chun tosaigh go maith, m.sh. san intriacht sd[äː]d stad!
So what he’s saying is that it is [a] or a bit raised [a] (which, fair enough, means
very close to [æ]) only in contact with only slender consonants (/g’ar’id’/, /at’/). But when in contact with a broad consonant, it is even further back than [a], namely [a̠], in eg.
fear /f’ar/, and it can be even further back [ä] in contact with /x/, eg.
teacht /t’axt/ – in those words, acc. to Ó Sé, in Dingle Irish it’s pretty far from [æ]. [a] or even central [ä] seems to be more accurate transcription of the vowel.
Between broad consonants it’s also not really [ɑ], but rather a bit more front vowel, so perhaps [ä̠], [ɑ̟], or [ɑ̈] (only the third given by Ó Sé) is a more precise transcription.
For the long vowel, he gives [ɑː] (
timpeall ar Ghuta Cairdineálta 5) and rounded to [ɒː] when next to labial sounds (eg. in
bán, or
bó bhán transcribed as
boː [wɒː]
n), some speakers even use here [ɔː] (
Bíonn an cháilíocht chruinn seo, ag druidim i dtreo [ɔː], ag roinnt cainteoirí go minic). He notes that between slender consonants more fronted [aː] or [a̠ː] happen eg. in
a Sheáin, but always [ɑː] in
Meáig.
So, from
Gaeilge Chorca Dhuibhne, from what Google let me browse in Ó Cuív’s
The Irish of West Muskerry, from Doyle’s and Gussman’s
An Ghaeilge, and from Wikipedia vowel chart of Munster Irish, it seems to me that:
Either Munster Irish has its a-like vowels a bit retracted compared to Connacht Irish and doesn’t really use [æ] anywhere (or perhaps only as a rare allophone between two slender consonants), and the typical sound in words like
fear,
teacht,
bean, etc. is rather more central and open [a ~ ä]. And Ó Cuív’s and Doyle’s transcription of the vowel as /a/ when in contact with a slender consonant and /ɑ/ between broad ones is a good phonological approximation (but Ó Sé uses /a/ everywhere, since it is a single phoneme with no minimal pairs between /a/ and /ɑ/). The long /ɑː/ is more retracted than short /ɑ/ and might be rounded next to labial sounds.
Or researchers of Munster Irish can’t hear the vowel [æ] and consistently deny its existence (except for one small remark that [a] in
gairid is
[a] nó beagán níos airde).