Sunday 18 April 2021

Pronunciation Quirks

Here are some quirks of pronunciation. Because there such little audio available to me, the very finest details of pronunciation are impossible to know just yet. This is by no means comprehensive but just a few distinctive features to remember.

-Slender R
The slender r varies across Gaelic dialects but Ossory is unique in having [ʒ] for /r´/. This sound is essentially the same as the g in beige or j in the French word je. This same sound is also found in the dialects of Oileán Chléire and East Galway as an eclipsed/urú form of slender s /s´/. I should also say that this is incredibly similar to the slender r found in speakers particularly from the Dún Chaocháin peninsula in Erris, Co. Mayo.

From the snippet of a recording that I posted earlier, I would say that this sound is produced with the blade of the tongue pressed flatly on the alveolar ridge behind the teeth.

-Slender NN as NG
Just like in Ring, Co. Waterford, slender fortis n /N´/ merged with /ŋ´/ rather than /n´/. This is seen in the words Rinn, linn, tinn or éinne which are pronounced with /ŋ´/ and are often spelled Ring, ling, etc.

-Raised vowels
Many vowels are raised in the presence of a nasal consonant. This is a common shift in many languages. /o:/ often becomes /u:/ such as in the words mór, feirmeoirtóin etc.

The shift from /a:/ (or /ɑ:/) to /o:/ beside nasals was not something either Pádraig de Paor or 'W'* had but Matthew Byrne did have it such as in the words amháin /ə'vo:n´/, lámh /lɔ:/ and lán /lɔ:n/. Remember that Wagner's transcription is much narrower. This shift was also found in Clare/Thomond.

-ai to aoi
Where the diphthong -ai- is written the vowel is often /i:/ such as in the words caint, crainn, saill, strainséir etc. These words were frequently written with the vowel as -aoi- so caoint /ki:n´t´/, craoing*2 /kri:ŋ´/, saoill /si:l´/, straoinséir /stri:n´s´e:r´/.

-Stress
Stress is generally strong on the final syllable.

-Word final -amh
Final -amh and -(a)bh varies generally and could be either /ə/ and /əv/. Pádraig de Paor had both of these and spelling evidence suggests both too with it likely varying from speaker to speaker and region to region. tala, cúna and leana are attested forms of talamh, cúna and leanbh.

-Word final -th and -ch
When th comes at the end of a word, it is often pronounced /x/ and written ch such as in the words blách for bláth /bla:x/, leach for leath /l´ax/ etc. Sometimes the opposite is done where ch at the end of a word becomes /h/ or is elided. This is seen with gach aon which was /ga he:n/. This is used today in Ring and often spelled ga haon.

-Devoicing of final consonant
Like in Connacht and Ulster, the final consonant is often devoiced. This is mainly found in words ending in a g changing to c in those two provinces however this is done more liberally in Kilkenny with words such as eilic /el´ik´/ for uilig etc. This is never seen with Gaeilig /ge:l´ig´/ which seems to always retains its voiced g.

*Of the three informants referenced in Wagner's Kilkenny material for points 6 and 6a in LASID, there's Matthew Byrne from Tallowbran (point 6), Pádraig de Paor (6a) and a mysterious third informant listed only as 'W' (also 6a).

*2 This spelling is unattested but the pronunciation is taken from LASID Vol. I, page 248. This also works as an example of the second pronunciation feature.

Friday 16 April 2021

At long last, audio of Pádraig de Paor

Here we have a tiny clip of Pádraig de Paor. From speaking with people who are interested in or are studying Gaelic dialects with few or no living speakers, I understand that trying to learn said dialect is very difficult without a good bit of audio recording. Reading about its phonology is helpful but it would be impossible to properly know what a language or dialect sounded like without audio to supplement.

The clip itself lasts 29 seconds with only 14 of those being uninterrupted since a man and woman begin to talk about something to do with transcribing. I got the clip from this 'Gaeilg Chill Chainnigh' blog post. The link was dead but thankfully somebody in the comments posted a new link to it, so thank you Cathal!

Unfortunately, I don't know what radio programme this is from or if more of the complete audio is available online apart from this Twitter post from Seaghan Sionnach. I'd guess from the dated and almost British sounding prestige register of Hiberno English that the gentleman uses that this is from 20+ years ago but I wouldn't be sure. I have contacted the UCD archive and received news that I may receive all of the audio at some point in May. Unfortunately though, I may not be able to share it due to it not being my audio.

Here I'll be giving my attempt at a phonemic transcription along with the original Irish and English translation. I have used /r´/ instead of /ʒ/. My skill in this is not great so I would appreciate feedback.

I don't know how to insert audio into blogger so here's a video.


"Bhí feirm- ... bhí feirmeoir ann fadó, agus chua' fear bocht ag gearradh brosna aitinn ar chnoc. Tháinig an feirmeoir air agus d'fhiafraigh sé ..."

"There was a farmer long ago, and a poor man went cutting gorse wood on a hill. The farmer came and he asked ... "

/v´i: f´er´əm´ ... v´i: f´er´ə'm´u:r´ u:n [?] fado agəs xuə f´ar boxt ə g´arə brosnə at´əŋ´ er xruk .../

Here we can see a number of the features I described in my previous post. An interesting point is how vowels are generally raised in the presence of a nasal consonant such as in feirmeoir and ann. The fortis/lenis distinction was not made at this time in Kilkenny and no such phonemic distinction had been present for easily 100 years. As like the rest of Munster, the vowels here shifted or broke in order to stop the merging of words. This happened differently in different areas.

Some 'seanfhocail agus seanchainteanna' from Iarsmaí de Ghaeilig Chontae Chill Choinnigh

Risteard Breatnach (R. A. Breatnach), born in 1911 in County Kilkenny, was a linguist who wrote prolifically on Gaelic dialects and is said to have had a mastery unrivalled by few other scholars in the history of the language's scientific study. He contributed significantly to Éigse covering many topics and published several books. Another scholar also confusingly called Risteard Breatnach (R. B. Breatnach) wrote a study on Ring Irish which is a useful resource for anybody interested in that dialect and likely a useful resource for people interested in East Munster Irish generally. 

Being a Kilkenny man, Breatnach was drawn to conduct a study of the dialect from his own home county which he did as his master's dissertation. As useful as this would be for my own effort, I'm not able to get it as it remains unpublished in UCD's library. Part of this study, however, was published in Éigse Volume XXVI in 1992 under the title 'Iarsmaí de Ghaeilig Chontae Chill Choinnigh' meaning 'Relics/Remnants of Irish from County Kilkenny'. The article contains some useful information about the dialect as well as the main informant and likely last speaker, Pádraig 'an Paorach' de Paor. De Paor, who has gone by many names, was said to have been a very fluent speaker and provided Breatnach with a number of seanfhocail (proverbs, 'old-words'), seanchainteanna ('old-sayings') as well as several verses and prayers.

In this quick post, I will be showing some of the seanfhocail and seanchainteanna that I like and I will go through the meaning, pronunciation as well as cool dialectal features. Breatnach gives all of these using spelling that is a fair mix of dialectal Munster and standard but I will put into brackets any spelling that could also work. Most of these are fine and no majorly disastrous standardisation has occurred but spelling is a complicated issue in Irish dialects that I won't get into.

The transcription here is phonemic and in Celticist. Heinrich Wagner also listened to tapes of de Paor and gave much narrower phonetic transcriptions in his Linguistic Atlas and Survey of Irish Dialects (LASID). I am currently trying to get a hold of the audio but have only got a tiny bit as writing this.


Mara bhfuil (buil) ba ar na cnoic agam, thá suaineas (súineas) insa sop agam.

/mɑrə bil´ bɑ eʒ nə knik´ əgom, hɑ: su:n´əs ənsə sop əgom/

This one means 'If I've cattle in the hills, I can rest (lit. have safety in the sop)'. This is almost identical to a phrase my father once used, 'If you've cattle in the hills, you've money in the bank'. Cattle are valuable things and keeping them in the commonage in the hills will keep them safe because as my dad put it, they're untouchable and they can only multiply.

Straight away, we have some identifiably East Munster features. thá and buil as variants of  and bhfuil are known to be features of Déise Irish currently spoken in Ring Co. Waterford and the two were used in the South Tipperary variety of this dialect too. However, thá is occasionally used in most dialects varying by the speaker and is the most common form in Scotland as tha. In Kilkenny texts that do use both  and tháthá is overwhelmingly dominant. It seems that buil was always spelt bhfuil, but I have never encountered the latter in phonetically transcribed material.

A distribution of phonetic [a] and [ɑ] for phonemic /a/ is common to Irish dialects, however, Breatnach lists these as two separate short vowels /a/ and /ɑ/ without outlining what exactly the difference is. These were not distinguished in any historic Gaelic variety that I know of, so it's not possible that this is an older distinction which just happened to be lost in every other modern recorded Gaelic dialect. The other Breatnach (R. B. Breatnach) also does not give these as separate in his study of Ring Irish. I cannot find a single minimal pair and there are multiple places where Breatnach notes that de Paor had /a/ where he would have expected /ɑ/ like in /sagərt/. I suspect the difference might be that /ɑ/ is used in the presence of dorsal consonants (/k/, /g/, /x/ etc.) but that would be a complementary distribution so that doesn't make sense either. This all leads me to believe that the distinction was in fact not phonemic, but who am I to question Breatnach? He was a scholar and knew the dialect better than I do so I'll have to trust his judgement for now.

For the preposition ar which corresponds roughly with on, we have another strikingly Kilkenny/Ossory feature. This word is pronounced with a final slender r /r´/ and formerly spelt 'air' as in Scotland today*. The pronunciation of slender r /r´/ as something like [ʒ] is very peculiar. It looks like the consonant became so palatalised that became a full fricative, similar to what occurs in parts of Erris Co. Mayo, however, the pronunciation is slightly different and I will cover this in a future post with audio. This is particularly strong in speakers from the Dún Chaocháin peninsula and can be heard used by Pádraic Ó Gionnáin from Cill a' Ghallagáin here

Breatnach makes a habit of using /ʒ/ rather than /r´/ in his transcriptions. This is an interesting choice as it allows him to note places where this sound is devoiced and so is pronounced the same as /s´/. This has certainly lead to a number of words being reanalysed so I suspect that this is how Pádraig de Paor and other speakers would have phonemically analysed it but I can't say for certain.

For cnoic, we have another peculiar one. This word as well as many others beginning with an initial consonant cluster composed of any plosive + /n/ (also all lenited and eclipsed forms) is often pronounced with the /n/ as /r/ outside of Munster.  T. F. O'Rahilly wrote in Irish Dialects Past and Present (1932) that this was the case in North Kilkenny whereas it was not in the South of the county. This is also repeated by Wagner in LASID but in a small clip of de Paor (who is from the South), it does sound like he is using /xr/ for chnoc. The audio, however, is very poor and it's entirely possible that I am mishearing or that both pronunciations were used as is the case sometimes in Galway or the Aran Islands. 

As for /su:n´əs/, I think it's simply the case that the vowel has been smoothed. This happens with virtually every /uə/ in the article and Wagner notes several words having both smoothed /i:/ and broken /iə/ (or /i:ə/) such as scian.

insa is a common dialectal spelling for the standard sa. This older and dialectal form is written in Scotland as anns a' but as far as I can tell, is also found in every Irish dialect.


Beas an oíche ag dul i ngiorracht (i giorracht)  
agus an lá ag dul abhfad
agus na héiníní tuirseach
ó bheith ag déanamh nead.

/b´es ən´ i:hə dol ə g´irəxd
ɑgəs ə lɑ: dol ə vɑd
ɑgəs nə he:ˈn´i:n´i tirs´əx
o: v´e d´ianə n´ad/ 

This is a short rhyme relating to St. Brigid's day and Spring. In idiomatic English (prose?), it would be as follows.

The night will be getting shorter
and the day getting longer
and the birds tired
from making nests.

Sounds better in Irish in my opinion. The first thing that stands out to me is the rhyme scheme. Rhyming in Irish poetry is more like assonance in that only the vowel has to match up for a rhyme rather than the full 'rime' of the syllable. The same as in English, two words I saw in a rhyme from Erris such as /g´l´aN/ and /t´aN/ rhyme because the 'nucleus' (the middle part/vowel) and the 'coda' (the optional final part/consonant) are the same. Collectively, these two are called the rime. Different from English, however, is that only the nucleus (the vowel) needs to be the same for a rhyme to work here. In Classical Gaelic, this sytem worked a little bit differently with more rules about what could and could not rhyme in the coda but I believe that is no longer the case in modern verse. 

We have some interesting features here again. People familiar with Connacht or Ulster varieties of Irish will recognise beas, the future relative of . This form is usually used as a bheas in relative sentences meaning 'that will be' as in an uair a bheas tú which means 'the time/when you will be'. Interestingly, relative forms are often used in non-relative sentences in Kilkenny but only in the future tense. This is rather odd given that relative forms don't exist generally in other Munster dialects and the classical language from Ossory verse often agrees with this practice and does not use the relative.

For i giorracht, we have an interesting example of the loss of eclipsis (urú). What we would normally expect here is i ngiorracht /ə ŋ´irəxd/ with eclipsis on the g because of the preposition i meaning in. Whilst I wouldn't say that eclipsis was entirely lost in Kilkenny, it's clear that it wasn't as productive as it was/is in other parts of the country. Seosamh Laoide, writing in An Claidheamh Soluis in 1910, noted that eclipsis was lost as in Rathlin and Scotland such as in the phrase go bí tú and there are a number of phrases where the mutations are mixed up or eclipsis is not present. Breatnach has seach ndiabhal and go mbeirimid Wagner has /gə v´anə/ for go mbeannuighidh. Eclipsis is always included in the verses in Dáithí Ó hÓgáin's Duanaire Osraíoch in what was then the older literary register. This literary style differed in vocabulary and grammar and the pronunciation was likely older too as in other Gaelic speaking regions, so not exactly representative of the spoken dialect. 

There's not much particularly interesting in the next two lines apart from the stress on héiníní. Taking again from Wagner and Laoide, stress in Kilkenny generally agreed with the rest of Munster without major exception. However, Matthew Byrne from Tullowbran Co. Kilkenny, another informant to Breatnach's thesis, had slightly more unusual stress patterns that differed from his contemporaries in Glenmore.

In the final line, we have two interesting points. déanamh here is the verbal noun of the Connacht and Ulster verb déan rather than the usual Munster dein. In this respect, Kilkenny differs from Ring Co. Waterford, however, déan and rinne were both found in the Déise dialect of South Tipperary. Although I think déan was more common, dheineadar is attested by Wagner in LASID.

We also see in this word, the breaking of the original /e:/ vowel into the diphthong /ia/. Before a slender consonant, this vowel broke in Munster and parts of Scotland as well as being present in some words in Ulster and Mann. It'd probably be more accurate to just say that this unbroken vowel was preserved in Connacht and Donegal. Originally, the diphthong was separate from /iə/ but over the last century, they have essentially merged with certain older speakers sometimes variably having them separate. This distinction is sometimes made by de Paor with him leaning towards having them unmerged. AnLoingseach explains this better in this video of his.

You can see in this image from Wagner's LASID
that the vowel is broken in all of Munster.
It also being sometimes broken outside of the
province but this is not phonemic. If the image
quality it good enough, you can also see that
in the data gathered between 1950 and 1954,
many of the speakers had /iə/.

Pádraig de Paor also has the final -amh as /ə/ but he does not always do this. He and the two other informants in LASID use both /ə/ and /əv/ for words such as leanbh, talamh, annamh, etc. In Wagner's entry for this verb, none of the informants used /əv/.

nead is /n´ad/ but sometimes the /a/ vowel is /e/. This is an interesting preservation that AnLoingseach also talks about in the video I linked above. It is preserved in North Donegal to some extent.


Níl (nín) lao ná leanbh ná leanfadh a mháthair.

n´i:n´ le: nɑ: l´anə(v) nɑ: l´anhəx ə vɑ:həʒ

This would mean 'There's no calf nor baby that won't follow its mother.'

Here with nín we can see another unique East Munster word. This is a variant of níl and not . I'm not knowledgeable enough yet to talk about how this came about but one theory is that it was nasalisation. Another states that ní fhuileann being shortened.

The /e:/ vowel in lao is typical of Munster. As is the final /-əx/ in leanfadh.


In conclusion, we can see that the Irish spoken in County Kilkenny stands apart in its dialect being a unique variety closely resembling but definitely separate from the Irish of Déise. Although these phrases aren't necessarily perfect examples of the dialect as it was naturally spoken, they are insightful into the features of the area.

This ended up a bit longer than I anticipated but I hope you've enjoyed it. I'll be continuing this post in multiple parts.

Sorry for having not posted in a while, I will put up a few things soon.

*For some reason, this was changed in an Caighdeán Oifigiúil, the official standard language and spelling. Although you could say that it stops confusion with the prepositional pronoun air, any difficulty in telling them apart is obliterated by using context.


Friday 9 April 2021

what's in a name

Update: Since writing this, I have discovered that a woman by the name Margaret Cody from Coolnahau, Mullinakill was recorded by Canon William Carrigan in 1933. I do not have this recording.

From now on I will be calling Pádraig de Paor exclusively by 'Pádraig de Paor' or 'de Paor'. Pádraig de Paor was one of the main informants for Kilkenny/Ossory Irish material, the only native speaker I know to have been recorded, the last fluent native speaker and quite possibly the last speaker altogether. Annoyingly, he has gone by an array of names which I have listed below.

  • Pádraig de Paor
  • an Paorach
  • Pádraig Paor
  • Patrick Paor
  • Paddy de Paor
  • Paddy Power
  • Patrick Power