As AnLoingseach, who notoriously is not quick at uploading videos, has uploaded not one, but two  great videos, I assume it’s time for me to update something on this blog. The first of these is on the pronunciation and history of the definite article while the second, which was written and filmed as a follow-up to a comment in the first, is on the etymology of the word “sionnach”, based on a tweet by Dr David Stifter. They are both great videos, in AnLoingseach’s usual style of super entertaining and informative rambling, and I recommend everyone watch them. Because of that, I was inspired to update this blog. We’ll be discussing the differences between subdialects of Cois Fharraige Irish.

Though Cois Fharraige is classified as one dialectal region according to Ó Máille (see Irish Dialects of Connacht), we can find splits and different features in these dialects.  Ó hUiginn splits it into Cois Fharraige and Iarthar Chonamara, whereas de Bhaldraithe adds a third dialect in the eastern part of Cois Fharraige, near Galway City. He mentions this one has several similarities with East Galway Irish, and includes it as a separate appendix in The Irish of Cois Fharraige. We will look at the main differences between the three, relying heavily on de Bhaldraithe’s book. The ‘standard’ form will be the central area, where de Bhaldraithe’s work was all done. All phonemes are transcribed as de Bhaldraithe uses them. See his book for more info.

Subdialectal Differences in the Western  Area of Cois Fharraige

We’ll start by looking at the western end. This is basically the dialect Ceantar na nOileán (information about their language plan). Or, to be more precise, it’s the dialect of Leitir Móir and Garmna, where the speakers were from. Here is what de Bhaldraithe (1975, 117) has to say as a preliminary remark.

The dialect of Leitir Móir and Graumna Islands, which are on the western fringe of Cois Fhairrge, appears to be substantially the same as that of Teach Mór area, except in a few points, the most important of which are given below.

The first difference he mentions is devoicing of nasals, laterals and rhotics (‘r-sounds’). Essentially, medial l,m,n,r are all partially devoiced if they occur before th or f of the future/conditional suffixes. He also mentions certain speakers have this devoicing in other situations, but its most consistent in the future/conditional stems.

He next mentions that there is a distinction between ‘unlenited, non-palatal n’, as opposed to ‘lenited, palatal n;. The lenited version is realized as a ‘voiced velarised interdental nasal’ (/ɴ/ in the Celticist transcription; link to Fuaimeanna for pronunciation). n is unlenited in three main cases: (a) when written nn, (b) when preceded by s or r, (c) when followed by t,d,l. Lenited non-palatal n is a ‘voiced velarised post-alveolar nasal (/n/ is used by de Bhaldraithe). As he explains it: ‘In making it the air-passage in the mouth is blocked by the tongue being raised to make contact with the back part of the teeth-ride.’ A key thing to note is that ‘initial n is lenited under the same circumstances as other consonants’. Thus lenited n does exist in Connemara (as does l), despite what teachers say. He does mention that this distinction is lost among the younger generation (and given that the first edition was in 1945, they’re likely all that is left. He mentions the same is true for the lateral — that the difference between the lenited and unlenited non-palatal l are more often maintained than in central Cois Fharraige, but they’re gone among the younger speakers.

Several changes have happened to vowels in this dialect. before s,t or when preceded by t,d,h and followed by d,t, the vowel a has changed to /æ/, which normally occurs half-long. Likewise u > /i/ when preceded by t,d and followed by s. In the future autonomous, the vowel of the ending is a schwa, something not found in the Teach Mór area. Medial non-palatal mh is a nasalized /w/, nasalizing the vowel in front of it. Finally, before /h/ long vowels tend to become half-long or short, while the half-long vowels tend to become short. Short vowels are sometimes lengthened to half-long based on analogy.


Finally, there are three miscellaneous differences. In the western part of Connemara, intervocalic th is retained as /h/ where it’s been lost in Cois Fharraige. When preceded by /i/ or /i:/ it can become /x/ as seen in ithe pronounced as if it was iche. However, where the old spelling would have had ghth, if it is preceded by an unstressed vowel, it is silent. Palatal intervocalic ch has become /h/. And f in the future and conditional has become /h/ when preceded by a lose fricative (tráighfidh, for instance)

Subdialectal Differences in the Eastern Area of Cois Fharraige

In the eastern area, the main difference is in the development and realization of the vowel phonemes. Where several vowels have been lengthened or diphthongized in the Teach Mór area, in the eastern part several short vowels are retained before ll,m,nn,ng,nr, unless a vowel immediately follows. Some examples of these include greim being pronounced with a short /i/ as opposed to a long one further west. However, some lengthening does take place. Liekwise, some speakers retain short vowels before rr,rn,rd,dr,gr,mhr, where they’d be lengthened or diphthongized further west.


Two vowels,  /æ/ (ea, ai) and /a/ (a, ai) do not occur in long forms with the one exception of /æ/when it appears before r, as in fearr. /o:/ is also retained before a nasal, though it has shifted to /u:/ further west; evidence points to this change taking place in the eastern part of the area however.


Finally, final unstressed schwa is something lengthened to /i:/, though it is still often found as a schwa. Likewise, unstressed ai > /i:/, Stress is found on the second or first syllable of the prepositional pronoun forms based on ag, and stress is often on the second syllable of words like coláiste, in which the first vowel is elided; stress can be found on the first syllable however.


Leave a Reply

Avatar placeholder

Your email address will not be published. Required fields are marked *