Talk:Irish phonology

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
Featured articleIrish phonology is a featured article; it (or a previous version of it) has been identified as one of the best articles produced by the Wikipedia community. Even so, if you can update or improve it, please do so.
Main Page trophyThis article appeared on Wikipedia's Main Page as Today's featured article on February 13, 2008.
Article milestones
November 29, 2006Peer reviewReviewed
November 30, 2006Good article nomineeListed
March 17, 2007Featured article candidatePromoted
Current status: Featured article


The chart has /sˠ/ and /ʃ/. One of the diagrams has /sˠ/ and /ɕ/. Since /ɕ/ = /ʃʲ/ (the IPA symbol is palatalized postalveolar), should the chart read /sˠ/ vs. /ʃʲ/ instead? kwami (talk) 08:48, 5 February 2008 (UTC)

The sound phonetically is apparently usually [ɕ] = [ʃʲ], but it's a little hard to pin down precisely. Some sources describe it as [ʃ], and since effectively all Irish speakers are bilingual with English, it's pretty likely they use exactly the same sound when speaking Irish as when speaking English. This article originally used /ɕ/, but I later changed it to /ʃ/ due to complaints (probably still on this talk page) that /ɕ/ was inaccurate and/or misleading. I figure for a broad transcription, /ʃ/ is good enough and less likely to be confused with /ç/, which is a completely different phoneme. /ʃʲ/ would work too, but it has a touch of OR since while some published works use /ʃ/ and others use /ɕ/, none uses /ʃʲ/. —Angr If you've written a quality article... 16:32, 5 February 2008 (UTC)
Per English phonology, /ʃ/ is often slightly labialised. This lip-rounding is even stronger in German. I'd expect the Irish phoneme to never exhibit labialisation, unless perhaps under the influence by English. (Does Hiberno-English show this lip-rounding?) As I am so used to [ʃ] showing labialisation (in Polish, which features a contrastive phoneme /ɕ/, the postalveolar fricative it contrasts with is actually retroflex, so it doesn't provide a true /ʃ//ɕ/ contrast, either), distinguishing /ɕ/ from an unlabialised /ʃ/ can be very difficult for me. So the Irish slender sibilant is indeed actually more palatalised that plain [ʃ], at least originally? Then the article should note this for the user not to be confused by the choice of symbol (as in be led to conclude it is completely equivalent with English /ʃ/, including potential labialisation). --Florian Blaschke (talk) 05:27, 25 July 2015 (UTC)


The article should make clear in the 1st sentence that it's talking about Irish Gaelic, not Irish English, rather than leaving it to a picture & the 4th paragraph. —Preceding unsigned comment added by (talk) 10:48, 13 February 2008 (UTC)

Something like "The phonology of the Irish language varies..." would be a quick fix to that. — Ƶ§œš¹ [aɪm ˈfɻɛ̃ⁿdˡi] 14:02, 13 February 2008 (UTC)

ok wtf these guyz are wierd seriously who has paragraph long discussions in articles like wtf —Preceding unsigned comment added by (talk) 23:47, 13 February 2008 (UTC)

The opening sentences are rather jarring. Variation of pronunciation according to recognized dialects is a given for all the languages I know. But saying there is no standard needs clarified because the article seems to be talking mostly about three standards or at least 'prestige' dialects of Irish Gaelic. —Preceding unsigned comment added by (talk) 14:21, 10 May 2008 (UTC)

Slender ng[edit]

Why [ɲ] instead of [ŋʲ]? I'd say that [ɲ] is the symbol for Spanish "ñ" or Italian "gn", a palatal nasal. But slender Irish "ng" is a different sound, a palatalized velar nasal. -- (talk) 23:05, 14 February 2008 (UTC)

Like all IPA characters, [ɲ] has some wiggle room in its reference. The Spanish and Italian sounds are slightly prepalatal, while the Irish sound is slightly postpalatal, but they're close enough that the same symbol can be used. Irish [ɲ] definitely has the same place of articulation as [c] and [ɟ], so it makes sense to use a symbol from the same series. Icelandic phonology also uses [ɲ] to indicate the same sound as in Irish. —Angr If you've written a quality article... 04:59, 15 February 2008 (UTC)
That's what I mean (…I think the anonymous commenter above was me…): I bet the whole "slender velar" series consists of palatalized (or otherwise fronted) velars, [ŋʲ gʲ kʲ], not of palatals, [ɲ ɟ c]. With the exception of [ç ʝ], which are indeed very similar to [xʲ ɣʲ], palatals sound very similar to palatalized alveolars, not velars; soundfiles of palatal stops (Ladefoged recordings from Hungarian) are here.
I have never heard Irish spoken, though, so I could obviously be in for a surprise… David Marjanović (talk) 22:35, 18 May 2008 (UTC)
Well, let's see what some of the experts say:
  • Describing the dialect of Tourmakeady, de Búrca says of the sound he writes ŋ′: "This symbol denoted a voiced palatal nasal, made by bringing the centre of the tongue into contact with the back part of the hard palate..."
  • Ó Cuív says of West Muskerry ŋ′: "This is a voiced palatal or palato-velar nasal consonant, and corresponds to ɡ′ with regard to tongue-position, lip-position and glides." Of ɡ′ he writes, "This is a voiced palatal or palato-velar plosive, formed by pressing the front [sic!] of the tongue against the roof of the mouth at the junction of the hard and soft palates, keeping the tip down behind the lower teeth."
  • Sjoestedt says of Kerry (Dingle Peninsula) ŋ′ that it has the same place of articulation as g′, and of g′ that it has the same place of articulation as k′. She describes k′ as "une occlusive palatale, sourde, forte, aspirée".
  • Finck says for the Aran dialect: "Der buchstabe ŋ́ bezeichnet einen stimmhaften nasalen verschlusslaut, dessen verschluss zwischen dem hinteren zungenrücken und der mitte des harten gaumens gebildet wird."
  • Quiggin says of ɲ (using that symbol): "This symbol denotes a palatal ng formed with the back of the tongue against the place where the hard and soft palates meet and is therefore similar to the French gn in 'signe'."
  • Ní Chasaide in the Handbook of the IPA, describing the dialect of Gweedore, doesn't go into detail about the precise place of articulation, but the chart has only a column for "Palatal" and the sound is written ɲ.
So all these authors describe it either as "palatal" or "palatal or palato-velar". And other languages also conventionally use the symbols [c ɟ ɲ] for sounds just as far back as the Irish sounds, see Icelandic phonology and Turkish phonology for some examples. There's really no reason not to use /ɲ/ for the Irish sound, except that it may not correspond exactly to a sound in other languages that uses the same symbol. But that's no reason, since IPA symbols never mean precisely the same thing from one language to the next.
The Hungarian sound files you linked to are interesting. The word-initial examples sound just like the Irish palatals to me, while the word-medial ones sound just a tiny bit further forward, like [c̟ ɟ˖ ɲ˖] or [t̠ʲ d̠ʲ n̠ʲ]. —Angr 18:13, 22 May 2008 (UTC)

Congratulations on FA[edit]

Impressive work! Krym66 (talk) 23:41, 16 February 2008 (UTC)

Thank you! —Angr If you've written a quality article... 06:34, 17 February 2008 (UTC)

I'd say the article needs considerable improvement. For example, looking at the chart of consonants, it's obvious that many come in pairs based on some sort of hard palate vs. soft palate/velum distinction, but it could benefit from an explanation about secondary articulation vs. primary (since quite a few consonant sounds are primarily in these parts of the mouth). Also, how about an explanation of this distinction as it is reflected in the grammar?

It also seems like a language that could benefit from a syllable-based account, since the vowels are affected too. —Preceding unsigned comment added by (talk) 14:27, 10 May 2008 (UTC)

The broad/slender distinction in consonants is discussed throughout the article. The reflection of the distinction in the grammar is outside the scope of this page, which is just about the phonology, but could be brought up in the various pages relating to Irish morphology. A syllable-based account would be appropriate for a linguistics paper, but not for Wikipedia, which rejects original research. —Angr 15:31, 10 May 2008 (UTC)

Phonology is part of a language's grammar--see Chomsky. The relationship of phonology with language structure at other levels is IRRELEVANT or out of the scope of the article? I don't think so. I don't think mentioning how it affects the pronunciation beyond a single segment requires original research. —Preceding unsigned comment added by (talk) 06:05, 15 May 2008 (UTC)

This article is a good case in point about what is wrong with Wikipedia. It has far too much information--much of it debatable (for example, why structarlist phonemics is such an important part of the phonological analysis -- to be understood by the non-specialist). And yet it has jarring idiosyncracies to it, showing a lack of actual editorial unity across related topics work (inconsistencies or ommissions or lack of links to related treatment in topics like Irish language, Gaelic, Celtic, Goidelic branch of Celtic languages, Irish as a Second language, etc) to work. I really doubt that in areas like linguistics if Wikipedia will ever get its act together.

Rather than gloating on what a good article this is (in terms of linguistics, it is so-so, in terms of knowledge of Irish/Irish Gaelic it is good because there are very few phonologists with a speaker's knowledge of the language), it ought to be cut down and incorporated as a section into Irish language and could also be used to enlighten entries on such topics as Celtic languages, Goidelic branch, etc.).

Wiki project has lost its way. This article is a good example. If Wikipedia would put together a more consistently edited and limited encyclopedia, then articles like this might have a place in a more comprehensive data base of human knowledge. —Preceding unsigned comment added by (talk) 04:14, 17 May 2008 (UTC)

Cutting down the article and incorporating it back into the main Irish language article would obviously be a change for the worse that would serve no purpose at all. If you feel you can improve or expand this article, go ahead, but remember to avoid original research (i.e. you can only discuss issues that have already been discussed in other people's published work) and to cite your sources. —Angr 06:12, 17 May 2008 (UTC)


Are the diphthongs rising or falling? That is, should the <  ̯> be placed under the first or second component? kwami (talk) 20:13, 3 June 2008 (UTC)

They're all falling. Both /iə/ and /əi/ are [iə̯] and [əi̯]. Scottish Gaelic has some rising diphthongs, I think, but Irish doesn't. —Angr 20:43, 3 June 2008 (UTC)

Slender labials in Scottish Gaelic?[edit]

"Many of the phonological processes found in Irish are found also in its nearest relatives, Scottish Gaelic and Manx. For example, both languages contrast "broad" and "slender" consonants, but only at the coronal and dorsal places of articulation; both Scottish Gaelic and Manx have lost the distinction in labial consonants." This is contradicted in the Scottish Gaelic article, which lists slender versions of b, bh, p, and ph. I assume one of the articles is wrong, though I am not knowledgeable enough to tell which one. Lesgles (talk) 01:30, 16 December 2008 (UTC)

As that article shows, the "slender labials" of Gaelic are actually clusters of labial consonant + /j/ in initial position. It claims there are also clusters of /j/ + labial consonant in final position, but I'm not sure about that. But a cluster isn't the same thing as a single palatalized consonant, which is what Irish has. —Angr 08:25, 16 December 2008 (UTC)
I see. Thanks for clearing that up! Lesgles (talk) 00:49, 17 December 2008 (UTC)

broad/slender vs. Slavic hard/soft consonants[edit]

Palatalized consonants normally are not considered part of the phonetic inventory of the Bulgarian language. — Klimenok (talk) 18:55, 19 November 2010 (UTC)

IPA palatal notation[edit]

I had changed all references of [i̯] to [j] because the International Phonetic Alphabet defines them as absolutely canonical. But they were reverted because this article uses [j] for slender dh/gh, which has more friction. That's what we call a fricative consonant, and I recommend the latter be changed to [ʝ] which is the dedicated canonical symbol for the voiced palatal fricative. [i̯] should be changed to [j], as it actually is. - Gilgamesh (talk) 18:38, 21 June 2012 (UTC)

No, it shouldn't. There are degrees of friction. The fricative [ʝ] is an allophone of /j/, which has an approximant allophone [j] in other contexts, as explained in the article. [i̯] has even less friction than that, being merely a nonsyllabic vowel. But quite apart from the phonetics, using [j] for the onglide is misleading because the onglide isn't an allophone of /j/, it's just a very brief nonsyllabic vowel, which is much more intuitively written [i̯]. For the sake of comprehensibility, the symbols should be kept separate. There is no reason whatsoever not to use [i̯] for the onglide, but there are good reasons not to use [j] for it. Angr (talk) 23:01, 21 June 2012 (UTC)
I take it that Irish speakers can distinguish between [i̯] and [j], right? It might be relevant to mention that at semivowel or approximant. — Ƶ§œš¹ [ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ] 00:13, 22 June 2012 (UTC)
This...sort of flies in the face of IPA orthodoxy. An asyllabic [i] is always strictly equivalent to [j]. It's nonstandard IPA to say otherwise. If there are variable levels of frication, why not use [j˔] or [ʝ˕] where this has more frication than [i̯/j]? - Gilgamesh (talk) 12:44, 22 June 2012 (UTC)
There's a similar contrast between the two in Spanish, so it's not unheard of. — Ƶ§œš¹ [ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ] 14:08, 22 June 2012 (UTC)
The very fact that [i̯] is equivalent to [j] means that we can use either one, and indeed we can use both. No law says we're required to use only one to the exclusion of the other. In this case, the function of the two sounds is so different (one is an onglide between a back vowel and a palatalized consonant, the other is an allophone of a consonant phoneme) that it is beneficial for presentation purposes to distinguish them visually. The phoneme /j/ would never occur in the positions where the onglide occurs, so it's not really possible to say whether Irish speakers can distinguish them by their acoustics alone: their position in the word would always be sufficient. Angr (talk) 17:55, 22 June 2012 (UTC)
[j] is generally understood to be a consonant and [i̯] a vowel, which fits the existing convention of this article. — kwami (talk) 18:47, 22 June 2012 (UTC)

I see... - Gilgamesh (talk) 04:35, 23 June 2012 (UTC)

Audio samples[edit]

I do not speak Irish and the distinction broad-slender is not very clear for me. What do you think about adding some audio samples? Xpicto (talk) 07:03, 24 June 2013 (UTC)

I agree that audio examples would be helpful, here, as well as in the article velarization. --Lundgren8 (t · c) 12:08, 24 June 2013 (UTC)
Maybe one of our native speakers can be prevailed upon to make some audio samples. Angr (talk) 12:37, 24 June 2013 (UTC)

Two-constant clusters?[edit]


Hello! Can someone explain what is meant by ”constant” in this picture to the right (and its brother)? What’s the difference here between ”constant” and ”consonant”? To me ”two-consonant cluster” would make as much sense, but I haven’t encountered ”constant” in this context. Thanks for explaining! --Lundgren8 (t · c) 12:06, 24 June 2013 (UTC)

It's a typo. — Lfdder (talk) 12:10, 24 June 2013 (UTC)
Rather embarrassing no one caught that before. Fortunately, as an SVG it's easy enough to fix. Unfortunately I don't have an SVG editor so I can't fix it. Maybe one of you can? Angr (talk) 12:33, 24 June 2013 (UTC)
I was suspecting it was a typo, but since it occurred in both images, and since I’m not a native speaker of English, and since this is a featured article, I didn’t want to criticise. I can maybe fix it on Inkscape if I get it to work on my new computer; the online SVG translation editor is down at the moment. --Lundgren8 (t · c) 12:58, 24 June 2013 (UTC)
For the record, it was correct in the original PNGs that I made. I'm not responsible for these SVGs that someone thought would be a good idea. Angr (talk) 13:06, 24 June 2013 (UTC)
Whoever made these converted text to objects, so it was just like editing a raster. — Lfdder (talk) 13:19, 24 June 2013 (UTC)
Ah, you edited them? I was just about to upload them myself, good. But yes, a new text layer had to be made. --Lundgren8 (t · c) 13:24, 24 June 2013 (UTC)
Thanks for fixing them, Lfdder. Back when I used to make SVG files, I always had to convert text to objects, otherwise the text would just appear as black boxes. Angr (talk) 14:38, 24 June 2013 (UTC)


I suggest the word "normalized" be used in place of "standardized". Speling12345 (talk) 7:41, 13 December 2013 (UTC)


The article says they are realised universally as taps, but I’ve heard (I think predominantly) approximants. Siúnrá (talk) 13:45, 11 January 2015 (UTC)

The article is based on sources in which the speech of native speakers from Gaeltacht areas was analyzed, especially those who lived in the first half of the twentieth century. I wouldn't be at all surprised if Irish people nowadays (most of whom aren't native speakers to begin with) used the English /ɹ/ sound when speaking Irish, but the sources consulted for this article don't mention that if so. —Aɴɢʀ (talk) 16:51, 2 May 2015 (UTC)

slender c and g and IPA c and ɟ[edit]

For me, IPA [c] is rather a t-like sound, e.g. Hungarian orthography ty (the same for [ɟ] which is rather a d-like sound though written gy in Hungarian). There are only a few cases of mixing between slender /k´/ and /t´/ in Irish (e.g. / as a wh-word in Connacht). But probably never ceist and teist would be confused. For me, /k´/ is clearly recognizable as a k-sound and /g´/ is clearly a g-sound and both are clearly differentiable from broad /k/ and /g/. Perhaps they aren’t really palatal plosives but rather palatovelar ones. So, I’d describe them in IPA rather as [kʲ] and [gʲ]. -- (talk) 12:28, 13 February 2015 (UTC)

They vary between palatal and palatovelar; also, the symbols /c/ and /ɟ/ are flexible enough that they can legitimately be used for k-like palatals as well as for t-like ones. The /c/ and /ɟ/ of Icelandic and Greek, for example, are very much like the Irish ones. —Aɴɢʀ (talk) 16:49, 2 May 2015 (UTC)

External links modified[edit]

Hello fellow Wikipedians,

I have just modified one external link on Irish phonology. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

As of February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete the "External links modified" sections if they want, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{sourcecheck}} (last update: 15 July 2018).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 22:34, 23 September 2017 (UTC)


In many places in different wikipedia language versions, the Irish "broad" [s] is marked with the velarization diacritic: [sˠ]. But is it for sure that, phonetically, a voicless alveolar fricative *can* be velarized? I find this very difficult if I try to articulate it. Is there really anything that goes beyond simple coarticulation in [sa] combinations? Given that the "slender" s is [ʃ] anyway, the velarization mark strikes me as unnecessary, even if it were phonetically possible. What do the sources say? --Alazon (talk) 13:19, 1 November 2017 (UTC)