Cwyc Cwestsiyn about Welsh pronounciation

[Zumir]

Is there a way to tell if a "wy" is pronounced [uj] or [wɪ]?

[linguoboy]

Is there a way to tell if a "wy" is pronounced [uj] or [wɪ]?

If it's /uj/, there should be a circumflex on the w, e.g. ŵy gŵydd /'uj'gujð/ "goose egg" . But this isn't foolproof because the diacritic is often left off. Best just to learn to the PIE etymon of every individual vocabulary item.

[Bristel]

Is there a way to tell if a "wy" is pronounced [uj] or [wɪ]?

If it's /uj/, there should be a circumflex on the w, e.g. ŵy gŵydd /'uj'gujð/ "goose egg" . But this isn't foolproof because the diacritic is often left off. Best just to learn to the PIE etymon of every individual vocabulary item.

Isn't it just easier to learn the instances of this by context instead of looking at the PIE etymon?

That would be like learning the PIE etymon of Irish just because you can't figure out which pronunciation is used... (which is different due to local dialects)

Of course, you could be making a joke, of which I approve...

[Yng]

What linguoboy said.

And it's 'cwic cwestiwn'

[marconatrix]

This is one of the very few weaknesses in Welsh spelling, but a good dictionary ought to distinguish the two pronunciations, and there are a few rules of thumb but they're not 100%. Added to which some dialects tend to flip /uI/ over into /wI/ anyway. Btw. the Welsh use of 'w' for /u/ works elsewhere because there are no cases of */wu/ (indeed /w/ hardly ever occurs before any rounded vowel).

[Drydic]

Is there a way to tell if a "wy" is pronounced [uj] or [wɪ]?

If it's /uj/, there should be a circumflex on the w, e.g. ŵy gŵydd /'uj'gujð/ "goose egg" . But this isn't foolproof because the diacritic is often left off. Best just to learn to the PIE etymon of every individual vocabulary item.

Isn't it just easier to learn the instances of this by context instead of looking at the PIE etymon?

Sure maybe it might be 'easier', but which is going to be of more use, memorizing some welsh pronunciations or the various PIE etyma?

[Yng]

This is one of the very few weaknesses in Welsh spelling

That and the exceptions and the failure to represent actual pronunciation, yes.

[marconatrix]

If you represented casual everyday speech you wouldn't have a Welsh language, you'd probably have half a dozen at least The only 'deliberate errors' I'm aware of are ei, ein, eich and the artificial distinction between ac 'and' and ag 'with', but these do aid reading. Any practical orthography has to be some kind of compromise between the phonetic, phonemic and morphemic representation. These days there's a developing speciality and literature of 'orthography design' (google it), but IMO Orgraff yr Iaith Gymraeg was way ahead of its time in the way it dealt with the issues. 'Dych chi'n wedi ei ddarllen o?

[Yng]

If you represented casual everyday speech you wouldn't have a Welsh language, you'd probably have half a dozen at least The only 'deliberate errors' I'm aware of are ei, ein, eich and the artificial distinction between ac 'and' and ag 'with', but these do aid reading. Any practical orthography has to be some kind of compromise between the phonetic, phonemic and morphemic representation. These days there's a developing speciality and literature of 'orthography design' (google it), but IMO Orgraff yr Iaith Gymraeg was way ahead of its time in the way it dealt with the issues. 'Dych chi'n wedi ei ddarllen o?

Do. Wel, ie - ond pa dafodiaith ma'r orthograffi i fod i'w chynrychioli? Dim tafodiaith Caerdydd - dw i'm yn siwr oes tafodiaith yno bellach, beth bynnag, ond os oes, dim golwg na swn yr iaith wedi'i hysgrifennu sydd arni hi.

The problem is, it's an orthography of some northern dialect spoken c. 1500 which has been forced into service as the orthography of the standard language, which itself can't decide what on earth it wants to be based on. Obviously representing all dialects is impossible - but the current orthography doesn't even have decent conventions for DEALING with certain things that happen in all sorts of colloquial dialects, e.g. palatalised final /s/ (<si> just doesn't quite cut it there does it ), and although generally its phonemic representation is reasonably good - although since the various realisations of /ai/ differ so widely and are, thanks to hypercorrection etc, no longer strictly allophonic in many dialects, it's still not 100% - my bugbear is 'eisiau' (which no-one speaking a natural dialect pronounces how it's spelt). The other issue is borrowed phonemes - mainly /tʃ dʒ/ - which are usually spelt ts j respectively. This is an issue because in many dialects, one mutates into the other - tsips > dy jips - and then how do you represent the nasal version? And again, of course, the lack of a final /ʃ/ grapheme - thus <wats>, actually pronounced /watʃ/.

[marconatrix]

I think the problem behind representing the borrowed phonemes /tS, dZ/ and the extension of [S] so that it's now probably a phoneme in its own right, not just a palatal allophone of /s/, is that not everyone has these sounds, or at least didn't when the orthography was finally sorted c100 years ago. This may no longer be true now that everyone speaks English frequently. Cornish took these phonemes on board earlier and more thoroughly than Welsh, mostly in English loans, but some native developments sporadically produced these sounds too. And like spoken Welsh /tS > dZ/ was added to the mutation system, e.g. /tSi/ 'house' /a'dZi/ 'inside'. Spelling them 'ch, j, sh' is one of the few things that hasn't been an issue for us Welsh has long since adopted 'j', and surely also 'sh' finally in 'brwsh' etc. is hardly a problem. You'd have to write 'tsh' for /tS/ though since Welsh already uses 'ch' for /x/. But where is the problem in adding an 'h' on to 'wats(h)? How does the nasal mutation of /tS, dZ/ go, btw? I've not heard of that and it could be fun.

Every language I've come across seems to have one or two eisiau words. Does anyone know what happened in this case? Anyhow in the South they use moyn < ymofyn don't they?? But again if it's a problem then surely isio or whatever would become acceptable through use. Like nobody writes alarum in English any more, not unless they're trying to create an antique flavour.

As I understand it, the Welsh literary standard is simply the development of a long established oral standard going back at least to Early Modern times and particularly exemplified in the classical strict metre poetry. This seems to have evolved by consensus rather than favouring any particular dialect, but then being conservative it's bound to resemble the more conservative modern dialects. Middle Welsh prose is not too far from Lit. Welsh and yet some of the mss were apparently written in Gwent.

But at least Welsh has a long-established standard which has held the language, and to some extent at least the country together as a unit. You only have to look at Brittany to see what happens in the absence of such a standard. The main problem with Welsh now is that learners aren't tipped off about ein and eisiau and the few other traps and so may become fixated on spelling pronunciations.

[Yng]

Not just learners. Ein incidentally is not a mis-spelling - although in the south it is largely pronounced as /in/, in most of the north the most common pronounciation is pretty much how it's spelt.

/ʃ/ isn't just from borrowing English, though - it's found in many native contexts. It's allophonic natively in the South, probably, following /i/ - /miʃ/ etc - but in the north you have pairs like /neʃ/ 'gwneuthum' (well, to be fair, 'gwnes') and /nes/ 'nes' (the comparative of 'agos'), with the palatal s appearing in all 1sg preterites, with the regular ending -is /iʃ/. I have no idea where this is from - perhaps historic [ais] > [aiʃ] and then vowel screwery, but I dunno.

You're right to an extent - my main issue with representing the letters with borrowed consonants is that they don't show the same connection as the other mutated forms. When I'm writing, I tend to spell /tʃ/ in loanwords with <tsh>, etc - a convention which is reasonably well established - and <sh> for /ʃ/.

How does the nasal mutation of /tS, dZ/ go, btw? I've not heard of that and it could be fun.

I think it's /ŋʰ/ or somethinɡ like that. I've only ever heard it from a couple of people, but there are dialects where it's common (around here the nasal mutation is largely replaced with the soft mutation anyway, it seems).

[marconatrix]

My guess would be that nesh is from (gw)nais i which if treated as a unit would give the usual allophonic [S]. Then once /S/ became established on the basis of loans the form would be reinterpreted as nesh (i) with the sh now independent of the pronoun. Ei, ein, eich are written y, yn, ych in Middle Welsh /1, @n, @x/ which are apparently the genuine forms. The 'ei' is a hangover from Owen Pugh's crackpot ideas retained simply because they aid reading, there being other common words spelled y and yn. Part of the difficulty reading Middle Welsh is that y can be the verbal particle, the def. article, the prepostion 'to' or the possessives 'his, her', all very common words.

If [dZ] were interpreted as a realisation of /dj-/ then I suppose you might get /-n nj-/ ? And I suppose /tS ~ tj/ might give /-n nhj-/ or /-n njh-/, a voiceless/aspirated palatal nasal if that's possible.

[linguoboy]

The 'ei' is a hangover from Owen Pugh's crackpot ideas retained simply because they aid reading, there being other common words spelled y and yn.

I thought there was also an element of false etymologising on the basis of Latin eius.

[Yng]

My guess would be that nesh is from (gw)nais i which if treated as a unit would give the usual allophonic [S]. Then once /S/ became established on the basis of loans the form would be reinterpreted as nesh (i) with the sh now independent of the pronoun

As far as I know there was never a 'gwnais i' - it was 'gwneuthum', and then was restructured as 'gwnes', when the diphthongs had already long reduced (this is an issue with register, too, since writing 'gwnes' looks informal but 'gwneuthum' is a bit too literary for the same middle register as e.g. gwelais). Still, you might be right - the pronouns have been historically treated as a unit with verbs, in some cases - but in that case why -isht? Analogy, I suppose.

Ei, ein, eich are written y, yn, ych in Middle Welsh /1, @n, @x/ which are apparently the genuine forms. The 'ei' is a hangover from Owen Pugh's crackpot ideas retained simply because they aid reading, there being other common words spelled y and yn. Part of the difficulty reading Middle Welsh is that y can be the verbal particle, the def. article, the prepostion 'to' or the possessives 'his, her', all very common words.

Yes. But nevertheless, throughout the North the usual pronunciation is /e/, whether or not this was originally via spelling pronunciation.

I thought there was also an element of false etymologising on the basis of Latin eius.

[/quote]

Yes, that's the explanation I've always heard.

[marconatrix]

The 'ei' is a hangover from Owen Pugh's crackpot ideas retained simply because they aid reading, there being other common words spelled y and yn.

I thought there was also an element of false etymologising on the basis of Latin eius.

Correct, that was one of his odd ideas.