Questions not about German Thread

Discussion of natural languages, or language in general.
User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

OK, since I've progressed from staring longingly at Irish to "learning" it (so far only duolingo and memrise), and since I don't want to monopolise the linguistic struggles thread, I thought I might put some questions here.

Ubh - this seems to have a /v/ in it, but shouldn't it be /w/? Is this just an irregular spelling, or is there something i'm missing?

Broad vs slender is still killing me. But my most pressingly problematic phoneme is /r_j/. Soundfiles seem quite clearly to me to show this being a voiced fricative (that I interpret as /Z/, but presumably is more forward than my /Z/), or sometimes what sounds very much to me like /d/... but only sometimes. It seems the official (that is, dictionary, presumably gaelteacht) Munster dialect doesn't do this, but Ulster and Connact seem to do it more, and the forvo climbs from random actual irish people seem to do it even more - maybe this is an L2 thing as well as a dialect thing? I can't work out where the dictionary soundfiles do /Z/, when they do /d/, and when they do /r/, except that the stronger pronunciations are much more common at the ends of words.

I wouldn't be surprised if something was going on here, because how I'm meant to palatalise a tap, I don't know. So I'm to do one thing with the tip of my tongue, moving it quickly near the front of my mouth, while doing something totally different in a totally different part of the mouth with the bit just behind the tip? How is that possible!?

Come to think of it, does anybody know what the Irish flap is meant to be anyway? I've always mostly ignored the detail of flaps, but it seems there's quite a variety possible. I can tap against the flat bit behind the teeth, I can flap backward over the ridge, or I can flap forward over the ridge - the latter seems less natural, but apparently it's the only thing that's "officially" a flap. What does Irish use? My instinct has been to try to use the post-dental tap for broad and the ridge-flap for slender, but wikipedia suggests this is the only coronal pair NOT distinguished by POV...


Oh, and don't worry about the first question, i think I've worked it out now. It's because although everyone tells you that broad <bh> is /w/, /w/ is actually [v] (velarised) in word-final (or every) position, yes?



[On the positive side: the grammar seems pretty easy, so far. Not coming naturally due to everything being a different order to every other language I know, but makes sense, seems like it'll come naturally eventually. And I think I've worked out the theory of velarisation, although actually applying it in practice is another matter...]


EDIT: ok, saibhir - everyone seems to agree that the first vowel is something like [E]. But it seems like it 'ought' to be [{] at most, and probably a centralised [a]. Am I missing a rule, or is this an exception, do you know?
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Salmoneus wrote:Broad vs slender is still killing me. But my most pressingly problematic phoneme is /r_j/. Soundfiles seem quite clearly to me to show this being a voiced fricative (that I interpret as /Z/, but presumably is more forward than my /Z/), or sometimes what sounds very much to me like /d/... but only sometimes. It seems the official (that is, dictionary, presumably gaelteacht) Munster dialect doesn't do this, but Ulster and Connact seem to do it more, and the forvo climbs from random actual irish people seem to do it even more - maybe this is an L2 thing as well as a dialect thing? I can't work out where the dictionary soundfiles do /Z/, when they do /d/, and when they do /r/, except that the stronger pronunciations are much more common at the ends of words.
Ulster actually has [j] or [ʝ], at least in Gaoth Dobhair. I'm not sure exactly what the Connacht value is.
Salmoneus wrote:I wouldn't be surprised if something was going on here, because how I'm meant to palatalise a tap, I don't know. So I'm to do one thing with the tip of my tongue, moving it quickly near the front of my mouth, while doing something totally different in a totally different part of the mouth with the bit just behind the tip? How is that possible!?
If it helps, this isn't just a problem in Irish. I was reading up a bit about Russian /rʹ/ today and apparently, due to the articulatory difficulties, it's rare for it be tapped more than once.
Salmoneus wrote:Come to think of it, does anybody know what the Irish flap is meant to be anyway? I've always mostly ignored the detail of flaps, but it seems there's quite a variety possible. I can tap against the flat bit behind the teeth, I can flap backward over the ridge, or I can flap forward over the ridge - the latter seems less natural, but apparently it's the only thing that's "officially" a flap. What does Irish use? My instinct has been to try to use the post-dental tap for broad and the ridge-flap for slender, but wikipedia suggests this is the only coronal pair NOT distinguished by POV...
Here's Ó Cuív's description for Muskerry:
[/r/] is a velarized voiced post-alveolar fricative consonant, formed by raising the tip of the tongue towards the back of the teeth-ridge, with simultaneous raising of the back of the tongue toward the soft palate giving a velarized effect.
[/r'/] is a palatalized voiced post-alveolar fricative consonant, formed by raising the tip of the tongue towards the back of the teeth-ridge, with simultaneous raising of the front of the tongue toward the hard palate giving a palatized effect.
He calls the allophones found in slender onsets and broad clusters of all sorts "voiced flapped alveolar consonant[s]". So, yeah, same POA.
Salmoneus wrote:Oh, and don't worry about the first question, i think I've worked it out now. It's because although everyone tells you that broad <bh> is /w/, /w/ is actually [v] (velarised) in word-final (or every) position, yes?
In Connacht and Ulster, from what I understand, this really is a semivowel. Ó Cuív calls Muskerry /v/ and /v'/ "voiced bilabial fricative[s]", with the note that "after the accent (i.e. at the end of a word or syllable), or when it precedes l, n, or r, the friction is quite marked." It only has the semivowel pronunciation "under the accent (i.e. at the beginning of a word or syllable), or following initial g".
Salmoneus wrote:EDIT: ok, saibhir - everyone seems to agree that the first vowel is something like [E]. But it seems like it 'ought' to be [{] at most, and probably a centralised [a]. Am I missing a rule, or is this an exception, do you know?
Saibhir is simply exceptional. For Muskerry, the spelling would lead you to expect */səir'/, but the actual pronunciation is /sev'ir'/ (i.e. [sˠɛβʲɪʑ˔]).

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

linguoboy wrote:
Salmoneus wrote:Come to think of it, does anybody know what the Irish flap is meant to be anyway? I've always mostly ignored the detail of flaps, but it seems there's quite a variety possible. I can tap against the flat bit behind the teeth, I can flap backward over the ridge, or I can flap forward over the ridge - the latter seems less natural, but apparently it's the only thing that's "officially" a flap. What does Irish use? My instinct has been to try to use the post-dental tap for broad and the ridge-flap for slender, but wikipedia suggests this is the only coronal pair NOT distinguished by POV...
Here's Ó Cuív's description for Muskerry:
[/r/] is a velarized voiced post-alveolar fricative consonant, formed by raising the tip of the tongue towards the back of the teeth-ridge, with simultaneous raising of the back of the tongue toward the soft palate giving a velarized effect.
[/r'/] is a palatalized voiced post-alveolar fricative consonant, formed by raising the tip of the tongue towards the back of the teeth-ridge, with simultaneous raising of the front of the tongue toward the hard palate giving a palatized effect.
He calls the allophones found in slender onsets and broad clusters of all sorts "voiced flapped alveolar consonant[s]". So, yeah, same POA.
*bangs head on table*
That's got to just be West muskerry being weird, though, right? I haven't found any soundfile that has velarised /r/ being a fricative...
Salmoneus wrote:Oh, and don't worry about the first question, i think I've worked it out now. It's because although everyone tells you that broad <bh> is /w/, /w/ is actually [v] (velarised) in word-final (or every) position, yes?
In Connacht and Ulster, from what I understand, this really is a semivowel. Ó Cuív calls Muskerry /v/ and /v'/ "voiced bilabial fricative[s]", with the note that "after the accent (i.e. at the end of a word or syllable), or when it precedes l, n, or r, the friction is quite marked." It only has the semivowel pronunciation "under the accent (i.e. at the beginning of a word or syllable), or following initial g".
Again, soundfiles - including focloir - seem to have this as /v/ when final, and wikipedia agrees. W also says that for some munster speakers it can be a bilabial fricative when initial.

Have to say, it would really help if there was some sort of standardised Irish pronunciation, rather than twenty different dialects that seem to be barely mutually comprehensible.
Salmoneus wrote:EDIT: ok, saibhir - everyone seems to agree that the first vowel is something like [E]. But it seems like it 'ought' to be [{] at most, and probably a centralised [a]. Am I missing a rule, or is this an exception, do you know?
Saibhir is simply exceptional. For Muskerry, the spelling would lead you to expect */səir'/, but the actual pronunciation is /sev'ir'/ (i.e. [sˠɛβʲɪʑ˔]).
Great! I mean sure, it's a right bugger having to remember irregular spellings, but it's better than failing to understand the rule in the first place, so that's good to know. And is the same also true of ag? This also seems quite clearly to have [E] in all soundfiles.
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Salmoneus wrote:That's got to just be West muskerry being weird, though, right? I haven't found any soundfile that has velarised /r/ being a fricative...
Actually, modern West Muskerry speakers commonly have a retroflex in those positions where Ó Cuív describes a "fricative". (His account is a century out-of-date, after all.)
Salmoneus wrote:Great! I mean sure, it's a right bugger having to remember irregular spellings, but it's better than failing to understand the rule in the first place, so that's good to know. And is the same also true of ag? This also seems quite clearly to have [E] in all soundfiles.
Yeah, the pronunciation of ag is influenced by (3.SG.M) aige. In Scottish Gaelic it's even spelled aig. Same is true of ar, which is homophonous with air (at least in all the dialects I know).

In Muskerry, a similar thing happens with the preposition roimh, which is /rim'/ presumably under the influence of 3.SG.M /rim'is'/ (CO roimhe).

gestaltist
Lebom
Lebom
Posts: 125
Joined: Fri Feb 06, 2015 5:21 am

Re: Questions not about German Thread

Post by gestaltist »

Salmoneus wrote:Velarization isn't phonemic in my L1 (or any language where I have ever had access to a community of speakers in RL); it should hardly be surprising therefore that I have a problem knowing how to do it, or knowing whether I've done it or not.
I had exposure to native speakers of Russian who velarize their „hard“ consonants. Does your dialect of English distinguish the non-velarized /l/ in onset positions? If it does, try to pick up on the difference and apply it to other consonants. The way I intuitively do it when speaking Russian is to retract my tongue a bit.

Maybe try it this way: place your tongue in position to pronounce [g] and then pronounce another consonant that’s supposed to be velarized. The resulting sound is somewhat „darker“ than a non-velarized consonant.

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: Questions not about German Thread

Post by hwhatting »

I'm currently reading Nicoline v.d. Sijs, De geschiedenis van het Nederlands in een notendop, Amsterdam 2006 ("History of the Dutch language in a nutshell"). On p. 116, she decribes that in the 18th century a differentiation arose in forming relative sentences, using wie + adposition for persons and waar + adposition for things. Then she writes Dit kunstmatige onderscheid woordt momenteel veelvuldig verwaarloosd; in de spreektaal bestaat de regel eigenlijk überhaupt niet. ("This artificial distinction is currently often neglected; in the spoken language this rule actually doesn't exist at all.")
Is that kind of usage of German überhaupt really a thing? I mean, it wouldn't be the first German loan word in Dutch, but it still stands out, not even being adapted to Dutch orthography or calqued from native material like others (e.g. bewusteloos, waanzinnig).

Dē Graut Bʉr
Avisaru
Avisaru
Posts: 593
Joined: Sun Jul 07, 2013 3:03 pm
Location: Nijmegen, Netherlands

Re: Questions not about German Thread

Post by Dē Graut Bʉr »

hwhatting wrote:Is that kind of usage of German überhaupt really a thing?
Yep.

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: Questions not about German Thread

Post by hwhatting »

Dank je wel!

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: Questions not about German Thread

Post by Qwynegold »

I'm looking at these Japanese verb paradigms, and the line [辞書形]だけ is written there twice, with slightly different data. Is this two だけs with different meanings, or what is going on?
Image
My most recent quiz:
Eurovision Song Contest 2018

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

Irish again...

First: despite the long vowel, it seems like most dialects pronounce <damhán> (as in 'damhán alla') as /d@un/. OK, the mh takes precedence here and forces the diphthong. But in Munster, stress is attracted to that long vowel. So my question is, what order does this happen in? Does the diphthong form first, eliminating the long vowel, so stress isn't attracted to it, so Munster has the same pronunciation as everywhere else? Or does stress move to the second syllable first, so the first syllable is unstressed and so doesn't form the diphthong? So in Munster is it /d@un/, or /d@"BAn/? (or something else entirely because this is all impossible to understand?)

Second: <cait>, plural of <cat>. I would expect this to be /cat_j/ (although I think I'll have to have something closer to [{], because distinguishing four subtly different locations of /a/ just doesn't seem feasible to me). However, some people over on Duolingo were saying it's more like /k_eIt_j/ (i.e. treating the <i> as the main vowel and the <a> as just indicating a broad consonant). Is this just nonsense, or is that a dialect form, or what?

Third, and more generally: how OK is it to mix and match dialect features? I think I'd like to use some Munster features, like stress attracted to heavy second syllables, and rounded /a:/, partly because that's the part of the country I have the most connexion to, and partly because it would just make things easier. But I'm not sure I want to go full broad-Munster-dialect, and what's more I'm also tempted to use some specifically non-Munster features, like alveopalatal slender <ll> and <nn> (since these seem standard outside Munster, and because I need everything I can get to try to get clear blue water between the broad and the slender and that seems like an easy way at least for the geminates).
Is that sort of 'all-over-the-place' learner's accent considered OK, or just abominable?
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

Yng
Avisaru
Avisaru
Posts: 880
Joined: Fri Jul 03, 2009 3:17 pm
Location: Llundain

Re: Questions not about German Thread

Post by Yng »

I suspect that just being able to distinguish broad and slender in the first place will probably get you a 'gosh u have a good accent' response
كان يا ما كان / يا صمت العشية / قمري هاجر في الصبح بعيدا / في العيون العسلية

tà yi póbo tsùtsùr ciivà dè!

short texts in Cuhbi

Risha Cuhbi grammar

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Salmoneus wrote:First: despite the long vowel, it seems like most dialects pronounce <damhán> (as in 'damhán alla') as /d@un/. OK, the mh takes precedence here and forces the diphthong. But in Munster, stress is attracted to that long vowel. So my question is, what order does this happen in? Does the diphthong form first, eliminating the long vowel, so stress isn't attracted to it, so Munster has the same pronunciation as everywhere else? Or does stress move to the second syllable first, so the first syllable is unstressed and so doesn't form the diphthong? So in Munster is it /d@un/, or /d@"BAn/? (or something else entirely because this is all impossible to understand?)
(2) with a little bit of (3) (at least in Muskerry). There's a minor exception to the Munster stress shift, to wit: "When the stress is generally on the last syllable of a word, there is a tendency to shift it to the previous syllable when a word with strong initial stress immediately follows." (Ó Cuív, p. 67). So, for instance, cipín beag [ˈcɪpʲiːnʲ ˈbʲɔ̝̈g], even though cipín in other contexts is normally [cɪˈpʲiːnʲ].

So the only thing really inexplicable about the pronunciation of damhán alla in Muskerry is that the first vowel is raised and labialised, i.e. /ˈduvaːn ˈalə/ whereas damh retains the expected /a/. (And perhaps this isn't so inexplicable after all. See p. 39 of Ó Siadhail's Modern Irish for minor rule which raises /a/ to /u/ or /i/ "where there is a long a in the second syllable and no morpheme boundary", e.g. cabáiste > /gubasʹtʹə/ (Cois Fhairrge). He specifically says that this rule doesn't apply to Munster because of the neutralisation of pretonic vowels there, but perhaps it does and is subsequently obscured by it. Blocking the usual stress shift would then prevent vowel reduction from taking place, preserving the raised vowel.)
Salmoneus wrote:Second: <cait>, plural of <cat>. I would expect this to be /cat_j/ (although I think I'll have to have something closer to [{], because distinguishing four subtly different locations of /a/ just doesn't seem feasible to me). However, some people over on Duolingo were saying it's more like /k_eIt_j/ (i.e. treating the <i> as the main vowel and the <a> as just indicating a broad consonant). Is this just nonsense, or is that a dialect form, or what?
Dunno. AFAIK, in Munster it's [ˈka̠tʲ]. Historically, there are several cases where /a/ before a slender consonant has been raised and fronted to /i/ (e.g. glas "grue", comparative /glisʹə/ [as if *gluise), but I've never heard that this is one of them.
Salmoneus wrote:Third, and more generally: how OK is it to mix and match dialect features? I think I'd like to use some Munster features, like stress attracted to heavy second syllables, and rounded /a:/, partly because that's the part of the country I have the most connexion to, and partly because it would just make things easier. But I'm not sure I want to go full broad-Munster-dialect, and what's more I'm also tempted to use some specifically non-Munster features, like alveopalatal slender <ll> and <nn> (since these seem standard outside Munster, and because I need everything I can get to try to get clear blue water between the broad and the slender and that seems like an easy way at least for the geminates).
Is that sort of 'all-over-the-place' learner's accent considered OK, or just abominable?
By whom? I think in general, native speakers will just be impressed if you aim for any sort of non-anglicised pronunciation. As Yng says, they're used to having people ignore the broad/slender distinction entirely that they're not going to cavil. As for learners, most of them don't even know where the features they use come from. Most all the L2 speakers I've talked to have had teachers from at least two different dialect regions and since pronunciation doesn't seem to be taught in any kind of structured way they just imitate what they hear. So they might use a Connacht pronunciation with one word and a Donegal pronunciation of another within the same sentence, depending on who they learned them from, but more likely the whole thing will be said with such a heavy English accent that you'd be hard-pressed to detect any other sort of blas at all.

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

Thanks guys.
linguoboy wrote:There's a minor exception to the Munster stress shift, to wit: "When the stress is generally on the last syllable of a word, there is a tendency to shift it to the previous syllable when a word with strong initial stress immediately follows." (Ó Cuív, p. 67). So, for instance, cipín beag [ˈcɪpʲiːnʲ ˈbʲɔ̝̈g], even though cipín in other contexts is normally [cɪˈpʲiːnʲ].

So the only thing really inexplicable about the pronunciation of damhán alla in Muskerry is that the first vowel is raised and labialised, i.e. /ˈduvaːn ˈalə/ whereas damh retains the expected /a/. (And perhaps this isn't so inexplicable after all. See p. 39 of Ó Siadhail's Modern Irish for minor rule which raises /a/ to /u/ or /i/ "where there is a long a in the second syllable and no morpheme boundary", e.g. cabáiste > /gubasʹtʹə/ (Cois Fhairrge). He specifically says that this rule doesn't apply to Munster because of the neutralisation of pretonic vowels there, but perhaps it does and is subsequently obscured by it. Blocking the usual stress shift would then prevent vowel reduction from taking place, preserving the raised vowel.)
OK, I think I get that.

I think in general, native speakers will just be impressed if you aim for any sort of non-anglicised pronunciation. As Yng says, they're used to having people ignore the broad/slender distinction entirely that they're not going to cavil. As for learners, most of them don't even know where the features they use come from. Most all the L2 speakers I've talked to have had teachers from at least two different dialect regions and since pronunciation doesn't seem to be taught in any kind of structured way they just imitate what they hear. So they might use a Connacht pronunciation with one word and a Donegal pronunciation of another within the same sentence, depending on who they learned them from, but more likely the whole thing will be said with such a heavy English accent that you'd be hard-pressed to detect any other sort of blas at all.
Good to know!
It's only hypothical, obviously, given that a) I'm not going to learn this well enough to speak to anyone anyway, and b) even if I learned it, I'd probably never be in a situation where I'd be with anyone I could speak it. But, you know, may as well do it right, or at least not completely wrong.
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
Pole, the
Smeric
Smeric
Posts: 1606
Joined: Sat Feb 11, 2012 9:50 am

Re: Questions not about German Thread

Post by Pole, the »

I have a question: is Welsh a real language or was it just created by the Basque monks so Basque wouldn't look so weird?
The conlanger formerly known as “the conlanger formerly known as Pole, the”.

If we don't study the mistakes of the future we're doomed to repeat them for the first time.

Yng
Avisaru
Avisaru
Posts: 880
Joined: Fri Jul 03, 2009 3:17 pm
Location: Llundain

Re: Questions not about German Thread

Post by Yng »

Pole, the wrote:I have a question: is Welsh a real language or was it just created by the Basque monks so Basque wouldn't look so weird?
8)

welsh is so not weird by IE standards even
كان يا ما كان / يا صمت العشية / قمري هاجر في الصبح بعيدا / في العيون العسلية

tà yi póbo tsùtsùr ciivà dè!

short texts in Cuhbi

Risha Cuhbi grammar

Vijay
Smeric
Smeric
Posts: 2244
Joined: Sat Feb 06, 2016 3:25 pm
Location: Austin, TX, USA

Re: Questions not about German Thread

Post by Vijay »

Pole, the wrote:I have a question: is Welsh a real language or was it just created by the Basque monks so Basque wouldn't look so weird?
But Basque still looks so weird! ;) Does that stop me from wanting to or even trying to learn it anyway? Nope.

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Yng wrote:
Pole, the wrote:I have a question: is Welsh a real language or was it just created by the Basque monks so Basque wouldn't look so weird?
welsh is so not weird by IE standards even
People really seem to get hung up on the orthography. Much as they do with Irish.

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

A similar Munster Stress question...

Do diphthongs count as heavy for the purposes of attracting stress?
And do diphthongs and/or long vowels not shown as such but emerging before final geminate sonorants or clusters still attract stress, or does this process happen after the stress shift?

for instance, verbs. caitheann sibh - is the first syllable stress and the second schwa, or is the second stressed and long/diphthongised?
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Salmoneus wrote:Do diphthongs count as heavy for the purposes of attracting stress?
Yes.
Salmoneus wrote:And do diphthongs and/or long vowels not shown as such but emerging before final geminate sonorants or clusters still attract stress, or does this process happen after the stress shift?
These diphthongs only develop from stressed vowels anyway. So stress assignment not only precedes this process, it determines it. (Not all examples involve final syllables, e.g. francach > /ˈfraunkəx/, not */frənˈkax/.)
Salmoneus wrote:for instance, verbs. caitheann sibh - is the first syllable stress and the second schwa, or is the second stressed and long/diphthongised?
Well, verbs are a special case wrt to stress assignment anyway. Stress remains on the root even before endings that are inherently long, e.g. ˈbheidís, ˈcasfaí.

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

linguoboy wrote:
Salmoneus wrote:Do diphthongs count as heavy for the purposes of attracting stress?
Yes.
Salmoneus wrote:And do diphthongs and/or long vowels not shown as such but emerging before final geminate sonorants or clusters still attract stress, or does this process happen after the stress shift?
These diphthongs only develop from stressed vowels anyway. So stress assignment not only precedes this process, it determines it. (Not all examples involve final syllables, e.g. francach > /ˈfraunkəx/, not */frənˈkax/.)
Well, I knew that stress determined it, but I didn't know whether it was 'original' stress or shifted stress... (I guess, whether the diphthongisation/lengthening was the result of a non-productive diachronic process or an ongoing productive synchronic one). So thanks for making it clear.
Wait, there's a diphthong in francach? Is there any reason why? I thought it was only before nn/ll/rr and final m? [While we're at it: wikipedia says that /a/ also lengthens before rd, rn and rl, but the page on munster irish only mentions it diphthongising instead of lenghthening before nn/rr/ll and final m... so before rf, rn and rl, is it /a:/ or /au/?
Salmoneus wrote:for instance, verbs. caitheann sibh - is the first syllable stress and the second schwa, or is the second stressed and long/diphthongised?
Well, verbs are a special case wrt to stress assignment anyway.
*sigh*
Of course they would be...
Stress remains on the root even before endings that are inherently long, e.g. ˈbheidís, ˈcasfaí.
OK, that makes that simpler I guess.

Thanks again.
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Salmoneus wrote:
linguoboy wrote:
Salmoneus wrote:And do diphthongs and/or long vowels not shown as such but emerging before final geminate sonorants or clusters still attract stress, or does this process happen after the stress shift?
These diphthongs only develop from stressed vowels anyway. So stress assignment not only precedes this process, it determines it. (Not all examples involve final syllables, e.g. francach > /ˈfraunkəx/, not */frənˈkax/.)
Well, I knew that stress determined it, but I didn't know whether it was 'original' stress or shifted stress...
If there's no diphthong, then there's nothing to attract the stress.
Salmoneus wrote:Wait, there's a diphthong in francach? Is there any reason why? I thought it was only before nn/ll/rr and final m? [While we're at it: wikipedia says that /a/ also lengthens before rd, rn and rl, but the page on munster irish only mentions it diphthongising instead of lenghthening before nn/rr/ll and final m... so before rf, rn and rl, is it /a:/ or /au/?
The lengthening of vowels before fortis /r/ and some rC clusters[*] is probably best treated as a separate process from the lengthening/diphthongisation before fortis nasals and /l/ (not least of all because it alternates with epenthesis in a few words, e.g. [fʲaːɾˠ] and [ˈfʲaɾˠə] for fearr; cf. dorn [ˈd̪ˠɔɾən̪ˠ] vs doirne [ˈd̪ˠiːɾˠnʲɪ]).

When final, fortis sonorants are indicated in the orthography by doubling. But they also occur medially before plosives, in which case the spelling is morphophonemic. For example, seanda (< sean + -da) but ceanndána (< ceann + dána). Medial ng is inherently ambiguous, but in most cases this represents lenis [ŋ] rather than fortis [ŋg]. (It helps to know if the word is historically monomorphemic or not. So lenis in teanga, but fortis in angar from amh- + gar.)

So, back to francach. The /ŋ/ here is fortis and the word was historically written frangcach to make that absolutely clear. But clearing away "unnecessary" letters like this was one of the aims of the mid-20th-century spelling reform.

[*] Mainly /rd/, occasionally /rs/, plus /rn/ and /rl/ when medial. I can't think of any examples before /rf/; do you have any?

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

Thanks. (and rf was a typo for rd).

To clarify: with verbs, does this mean stress also doesn't shift to the -íonn (etc) in some present tense verbs? And can stress still shift to the second syllable of a two-syllable root, or is it just fixed in verbs entirely?

Oh, and while I'm at it, I've reached a few words with seemingly silent letters, like codlach and tatnaíonn. Am I right in thinking that these are just there for etymological/morphological reasons, and aren't pronounced?
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Salmoneus wrote:To clarify: with verbs, does this mean stress also doesn't shift to the -íonn (etc) in some present tense verbs? And can stress still shift to the second syllable of a two-syllable root, or is it just fixed in verbs entirely?
It shifts to the second syllable of a two-syllable root where appropriate, and that's exactly what's happening in this case. The ending is actually -(e)ann. But there's a historical change of -ighe- to [iː] which (as I assume you know) led to be it being eliminated from the "simplified" orthography.

So, for instance, the form tosnaíonn [Standard tosaíonn], whose pre-CO orthography tosnuigheann shows that the root is actually tosnuigh [Standard tosaigh]. Similarly, the future form tosnóidh is stressed finally, not initially.
Salmoneus wrote:Oh, and while I'm at it, I've reached a few words with seemingly silent letters, like codlach and tatnaíonn. Am I right in thinking that these are just there for etymological/morphological reasons, and aren't pronounced?
Half right. Codladh (for which I'm assuming codlach is a typo) is the result of syncopation of the stem codail following addition of the verbal noun suffix -adh. The d assimilates and I assume that varieties which maintain a distinction between the lenis and fortis sonorants in medial position (something foreign to Munster) would treat this word as if spelled *colladh.

AFAIK, *tatnaíonn doesn't exist; the CO form is taitníonn. In West Muskerry, however, the second vowel is short. On top of that, the t is lenited and metathesis takes place, which fortisises the sonorant in the stem, i.e. taitneann > *taithneann > *tainntheann (i.e. /ˈtaŋˊhən/). Cf. cruithneacht "wheat", which is /kriŋˊˈhaxt/ (as if *cruinntheacht).

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Questions not about German Thread

Post by Salmoneus »

linguoboy wrote:
Salmoneus wrote:To clarify: with verbs, does this mean stress also doesn't shift to the -íonn (etc) in some present tense verbs? And can stress still shift to the second syllable of a two-syllable root, or is it just fixed in verbs entirely?
It shifts to the second syllable of a two-syllable root where appropriate, and that's exactly what's happening in this case. The ending is actually -(e)ann. But there's a historical change of -ighe- to [iː] which (as I assume you know) led to be it being eliminated from the "simplified" orthography.

So, for instance, the form tosnaíonn [Standard tosaíonn], whose pre-CO orthography tosnuigheann shows that the root is actually tosnuigh [Standard tosaigh]. Similarly, the future form tosnóidh is stressed finally, not initially.
Ahh! That's very helpful!
I mean, you were very helpful. The spelling reform seems to just have been designed to be unhelpful...
Salmoneus wrote:Oh, and while I'm at it, I've reached a few words with seemingly silent letters, like codlach and tatnaíonn. Am I right in thinking that these are just there for etymological/morphological reasons, and aren't pronounced?
Half right. Codladh (for which I'm assuming codlach is a typo) is the result of syncopation of the stem codail following addition of the verbal noun suffix -adh. The d assimilates and I assume that varieties which maintain a distinction between the lenis and fortis sonorants in medial position (something foreign to Munster) would treat this word as if spelled *colladh.
Right. [yes, typo]
AFAIK, *tatnaíonn doesn't exist; the CO form is taitníonn. In West Muskerry, however, the second vowel is short. On top of that, the t is lenited and metathesis takes place, which fortisises the sonorant in the stem, i.e. taitneann > *taithneann > *tainntheann (i.e. /ˈtaŋˊhən/). Cf. cruithneacht "wheat", which is /kriŋˊˈhaxt/ (as if *cruinntheacht).
...
...
*bangs head*

[Yes, stupid typo again - although I'm kind of impressed with my subconscious for autocorrecting to maintain the broad-with-broad rule...]
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions not about German Thread

Post by linguoboy »

Salmoneus wrote:I mean, you were very helpful. The spelling reform seems to just have been designed to be unhelpful...
Well, unless the morphophonology of your language is dead simple, any orthography is going to have to strike some balance between morphological transparency and phonemicity. The alternatives here would've been to (a) maintain historical -ighe- throughout or (b) maintain it only when there were relevant morphophonemic alternations. Either of those choices requires one to learn that /iː/ is sometimes spelled í and sometimes ighe. Is that easier than learning that the -(e)ann ending alternates with -íonn on certain verbs? For some people, yes, for others maybe no.

My biggest beef with the spelling reform is that they didn't set out to construct a truly pandialectal orthography, instead choosing one which systematically favours the pronunciations of (certain) Connacht varieties. So you have cases like CO trá, crua, and aibí where the Munster pronunciations have a final [ɟ] (representing historical igh or idh, i.e. pre-reform tráigh, cruaidh, aibidh) which is not represented at all in the orthography. To me, this is quite different from a case like, say buíochas, where the Munster form shows an unexpected change to /eː/. The change of slender final gh or dh to /gˊ/ in Munster is as regular as their loss or the change to /iː/ in other dialects and should have been treated as such.

Post Reply