There are 15 messages in this issue.
Topics in this digest:
1a. Re: the symmetry of sound change
From: Leonardo Castro
1b. Re: the symmetry of sound change
From: Matthew Boutilier
1c. Re: the symmetry of sound change
From: Matthew Boutilier
2a. Re: Emphatic "even" in nat and conlangs.
From: Daniel Prohaska
2b. Re: Emphatic "even" in nat and conlangs.
From: David McCann
3a. Re: London meet?
From: Scar Cvxni
4a. Re: software vocal tract models?
From: Eric Christopherson
5a. Re: Equivalent to Grand Master Plans in Proper Linguistics?
From: Eric Christopherson
6. THEORY: Likelihood of special interaction of VC[+glottal]V
From: Eric Christopherson
7. Construct state markers without overt possessor marking
From: Eric Christopherson
8a. THEORY: Sound change of vowel-like sound into rhotic
From: Eric Christopherson
8b. Re: THEORY: Sound change of vowel-like sound into rhotic
From: Matthew Boutilier
9.1. Re: writing (almost) entirely in lower-case letters
From: BPJ
10. Fw: the symmetry of sound change
From: Roger Mills
11. History of some spelling pronunciations (Was: Equivalent to Grand Ma
From: Jyri Lehtinen
Messages
________________________________________________________________________
1a. Re: the symmetry of sound change
Posted by: "Leonardo Castro" [email protected]
Date: Sat Jun 22, 2013 7:50 am ((PDT))
I have a similar question in my conlang. As it is designed to be
simple, I initially chose to include only the following consonants:
labial - coronal - guttural
stop p t k
fricative f s h
nasal m n
lateral l
By guttural I mean everything "back" in the oral cavity. The only
reason to exclude /N/ is that I thought that it is possible more
difficult to pronounce than the others (more probably in the beginning
of syllables), but I'm not sure about that.
Até mais!
Leonardo
2013/6/22 Jyri Lehtinen <[email protected]>:
> You are right that it's not easy to justify voicing /s/ but not /ʃ/ and /ɬ/
> unless you assume some extra twists to make /s/ more distinct from the rest
> two. Then again it might not be such a huge crime to have no clear reason
> at all for the change to only affect /s/, at least if your sense of
> aesthetics allows that.
>
> but i'm not so sure i want /K/ and /S/ to follow suit, mainly since [K\]
>> sounds a hell of a lot like [l], which i have (which itself devoices to [K]
>> word-finally!), and because /Z/ is one of my least favorite phonemes.
>>
>
> It might be a bit easier to justify leaving the lateral fricative outside
> the voicing but let the voicing affect both /s/ and /ʃ/. You can easily get
> rid of the nasty /ʒ/ by leniting it quickly into /j/. This is an
> interesting change because if you would decide to apply a similar voicing
> process for the lateral fricative as well you'd likely get something like /ɬ/
>> /ɮ/ > /l/. With what you describe, this would lead you to have either [l]
> and [ɬ] as allophones of a single lateral consonant or only marginally
> distinct phonemes. Especially the route of two marginally distinct phonemes
> would be nice to see experimented in some dialect of a parallel closely
> related language.
>
> -Jyri
Messages in this topic (9)
________________________________________________________________________
1b. Re: the symmetry of sound change
Posted by: "Matthew Boutilier" [email protected]
Date: Sat Jun 22, 2013 11:38 am ((PDT))
On Sat, Jun 22, 2013 at 3:00 AM, George Corley <[email protected]> wrote:
> Hmm, it looks like you're right. Ah well. In any case /s/ by itself is a
> natural class, so it's theoretically possible for it to have a sound change
> that no other sound has. But it seems a little odd to me that it would be
> affected and other sibilants wouldn't.
>
>
> On Sat, Jun 22, 2013 at 2:12 AM, Galen Buttitta <
> [email protected]> wrote:
>
> > According to my Phonology class workbook, /S/ is [+strident]. /K/ is
> > [-strident].
> >
>
hm, that's interesting. the [+strident] thing hadn't even occurred to me.
perhaps i could *only* voice the [+strident] consonants (the "true
sibilants"?) and then have /Z/ > /j/, which would be a merger.
On Sat, Jun 22, 2013 at 6:06 AM, Padraic Brown <[email protected]> wrote:
> Well, then it's a perfect time for a little in mediis conlanging. If you
> want
> v.l. sibilants to become voiced, are okay with /s/ doing so but also are
> not so fond of /K/ and /S/ doing so, just set your temporal quandoquecator
> for some unspecified time after the series of changes has started (with
> /s/)
> but hasn't fully or even yet affected the other sounds.
ah, the old *in medio linguifaciendo*.
> It's not like the whole
> English nation woke up one morning and, looking at their calendars, said
> "oh bosh! Twenty second June, so today's the day we're all supposed to
> switch over to the glo''al stop!" Change is gradual and there's no reason
> a conlanger can't work on his conlang mid shift!
> >
>
your vignette is so entertaining i'm inevitably going to imagine sound
change like this from now on. but i appreciate your point.
On Sat, Jun 22, 2013 at 6:34 AM, Jyri Lehtinen <[email protected]>wrote:
> It might be a bit easier to justify leaving the lateral fricative outside
> the voicing but let the voicing affect both /s/ and /ʃ/. You can easily get
> rid of the nasty /ʒ/ by leniting it quickly into /j/. This is an
> interesting change because if you would decide to apply a similar voicing
> process for the lateral fricative as well you'd likely get something like
> /ɬ/
> > /ɮ/ > /l/. With what you describe, this would lead you to have either [l]
> and [ɬ] as allophones of a single lateral consonant or only marginally
> distinct phonemes. Especially the route of two marginally distinct phonemes
> would be nice to see experimented in some dialect of a parallel closely
> related language.
>
>
YES. i've been considering something like this. well, the above discussion
has illuminated a possible divorce of /K/ from the other two "sibilants,"
and it would really be easy for /S/ to become /Z/ > /j/.
and for a long time i've considered taking the voiced reflex of /K/ and
merging it with /l/; then while they both would appear as [l] in some
environments and [K] in others, they'd basically be one phoneme (like [G\]
and [G] in modern Persian, as i understand it).
i could also have the original voiceless-sibilants as allophones of the *
geminate* sibilants (since they wouldn't've voiced in those environments).
e.g. /z:/ = [s(:)]. and ... OOOOOHH ... /l:/ = [K(:)]. hot damn, today is a
good day.
anyway, this merging idea is cool, because i've been wanting to add some
non-phonemic archaisms to my language's orthography (an abugida). if /K/
and /l/ merge, i can keep them orthographically distinct to make for a
little homophony.
same with /j/ and /S/ - i can have one /j/ that alternates as [S] and one
/j/ that does not. sort of like modern Hebrew where *qof* is always [k] and
*chet* always [x], but *kaf* is [k] or [x] depending on environment.
back to the drawing board. thanks to all responders
matt
Messages in this topic (9)
________________________________________________________________________
1c. Re: the symmetry of sound change
Posted by: "Matthew Boutilier" [email protected]
Date: Sat Jun 22, 2013 12:03 pm ((PDT))
On Sat, Jun 22, 2013 at 9:49 AM, Leonardo Castro <[email protected]>wrote:
> I have a similar question in my conlang. As it is designed to be
> simple, I initially chose to include only the following consonants:
>
> labial - coronal - guttural
> stop p t k
> fricative f s h
> nasal m n
> lateral l
>
> By guttural I mean everything "back" in the oral cavity. The only
> reason to exclude /N/ is that I thought that it is possible more
> difficult to pronounce than the others (more probably in the beginning
> of syllables), but I'm not sure about that.
>
>
> Até mais!
>
> Leonardo
>
>
this is certainly not strange at all. in fact from a typological standpoint
it's a lot weirder to *have* /N/, for whatever reason.
all the languages that *do* have /N/ seem to either develop it from /ng/ >
[Ng] > /N/ (thus generally restricted to syllable codas) or inherit it from
Proto-Austronesian or something.
matt
Messages in this topic (9)
________________________________________________________________________
________________________________________________________________________
2a. Re: Emphatic "even" in nat and conlangs.
Posted by: "Daniel Prohaska" [email protected]
Date: Sat Jun 22, 2013 8:56 am ((PDT))
Cornish has a number of translations for even depending on the context. I found
them in the dictionaries by N.Williams and R. Gendall:
‹Kepar ev del devis, yndella an bows a wre› "even as he grew, so did the coat"
like he that grew (3sg.pret.) so the coat (rel.part.) did (impf.)
‹A ny wodhves unn prysweyth golyas.› "Could you not watch even for a moment."
(interrog.part.) (neg.part.) know (2sg.impf.sbj.) one instant watching (vb.n.)
‹Ty ynwedh a yll convedhes henna.› "Even you can see/understand that."
thou too (rel.part.) can (3sg.pres.fut.) understanding (vb.n.) that (m.).
‹Ny wonn y wul kyn fen ledhys› "I can't do it even if I died/I can't do it for
the life of me"
(neg.part) know-I (1sg.pres.) its doing (vb.n.) though be-I (1sg.impf.sbj.)
killed (vb.adj.)
‹Ha mar teffewgh whei ha ponya eneth.› "Even if you run."
And if come-you (2pl.impf.sbj.) you (pl.) and running (vb.n.) once
‹Me a vedn mos dei wos merwel.› "I'll go there even if I die."
I (rel.part.) will (3sg.pres.) going (vb.n.) thither because-of dying (vb.n.)
‹Nei a dal tria ken ven nehys.› "We must try even if we get refused."
we (rel.part.) must (3sg.pres.) trying (vb.n.) though be-we (1pl.impf.sbj.)
denied (vb.adj.)
‹Whath lakka vel hedna.› or: ‹Whath moy lakka vel hedna.› "Even worse than
that."
still/yet worse than that (m.); or: yet more worse than that (m.)
‹Na alje'ma anella eneth.› "I couldn't even breathe."
(neg.part) could (1sg.cnd.) breathe (vb.n.) once
‹Whath moy.› "even more"
yet more
‹Ken na'm bo'ma lowena 'nodho.› "Even if I get no joy out of it."
though (neg.part.) my (infixed prn.) be/have-I (1sg.pres.fut.sbj.) joy of-him
Dan
On Jun 22, 2013, at 11:35 AM, R A Brown wrote:
> On 21/06/2013 20:44, H. S. Teoh wrote:
>> On Fri, Jun 21, 2013 at 07:35:00PM +0100, R A Brown
>> wrote:
>>> On 21/06/2013 17:16, Daniel Bowman wrote:
> [snip]
>>> Interestingly, there is _no_ word in the Hebrew
>>> corresponding to "even" of the King James version (and
>>> several other, but not all, English versions):
>>> http://interlinearbible.org/isaiah/51-12.htm
>>
>> Now *that's* interesting!! I was just wondering exactly
>> the same thing -- what does the Hebrew say? It's
>> interesting that the pronoun is just repeated!
>
> Yes.
>
>> In that sense, it's not really a mirative as the English
>> would suggest, but IMO more of directing the attention
>> of the audience towards oneself, or emphasis on oneself.
>
> I thinks so.
>
>>> I thought I'd check other ancient versions. The
>>> Septuagint has: ἐγώ εἰμι ἐγώ εἰμι ὁ παρακαλῶν σε ... I
>>> am I am the [one] comforting you ...
>>>
>>> Once again no word corresponding to "even." :)
>>
>> That's curious, though. What does the repeated "I am"
>> signify in the Greek?
>
> My guess is that the translators found simply repeating ἐγώ
> (egṓ) simply "un-Greek" (ἐγὼ ἐγώ εἰμι ὁ παρακαλῶν ..). ἐγώ
> needs a verb, so I guess they found repeating subject and
> verb more acceptable. it has IMO the same effect: it
> directs attention to the subject - "I am, _I am_ the one
> comforts ...)
>
> [snip]
>>>
>>> The use of "even" in the various English versions
>>> seems to be to emphasize "I" which is simply repeated
>>> in the Hebrew version. One could have, I guess: "I,
>>> 'tis I, who comforts you ..."
>>
>> I think I like "I, 'tis I" better, in the sense of
>> closer correspondence with the original language.
>
> I think so, but I guess too colloquial (at that time) for
> the KJV translators.
>
> [snip]
>>>
>>> In other words, in those two languages there is nothing
>>> corresponding to the English "even" - we are just
>>> saying "I am I who ..."
>>
>> This is utterly fascinating. Why is it "I am I who ...
>> " rather than "I am he who ..."?
>
> Because "he" is specifically third person and I think in
> these two languages (TAKE & Outidic) the first person
> subject would attract the antecedent of the relative to
> first person as well.
>
> --
> Ray
> ==================================
> http://www.carolandray.plus.com
> ==================================
> "language … began with half-musical unanalysed expressions
> for individual beings and events."
> [Otto Jespersen, Progress in Language, 1895]
Messages in this topic (19)
________________________________________________________________________
2b. Re: Emphatic "even" in nat and conlangs.
Posted by: "David McCann" [email protected]
Date: Sat Jun 22, 2013 9:11 am ((PDT))
On Fri, 21 Jun 2013 14:49:29 -0300
Leonardo Castro <[email protected]> wrote:
> Well, as we see, what I call "emphatic even" could mean "oneself",
> "although", "on the contrary of one could expect", "only", "by myself"
> (?), etc. It's interesting that the Portuguese word "mesmo" also can
> be used with all these senses.
>
> "Mesmo chovendo, ele foi à festa." (Even raining, he went to the
> party.) "Eu o fiz, eu mesmo." (I did it by myself, by my own means.)
> "Eles foram mesmo lá." (They were really there [maybe against what was
> expected to happen].)
>
> So, I wonder if such a word is common in languages around the world.
We're talking about *focus particles* here:
Additive: also, even, -self
Particularising: especially
Exclusive: only, purely
Identity: exactly
Many languages have common strategies. The Latin use of 'et' for 'and'
and 'also' is paralleled in some Uralic languages. Constructions like
'all but he had fled' occur in Nahuatl and Thai.
Messages in this topic (19)
________________________________________________________________________
________________________________________________________________________
3a. Re: London meet?
Posted by: "Scar Cvxni" [email protected]
Date: Sat Jun 22, 2013 9:11 am ((PDT))
If you're arranging a London meet I'd be interested in getting together
with other UK conlangers, I'm only just starting out and would love to meet
some people within the community.
Scar
On 21 June 2013 18:50, Dirk Elzinga <[email protected]> wrote:
> This summer I am directing our dept's study abroad program in the British
> Isles. After a whirlwind tour of Ireland and Wales, we will be staying in
> London from 4-28 July. It would be fun to meet some UK conlangers while I'm
> there. My time is not entirely my own, but I will be free most weekday
> afternoons and Saturdays while in London. Let me know by private email (so
> as not to clutter the list) if you're interested and have time to meet up.
>
> Dirk
>
Messages in this topic (3)
________________________________________________________________________
________________________________________________________________________
4a. Re: software vocal tract models?
Posted by: "Eric Christopherson" [email protected]
Date: Sat Jun 22, 2013 10:02 am ((PDT))
On Jun 17, 2013, at 11:59 PM, Alex Fink <[email protected]> wrote:
> I have some ideas for a random phonology generator which would
> overcome certain features I came to find limiting about the model I
> used in Gleb.
>
> I understand that software models of the vocal tract exist, that allow
> you to specify the positions over time of the various articulators and
> then compute what the resulting frequencies / airflows / whatever are.
[...]
> Does anyone happen to know of one like this? Or indeed anything about
> what the modern state of affairs in the field is?
I don't, unfortunately, although I'm very interested. I'd particularly like
some code that can model likely vs. unlikely sound changes. (Perhaps I could go
through _Evolutionary phonology_ and build one myself without recourse to a
vocal tract model.)
Messages in this topic (2)
________________________________________________________________________
________________________________________________________________________
5a. Re: Equivalent to Grand Master Plans in Proper Linguistics?
Posted by: "Eric Christopherson" [email protected]
Date: Sat Jun 22, 2013 10:09 am ((PDT))
On Jun 20, 2013, at 6:41 AM, Jyri Lehtinen <[email protected]> wrote:
> One is whether the reflex of the weak grade of /t/ is /d/ as in the
> standard language of loss as in the eastern dialect area and the other is
> whether the standard language /ts/ gets replaced by /t:/ which is typical
> to many western dialects. For commonly used words I tend to use exclusively
> the dialectal variants unless I aim for a hyper corrected register to put
> someone off. Uncommon words haven't implemented this spread of dialectal
> features and get the reflexes of the standard language. There is however a
> middle ground between the two extremes where I'm never certain which reflex
> to choose and the choice depends sensitively from the register of the
> discourse. These examples are somewhat complicated by the fact that both of
> the standard language reflexes are in fact historic spelling
> pronunciations, but that's beside the point. I'm pretty certain that you
> can find similar partial sound changes from your own idiolects as well.
I'm curious as to what the "target" (I'm not sure of the correct term) of these
spelling pronunciations was; did Finnish itself have <d> and <ts> pronounced as
/d/ and /ts/? -- or did those spellings only occur in surrounding Germanic
languages, and it was from there that the spelling pronunciation caused the
*Finnish* graphemes to be pronounced likewise?
I know <d> is a grade variant of /t/; does <ts> occur in native words?
Messages in this topic (15)
________________________________________________________________________
________________________________________________________________________
6. THEORY: Likelihood of special interaction of VC[+glottal]V
Posted by: "Eric Christopherson" [email protected]
Date: Sat Jun 22, 2013 10:16 am ((PDT))
I remember reading in descriptions of at least two languages that, in those
particular languages, vowels separated by glottal sounds (i.e. [?] or [h])
often undergo total assimilation or, in at least one of those languages,
metathesis.
1. Is this fairly common cross-linguistically?
2. Is there any reason to suppose transglottal interactions of these sorts
would be more likely (either in specific languages or crosslinguistically) than
in pure sequences of two vowels?
2a. Or two vowels separated by some other kind of consonant?
I know that the "consonants" [?] and [h] are actually not quite consonantal
according to some analyses (and I think depending on the specific language);
and they actually affect surrounding vowels where e.g. [t] would not; so
perhaps this makes a difference.
Messages in this topic (1)
________________________________________________________________________
________________________________________________________________________
7. Construct state markers without overt possessor marking
Posted by: "Eric Christopherson" [email protected]
Date: Sat Jun 22, 2013 10:28 am ((PDT))
Hello, all. I'm wondering:
1. If it's attested for languages with construct state (e.g. Semitic languages)
to allow a noun phrase to appear *in* the construct state but *without* an
overt possessor (either an NP or possessive affix); and
2. What the semantics of that sort of construction might be.
The only language I'm familiar with for which #1 is true, *possibly*, is Ainu:
some Ainu nouns have suffixes that indicate that they are inalienably
possessed. There are personal affixes for 1st and 2nd persons singular and
plural (as well as an impersonal "person" which is often used for 1st plural
and, in mythology, singular). But Ainu has no 3rd-person affixes, and nouns
with this "possessed" suffix with no personal affix are interpreted as being
"his"/"her"/"their" whatever. I'm just not sure if this type of noun is exactly
the construct state, as it is understood in other languages.
For #2, I would hypothesize the existence of 3rd-person interpretations, as in
Ainu, but I have wondered too if a noun in construct state without overt
possessor marking might be construed in some languages some other way, e.g. as
being simply definite, or possessed by the 1st or 2nd person. In one of my
conlangs in progress, it's occurred to me to have the usual
non-overtly-possessed construct NP be interpreted as 3rd-person-possessed, e.g.
father-CONS "his/her/their father", but in the vocative have it be 1st-person,
e.g. father-CONS-VOC "O my father".
Thoughts?
Messages in this topic (1)
________________________________________________________________________
________________________________________________________________________
8a. THEORY: Sound change of vowel-like sound into rhotic
Posted by: "Eric Christopherson" [email protected]
Date: Sat Jun 22, 2013 11:19 am ((PDT))
There are several languages where syllable-final rhotics have developed into
vowel-like sounds and sometimes dropped out, e.g. English and German. Does it
ever happen that vowels develop into rhotics?
I am thinking of "rhotic" as a really broad group, including things that
actually are vowels/approximant like the *pre*vocalic resonant in English; but
I'm especially interested in development of vowels > vowel-like rhotics >
rhotic taps, flaps, trills, etc.
Messages in this topic (2)
________________________________________________________________________
8b. Re: THEORY: Sound change of vowel-like sound into rhotic
Posted by: "Matthew Boutilier" [email protected]
Date: Sat Jun 22, 2013 11:45 am ((PDT))
does conlang data count? i have something vocalic that becomes something
rhotic (well, at least i transcribe it as a rhotic).
i have a sound change where uvulars become velars, and leave a residual
offglide before (long) non-back vowels, which eventually develops into a
uvular R, basically.
*qiʔkɑn > *qīkan > *qʁīkan > kʁīkan 'i drank'
*χāθæn > *χʁāθə > *hʁāθə > ʁāθə 'bottle'
matt
On Sat, Jun 22, 2013 at 1:19 PM, Eric Christopherson <[email protected]>wrote:
> There are several languages where syllable-final rhotics have developed
> into vowel-like sounds and sometimes dropped out, e.g. English and German.
> Does it ever happen that vowels develop into rhotics?
>
> I am thinking of "rhotic" as a really broad group, including things that
> actually are vowels/approximant like the *pre*vocalic resonant in English;
> but I'm especially interested in development of vowels > vowel-like rhotics
> > rhotic taps, flaps, trills, etc.
Messages in this topic (2)
________________________________________________________________________
________________________________________________________________________
9.1. Re: writing (almost) entirely in lower-case letters
Posted by: "BPJ" [email protected]
Date: Sat Jun 22, 2013 11:29 am ((PDT))
2013-06-21 16:52, Michael Everson skrev:
> On 18 Jun 2013, at 05:58, Sasha Fleischman <[email protected]> wrote:
>
>> For my proto-language that I'm working on, I initially didn't
>> do capitalization because my romanization is essentially IPA,
>> and there's no capital ɦ.
>
> Yes, there is, U+A7AA LATIN CAPITAL LETTER H WITH HOOK, Ɦ.
Yes, but WTH is the hook doing on the *left* side? I'd expected
it to look like Ᏺ U+13F2 CHEROKEE LETTER YO, which I admit to
have used to fudge a capital ɦ. (Iˈve even used Ꮅ U+13B5 CHEROKEE
LETTER LI to fudge a capital ſ, though I prefer the insular S now! :-)
>
>> Since I started working on the language, I've gotten rid of that phoneme,
>> but the lack of capitals has stuck.
>
> Most IPA letters have capitals, actually.
It's easy to determine that that is not true. Of the 96 characters
in the IPA Extensions block only 29 have capitals, and even if we
allow for the fact that some characters in that block are obsolete
(some of which are among those having caps) or nonstandard, and
that _a-z æ ø ð β θ_ and perhaps a few more letters with caps are
also IPA that's far from a majority! Moreover there would be the
problem that at least b/β ð/ɖ and x/χ would have identical
capital shapes. That doesn't make a party IMHO!
/bpj
Messages in this topic (51)
________________________________________________________________________
________________________________________________________________________
10. Fw: the symmetry of sound change
Posted by: "Roger Mills" [email protected]
Date: Sat Jun 22, 2013 1:07 pm ((PDT))
this went to Padraic by mistake.....
----- Forwarded Message -----
From: Roger Mills <[email protected]>
To: Padraic Brown <[email protected]>
Sent: Saturday, June 22, 2013 2:29 PM
Subject: Re: the symmetry of sound change
I think, in distinctive feature terms, _s_ can be specified as [+Cons +Obs
+coronal +cont +strident -voi (redundantly -lateral)] and is distinguished from
_K_ (voiceless l IIRC) and _S_ in that K is [+Cons -Obst +cont +coronal
((maybe)) +lateral -voi] and _S_ is [+Cons +Obst -coronal* +cont +strident -voi
(redundantly -lat)] ((*or whatever features are used to distinguish
palatals)..... Thus it is a simple rule to change the voicing of _s_ without
changing the voicing ot the others.
If you did want K and S to voice as well, then the rule would have to specify
[+cont +strident -voi (plus whatever features would cover all three)]
--------------------------------------------
> From: George Corley <[email protected]>
>
> Hmm, it looks like you're right. Ah well. In any case
/s/ by itself is a
> natural class, so it's theoretically possible for it to
have a sound change
> that no other sound has. But it seems a little odd to
me that it would be
> affected and other sibilants wouldn't.
RM No, as I just showed, _s_ can be uniquely specified.
Messages in this topic (1)
________________________________________________________________________
________________________________________________________________________
11. History of some spelling pronunciations (Was: Equivalent to Grand Ma
Posted by: "Jyri Lehtinen" [email protected]
Date: Sat Jun 22, 2013 3:13 pm ((PDT))
2013/6/22 Eric Christopherson <[email protected]>
> On Jun 20, 2013, at 6:41 AM, Jyri Lehtinen <[email protected]>
> wrote:
>
> > One is whether the reflex of the weak grade of /t/ is /d/ as in the
> > standard language of loss as in the eastern dialect area and the other is
> > whether the standard language /ts/ gets replaced by /t:/ which is typical
> > to many western dialects. For commonly used words I tend to use
> exclusively
> > the dialectal variants unless I aim for a hyper corrected register to put
> > someone off. Uncommon words haven't implemented this spread of dialectal
> > features and get the reflexes of the standard language. There is however
> a
> > middle ground between the two extremes where I'm never certain which
> reflex
> > to choose and the choice depends sensitively from the register of the
> > discourse. These examples are somewhat complicated by the fact that both
> of
> > the standard language reflexes are in fact historic spelling
> > pronunciations, but that's beside the point. I'm pretty certain that you
> > can find similar partial sound changes from your own idiolects as well.
>
> I'm curious as to what the "target" (I'm not sure of the correct term) of
> these spelling pronunciations was; did Finnish itself have <d> and <ts>
> pronounced as /d/ and /ts/? -- or did those spellings only occur in
> surrounding Germanic languages, and it was from there that the spelling
> pronunciation caused the *Finnish* graphemes to be pronounced likewise?
>
> I know <d> is a grade variant of /t/; does <ts> occur in native words?
You asked for an explanation and you shall get one that hopefully isn't too
long. These two segments and their dialectal cognates happen to be among
the most discussed ones in Finnish dialectology.
Both of the segments have native origins. <d> and older <dh> were
originally used to write a voiced dental fricative which was the weak grade
of /t/. Swedish had lost its own dental fricatives earlier so for Swedish
speakers the most obvious pronunciation of <d> was [d] instead of [δ]. Such
a contact influenced development has also happened in some dialect pockets
on the west coast. The [d] pronunciation must have spread as a prestige
pronunciation from Swedish speakers and in general educated people knowing
other European languages. Modern reflexes of old /δ/ in traditional
dialects but also in much of the non traditional colloquial language are
mostly /r/ and loss with increasingly rare /l/ and only marginal /δ/.
There is actually a phonetic clue in modern pronunciation that the standard
language /d/ isn't a product of fully natural development. The
corresponding dental stop /t/ is clearly dental as according to all
evidence was the original voiced fricative. However, the modern
pronunciation of /d/ is distinctly alveolar (it shares its POA with the
modern /n l r/) with no apparent internal reason for it to be so.
The modern <ts> = /ts/ in the standard language originates from the digraph
<tz> in written Old Finnish where it was used to indicate /θ ~ θ:/. Yet
again you can explain the [ts] pronunciation with prestige influence by non
native speakers not familiar with the sound and distracted by its spelling
while nearly all dialects diverged away from the fricative pronunciation
into various different directions. The dialectal divergence left very
little native support for any "correct" pronunciation and gave room for
outside influences to step in.
The /ts/ segment rather bafflingly corresponds to a Proto Finnic dental
affricate. Dialectally the segment has a large variation of different
cognates, most common being /t:/ and /ht/, but also /ts/ is found in the
south west dialects. This dialectal /ts/ seems to go directly back to the
original affricate but both /t:/ and /ht/ are reflexes of /θ:/. This might
seem a strange development, but we actually do have data that gives firm
evidence for the existence of the intermediate /θ/. Firstly there are a
couple of dialects that have preserved the sound up to modern times.
Secondly there are place names that have Swedish phonetic (i.e.
non-spelling) substitutions for this sound that wouldn't make sense from an
affricate, such as Fi. Ähtävä ~ Se. Esse. Thirdly there are Finnish loans
in North Saami that preserve this sound, such as NSaa. Ruoŧŧa = /ruoθ:a/ ~
Fi. Ruotsi, "Sweden". And lastly there are written accounts from the Old
Finnish period that clearly describe the sound as a dental fricative.
Naturally both loaning and new internally coined words having /d/ and /ts/
solidify their status in Finnish phonology. It's worth a note that where
these segments don't have old native roots they are much less likely to get
substituted by dialectally influenced pronunciations.
-Jyri
Messages in this topic (1)
------------------------------------------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/conlang/
<*> Your email settings:
Digest Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/conlang/join
(Yahoo! ID required)
<*> To change settings via email:
[email protected]
[email protected]
<*> To unsubscribe from this group, send an email to:
[email protected]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
------------------------------------------------------------------------
