Re: Azurian phonology

> Ämne: Re: [CONLANG] Azurian phonology
> Från: Lars Finsen <lars.finsen@...>
> Datum: Sun, 19 Oct 2008 21:11:24 +0200
> Till: CONLANG@listserv.brown.edu
>
> Den 18. okt. 2008 kl. 19.37 skreiv Benct
> Philip Jonsson:
>
>> Both Icelandic and Faroese have an aspiration
>> contrast rather than a voicing contrast, as does
>> Danish. It is disputable whether there was a
>> shift from a true voiced--voiceless contrast
>> since even in Swedish and Norwegian accents
>> rather have a fortis vs. lenis stop opposition
>> with lenis stops realizad as fully voiced stops
>> only between two vowels or a vowel and a nasal
>> or liquid.
>
> I'm trying to work my way through this. I assume
> you are using 'accent' as a synonym for dialect,
> aren't you?
Yes, as is customary in English.  I also wanted
to imply that what I said applies not only to
traditional dialects but also to the different
regional pronunciations of the 'standard'
language (which after all is only standardized
in grammar and vocabulary and in that it censures
some more or less wideopread traditional dialect
features such as [4`] ('thick ell') and [3\]).

> I think there is something like what you are
> describing in the west and north of Norway. What
> Swedish dialects are you thinking of?
The dialects you mention as well as Swedish dialects
on roughly the same latitude -- in general isoglosses
in Scandinavia run east-west at more or less straight
angles with th national border! -- differ from other
dialects chiefly in that *phrase internal, word-initial*
lenis stops are also voiceless, but I daresay that there is
no Swedish or Norwegian dialect where post-pausal and
utterance-initial lenes are fully voiced; mind you
voiced--voiceless is not in phone_t_ic an absolute opposition
but an infinitely variable scale, with the opposition rather
is in terms of relative distance between fortes and lenes
in similar phonetic environments.

Phoneticians speak of
Voice Onset Time (VOT) with fully voiced at one end of the scale
and heavdly aspirated at the other.  For southern mainland
Scandinavian dialects you may say that in any given position
fortes are about 60% to 75% percent further away from fully
voiced, but the absolute position for both slides around
about 30%, for both in parallel. Intervocalically
lenes are 100% voiced and fortes are at about the 30%-40% mark
of the scale, while post-pausally they both slide down around
30% towards the aspirated end.  The dialects you mention
are marked out by sliding even more; the lenes maybe as far
as the 50% mark and the fortes to the 0% mark.
Interestingly mainland dialects differ in how they treat
pre-pausal fortes in particular: some have them at roughly
the same value as their post-pausal counterparts while others
have them very strongly post-aspirated (this is true of my speech)
while still others have them preaspirated -- unlike the insular
languages regardless of gemination.

BTW what I say here of stops in mainland Scandinavian dialects is
mainly true of English too, although in English prepausal
stops tend to be unexploded, so that there the fortis-lenis
distinction is rather one of voice **offset** time!  British
dialects tend to a relatively more fortis pronunciation and
American ones to a relatively more lenis pronunciation.  In the
latter the difference between intervocalic fortis--lenis stops, while
still fully discernible to native speakers, may be so small
that it appears lost to speakers of European Germanic languages,
including to speakers of European English -- and in the case
of /t/--/d/ it of course often is lost!

The insular
languages and Danish have their lenes roughly at the 50%
mark and their fortes up around 90% in all positions.
Characteristically many of these latter varieties -- notably
non-Northern Icelandic and most Danish -- lose the fortis--lenis
distinction for non-geminates in intervocalic position,
pronouncing the former fortes and the few intervocalic
lenes they had alike as voiceless lenes at the 50% mark.

NB that postvocalic single lenis stops were rare to begin with
since Old Scandinavian had voiced fricatives in that
position -- which usually simply were lost in Southern
mainland dialects and later restored from writing in formal Swedish. You have to go to
loanwords like _radar_
to find any, and that word might as well or rather be spelled
_ratar_ for most Icelanders.  Danish of course went as far
as turning *all* postvocalic coronal and velar single obstruents
into voiced fricatives and then on to approximants -- not a
few dialects did this to labials as well -- and *geminate*
fortis stops into single lenis stops, so that a Danish
_bakke_ sounds like a Swedish/Norwegian _bagge_ and an
Icelandic _baggi_ more like a Swe/No _ba[ck]ke_!


>> In the same position fortis stops are voiceless
>> unaspirated. Elsewhere fortes are voiceless
>> aspirated and lenes are voiceless aspirated,
>> though usually still more weakly articulated
>> than intervocalic fortes.
>
> So, elsewhere, the only difference between fortes
> and lenes is that the latter usually still are
> more weakly articulated than intervocalic fortes?
Yes.

>> One interesting discrepancy is that while
>> intervocalic lenis geminates are voiced, fortis
                                          ^
>> geminates in the same position -- and of course
>> after /s/ -- are usually aspirated!
>
> I think I can make sense out of this if there is a
> comma after 'voiced'.
Yes, now duly inserted.

>> Even preaspiration is found in that in some
>> accents any -- not only geminate -- postvocalic
>> aspirated stop can be realized as weakly
>> preaspirated.
>
> I think I get this too, if in that is synonymous
> to seeing that or similar.
Yes.

> Sorry if I sound rude. But I don't feel like a
> linguistic insider, and probably I never will.
> Decoding linguistic inside talk, especially in
> English, can be hard work to me. I don't have such
> a head for this stuff as some others here. But
> with a little patience, and some hard work, I
> think I can get by.
Well, just tell me if I get arcane!

>> These alternations are fully automatic, operate
>> across word boundaries and are usually
>> unconscious to speakers, but there all the same.
>
> No doubt there's a lot about my language that
> I'm unconscious of as well, though I've been
> more analytic about it than most. I really
> should study my own language better, especially
> now that I'm working with a conlang that's
> closely related to it. Next on my plough-through
> list is Gustav Indrebø's Norsk Målsoga,
> which I recently bought.
>
>> Of course most accents of English have a similar
>> fortis--lenis system.
>
> Interesting. Must learn a bit more on that as
> well.
>
>> OTOH Scots Gaelic has a system entirely
>> analogous to that of Icelandic and Faroese.
>
> That was an important insight. Possibly one of
> either Gaelic or Norse has caused a substrate
> effect on the other, or there is a common
> substrate. Pictish? Or Old Albic? Or Urianian? As
> Gaelic has only had a limited foothold in Uriania
> historically, I am wondering if I should limit the
> Azurian aspirated-unaspirated contrast system to
> those areas.
My pet theory is that voiced stops didn't transmit
very well through the storms of the Norwegian Sea,
so that at a not so long distance they sounded voiceless,
and people simply gave up bothering to voice them! ;-)

More seriously we know that in any slavery society
the laves outnumber the masters, and on Iceland in
viking times most slaves came from Ireland and
Scotland...

/BP

Replies