On Fri, 5 Jun 2009 13:10:31 +0200, Benct Philip Jonsson <[log in to unmask]> wrote:

>Andrew Jarrette skrev:
>> On Thu, 4 Jun 2009 16:41:20 +0100, R A Brown <[log in to unmask]>
>>> Mark J. Reed wrote:
>>>> But whence the English name "wye"?
>>> It must derive from Old English [y:], since
>>> the letter was used that way. As late Latin
>>> was also psilotic (i.e. they dropped their
>>> aitches) that would've been the contemporary
>>> Latin form also.
>>> Of course, Old English /y(:)/ became /i(:)/ in
>>> Middle English, but that would've meant that I
>>> and Y had identical names! Presumably when the
>>> vowel was unrounded, the letter name retained
>>> initial lip-rounding, i.e. became [wi:], hence
>>> modern English [waj].
>>> It did occur to me that there might have been
>>> dialect forms with initial non-phonemic glide
>>> in Old English, i.e. [Hy:], which would
>>> presumably have given Middle English [wi:].
>> I find it interesting that the Icelandic name
>> for <y> is, in their writing, 'uj', which they
>> pronounce [y:] (the only occurrence of this
>> phoneme in their language, otherwise <u> is
>> pronounced [Y]; <j> is like English consonantal
>> <y> in Icelandic normally). To me it looks like
>> if one interchanges the position of the
>> semivowel in the sequence 'uj', i.e. change
>> <u> (if taken in its original Latin
>>     pronunciation) to its corresponding
>>     semivowel <w>, and change <j> (if taken in
>>     its German/Scandinavian/Slavic/Hungarian
>>     etc. pronunciation) to its corresponding
>>     vowel <i> (Latin pronunciation), one arrives
>>     at 'wi', which looks like it could be the
>>     forerrunner of our 'wye'.
>> Not that any of this is important, relevant, or
>> informative, just wanted to mention the
>> similarity.
>> Andrew Jarrette
>Rather [y:] is the name of _-�_. 

Oh, woops.  I was writing from memory, from something read about 30 years
ago as a child in the library.  I probably should have been more circumspect
in my statement, since I clearly wasn't sure of what I was saying.  Plus I
was assuming naively that there were no Icelanders on the list who could
prove me wrong.  Maybe no Icelanders yet, but a Swede who knows his Icelandic.

The name name of
>_y_ is _yps�lon_ ['YfsilOn] or ['IfsilOn]. This
>would be the only word where _y_ not followed by
>/l/ may be [Y], unlike _pylsa_ and similar words
>where ['p_hYlsa] is fairly common. Evidently the
>unrounding of Old Icelandic /Y/ failed to apply
>before /l/, these cases merging with /Y/ < /U/

Interesting, didn't know that about Icelandic.

, but ['YfsilOn] is an outright archaism,
>probably because [Y] is how Icelanders
>traditionally read the Greek letter and its name.
>Actually there is arguably a diphthong /Yi)/ of
>limited occurrence which may be realized as
>[y:]. It typically occurs in _hugi_ /hYi):jI/
>[hy:ji] 'mind DAT.SG', usually cited in
>descriptions of Icelandc phonetics. Look at p. 9
>of Stef�n Einarsson's course if you have it! He
>writes the diphthong [(\Y)\y:] which is probably
>phonetically exact. He says that words _-ugi_
>and _-ogi_ /Oi):jI/ only rime with similarly
>spelled words, but words in _-�gi_ and _-augi_
>rime with each other /2i):jI/. Stef�n suggests
>that these diphthongs may be transcribed
>'phonetically' as _u�, o�, ��_ (p. 11), probably
>revealing native-speaker intuition rather than
>phoneticians' pedantry.

I appreciate the information, but your phonetic writing is a jumbled
confusion to me.  What are <Yi):> and <[(\Y)\y:] and <Oi):> supposed to
represent?  I don't understand all the ): and \, etc.

>Incidentally Finnish /y:/ is rather much on the
>[u\:] or [Y:] side, and to a Swedish speaker
>sounds like our /8:/ (The rather unwieldy CXS for
>this sound is [8_+_O:B_o], for which reason
>Andreas and I have come up with [8\] == [8_+_O]).

What about just <8>, as you wrote in "/8:/"?  When I checked CXS, that's
what I found for the close-mid central rounded vowel.  But is that the
phoneme you meant?  I thought Swedish "long u" was rather a high central
rounded vowel, CXS <u\>?

>The trick to teach a Finnish speaker to
>distinguish Swedish /y:/ from Swedish /8:/ is to
>tell him or her that Finnish _hyi_ 'ouch' sounds
>like Swedish _hy_ 'complexion' (which is _iho_ in
>Finnish), but means the same as Swedish _hu_, and
>the mnemonic _Hu, jag har d�lig hy_ = _Hyi,
>minulla on huonoa ihoa_ 'Ouch, I have a bad
>complexion'. ;-)

I once read an essay in a linguistic journal that talked about southern
Swedish dialects, and I remember it used the graphemes <yj> and <yw> when
talking about some of the Swedish vowels.  I wonder, could the first be an
attempt at your "long y", and the second an attempt at your "long u"?  If
so, are these representations at all accurate in their description of those
sounds? (I don't speak Swedish and have only heard it spoken maybe once in
1988, apart from checking out Lisa Ekdahl's songs on the internet (love "Vem
vet" - I notice she pronounces "vet" as though it were written "viet", or
"vee-et" in English spelling.  I also noticed this with the Swedes I met in
Germany in 1988.  I found it curious because the Swedish phrasebook I had
said to pronounce this vowel as though it were spelt "vate", i.e. IPA
[ve:t].  But I have never heard a Swede pronounce it this way.  And is "long
ä" really [E:] or is it closer to [e:]?  A commercial on Canadian TV that
features a Swedish or mock-Swedish tour guide has him pronouncing "Lake
Mälaren" as though the "ä" were pronounced [e:], as it seemed to me.)  By
the way I also own "Lisa Ekdahl sings Salvadore Poe", which is an excellent
CD especially for someone like me whose parents had lots of Bossa Nova
records around the house.

Andrew Jarrette