Re: Conlang Christmas

Henrik Theiling skrev:
 > Hi!
 >
 > Benct Philip Jonsson writes:
 >> On 29.12.2007 Henrik Theiling wrote:
 >>>    Þélk Nátli eð bon nó önn!
 >>>
 >>> This is pronounced as follows:
 >>> :  [TjEKk 'naU)htlI E: bo@):n noU): 9n]
 >> I can't remember ever having seen [o@)] for /o:/
 >> described for Icelandic. The closest thing I can remember
 >> is [o@)] for _á_ in Faroese. Is this a deliberate
 >> difference between Þrjótrunn and Icelandic
 >> pronunciation, or something you've actually heard
 >> colloquially? An effect of the following nasal perhaps?
 >
 > No, no sandhi, it sounds like that in most (all?)
 > positions to me. My main source would be Magnús
 > Péturson's recordings which I have accompanying a course
 > book of Icelandic. This is the clearest recording of
 > Icelandic I have listened to, apart from several movies
 > with somewhat more real-life Icelandic. The Icelandic /o:/
 > sounds very different from German /o:/ to me, the later I
 > would indeed transscribe as [o:]. This may be L1 bias, of
 > course. But German /o:/ does sound like a very pure
 > version of a non-changing, long vowel to me, while the
 > Icelandic version sounds like becoming laxer towards its
 > end. The increasing laxness I tried to express by using
 > [o@)]. The [@] part is brief; the [o] is strongly
 > prevailing.
 >
 > What transscriptions have you encountered?
 >
 > **Henrik

I have only ever seen [o:] which probably stands for [o_o:].
It is not unlikely that there is a laxing diphthongization
in Icelandic secondary long vowels, as such a
diphthongization is widely found in Mainland Scandinavia and
in Faroese. It may indeed be L1 bias which made me not
noticing it when I was in Iceland, since I wasn't aware of
long vowel diphthongization in my L1 at the time. I have
opening/laxing diphthongization of (secondary) long vowels
*in closed syllables only* as follows:

  - /i:/ -> [z\=:e)] (or [z=:e)], since I'm bidialectal).
  - /y:/ -> [z\_O=:2)] or [z_O=:2)]
  - /e:/ -> [e:E)]
  - /2:/ -> [2:3\)]
  - /E:/ -> [E:&)] or [E:e)]
  - /a:/ -> [a:&] or [&:E)]
  - /3\:/ -> [3\:6)]
  - /A:/ -> [Q:6)]
  - /o:/ -> [o_o:3\)]
  - /u:/ -> [u:U\)] or [u:8)] or even [uB_o8)], still somehow
    only one syllable!
  - /u\:/ -> [8_+_w:3\)] or even [8_+_wB_o3\)], also still
    one syllable.

NB that those items I've transcribed with a front second
    element may still have some centralization which escapes
    my naked ear, being [z\=:@\)], [z\_O=:8)], [e:3)], [E:3)]
    etc. However _lät_ 'sounded' is clearly *not*
    **[lE:6)t_h]. Most of the variants are part of my
    bidialectalism, so that for the most part one variant
    belongs to one dialect and the other variant to the other
    dialect, but not so with /E:/ -> [E:&)] or [E:e)]: _här_
    'here' and _där_ 'there' are always [hE:e)r\`] and
    [dE:e)r\`] while other words mostly have [E:&)] but
    sometimes [E:e)].

As allophonic variants these strike me as somewhat odd,
since diphthongization occurs in *closed* syllables but not
in open syllables: _vit_ 'white, sg.' is [v\z\=:et_h] but
_vita_ 'white, pl.' is ['v\z\=:tA] without any
diphthongization. I'd expect vowels in open syllables to be
longer and hence more prone to diphthongization. Maybe Dirk
can elucidate this?

[gA:d] how ugly narrow transcription with CXS is!


/BP 8^)>
--
Benct Philip Jonsson -- melroch atte melroch dotte se
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   "C'est en vain que nos Josués littéraires crient
   à la langue de s'arrêter; les langues ni le soleil
   ne s'arrêtent plus. Le jour où elles se *fixent*,
   c'est qu'elles meurent."           (Victor Hugo)

Reply