Re: THEORY: Tonogenesis (?) from PIE (Was: The rebirth of m"/21aw as mql21aw)

> Date:         Fri, 5 Nov 1999 18:15:30 +0000
> From: Paul Bennett <Paul.Bennett@...>
>=20
> tal>>>>>>
> Crash-course, Norwegian tonemes (pardon my stunt-translations):
>=20
> [snip]
>=20
> SAMPA:
>   \  -
> <b=3DF8nder> /'b2n@r/ "peasants"
>  /\  /
> <b=3DF8nner> /"b2n@r/ "beans"
>=20
> <<<<<<
>=20
> Are there any defined orthgraphic features (intervocalic cluster of
> different/same consonants?) that makes this predictable?
The usual theory about this phenomenon is that it arose back in the
late 800's or so, when Old Norse was having a heavy attack of
syncopation. A lot of bisyllabic stems were reduced to monosyllables,
but retained a 'bisyllabic intonation'. (I haven't seen any attempts
at reconstructing exactly what the difference was, however).

The point is that since then, the distribution of the two word accents
has changed a lot. And the rest of the phonetics has changed as well,
independently. So there's not much correlation between the current
form of a word and what word accent it gets.

There are localized rules in some corners of the lexicon, though ---
like that surnames in -sen always have the accent that's called
tonelag ett in Norwegian, and st=F8d in Danish.

Lars Mathiesen (U of Copenhagen CS Dep) <thorinn@...> (Humour NOT marke=
d)