Hailaz Llama.

> > Right. Essentially, if the ø (sometimes called ø2) resulting from
> > both i- AND u-umlaut of a was pronounced differently than the ø
> > resulting from i-umlaut of o, then ON had 10 vowels, not 9 (which
> > includes the short, long, and long nasalized vesions) ;)
> >
> > Konrad
> Or maybe 13, if you count the more open or lax pronunciation of the
> vowels later spelt <u> and <i> (earlier <o> and <e>) in unstressed
> possitions?! But these two only short + or - nasal.
> Llama Nom

;) 10, I think, only ø2 was quite rare. There are few words that had
it, as far as I can tell, as a stem-vowel, and I have not found any
example of long ø2. Gordon (and others) distinguish ø2 from ø1, and I
think for good reason, because it could hardly have been pronounced
the same as ø1 (think a both i-mutated and u-mutated vs. o i-mutated
only - we get something like short æ with ø-coloring ;). Still, one
can easily see why this rare vowel was slighted for extinction with
the first sign of vowel-collision/absorption in the language. About
the lax-pronounced u/o and i/e in unstressed positions. All that we
can reasonably do today is reckon stem-vowels. We have a good grasp
of the of how the PN stem-vowels became what they are in ON and how
they got there, but the vowels of inflectional endings are a wholey
different matter. OIce Homily-Book usually shows vowel-harmony, but
inconsistently; the contemporary ONor Homily-Book shows it only very
inconsistently (different hands wrote and copied, probably in both
cases, leaving us in the mire about the geographic specifics of the
vowel-harmony phenomenon in West Norse areas). East Norse evidence
hardly helps here, because we know that by this time that there were
different dynamics at work in this branch. So the questions remain:
what happened to the vowels in unstressed syllables in West Norse?
When did vowel-harmony develop? How come? In what areas did it exist
within the larger West Norse area? Was it universal? Were there some
local variations? In Icelandic, it disappeared. Is this a sign that
it came and went? Or only that it went? Some portions of the ONor
Homily-Book do not show it at all. Is this a sign that vowel-harmony
never developed in some West Norse regions? We can never know with
absolute certainty. The runic alphabet used only i and u here. While
we can reckon the stem-vowels from Latin-letter sources, research on
PN, the modern languages, etc., we hit a bump in the road when it
comes to the i/e and u/o vowels of inflectional endings. Given that
this is probably the only vowel-related problem that we do not fully
understand at this point (and probably never will ;), the simple
solution is to assume that there was no vowel harmony at all during
the 10th century, at least some areas were West Norse (otherwise the
same) was spoken. This makes things simple. We get i and u in the
vowels of inflectional endings, just like in runic inscriptions, just
like in MIce and just like in MFaro. A written convention that makes
sense in an area that can be i little cloudy. Just think, there are
thousands of examples of later poets rhyming i with e, é with e, etc
- all by folk who probably didn't speak this way in real life ;) Here
are a few by Hallgrimur Pétursson (which I promised that I would read
entire this winter ;):

Mig skyldi og lysta adh minnast thess
mínum drottni til thakklætis

Innra mig loksins angridh sker
æ, hvadh er lítil rækt í mér

Hvar fær thú gløggvar, sál mín, sédh
sanna gudhs ástar hjartagedh

uppteiknadh, sungidh, sagt og tédh.
Sídhan thess adhrir njóti medh.

or even é with i:

Bænarlaus aldrei byrjudh sé
burtfør af thínu heimili

and we have at least 3 more examples of é with e, one of o ith ó,
etc. - all from only the 1st psalm (27 verses) of 50 psalms! So, one
wonders, on the bases of later attitudes to these unstresed vowels,
if ancient vowel-harmony distiction were really that important in the
first place. I would bet that if you put 2 ON speakers in the same
room, one with total vowel-harmony and the other with none, that they
would have no problems speaking the 'same' language together ;) So I
vote for a simple o (empty vowel-harmony set, only i and u) - its
runic, its modern, its simple, its clean ;) So anyway - 10 vowels, I
think, the 10th very rare and short, the other 9 in both long and
short, nasal and non-nasal, versions....unless there is any evidence
to be had to the contrary ;)