Heill Keth,

> The medieval spelling of imported arabic words, (you mentioned
admiral) is of course common to all European languages, and thus does
not give a good demonstration of the peculiarities of English.

I think you misunderstood; I was giving an example of how the origin
of a word could get distorted by spelling change.

Anyway, I'd rather not discuss English spelling on this list, so I
won't dwell on that subject here. Let's not continue that discussion.

> For comparison, I'd like to state that the dictionaries we used for
Norwegian in school, did not include any etymologies at all. Nor did
they include any phonetic pronounciation guidelines. That is because
Norwegian is very much pronounced "as is".

I'm not Norwegian, but even if I were, I think I'd be careful with
such statements; first, which of the two official Norwegian
orthographies (or languages, so-called) do you refer to, Bokmål or
Nynorsk? Second, with the myriad Nowegian dialects, how can one make
a conclusive statement about Norwegian pronunciation?

Most languages have a variety of dialects, which often causes
problems in orhography. Spanish orthography, for example, fits to all
the dialects, yet to none of them; that is, in most versions of
Spanish, there are some two symbols that represent the same sound (or
rather, phoneme), yet in another version, the same two symbols
represent different phonemes. For example, in European Spanish, "se"
and "ce" are generally pronounced differently ("ce" sounding like
ON "þe" [Te]), while in American Spanish, they're pronounced the same
(both as "se"); meanwhile, in a typical European Spanish
dialect, "eya" and "ella" would sound the same (as "eya" [ejA]),
while an American Spanish dialect (such as, say, Ecuadorian) would
have two different sounds there, pronouncing "ella" as [eZA] or even
[edZA] ([dZ] being the "j" in "just"). The point is that in the case
of Spanish, and many other languages, the orthography serves all the
dialects equally, keeping all distinctions between sounds that are
made in the various dialects.

What I'm trying to say is that the quality of an orthographic system
is not just how well it transcribes "the spoken language" (which
spoken language?). Accurate transcriptions ultimately fall prey to
the great variety of the spoken language: between generations,
between dialects, sociolects, even idiolects (e.g. if we were to
transcribe English accurately, which "either" should it be?).

> But if the Icelandic school childeren all need
> to know SAMPA and IPA in order to learn to pronounce their own
> language, then that would of course modify ones view on how close
> written Icelandic is to an approximately phonetic spelling system.

How did you come up with that? Why not present it in question form

"Do Icelandic school children need to know SAMPA and IPA in order to
learn to pronounce their own language?"

Would at best have been a very silly question, but as a serious
statement, it baffles me.

You see, children learn to speak long before they learn to read or
write; they still do so, will do so, and have always done so. Even
outside of Norway, they do.

As regards SAMPA and IPA, then students of some high schools learn
rudimentary IPA at age 17, for one semester; nobody has heard of
SAMPA, though.

Now, I'm not particularly sensitive about criticism of Icelandic
orthography, or of anything Icelandic; erraneous conclusions are what
I am sensitive about, however. I wouldn't involve myself in this
discussion, except that I have a very hard time containing myself in
the face of such error.