Re: Lojbanizing umlaut

[Logical Language Group <lojbab>]

la lojbab. cusku di'e

>> Gag.  Goethe as Geute [goit@]?  Unspeakable.
> 
> I don't see why it is more unspeakable than [got@] or [get@]. To my untrained
> ear, o and u umlaut sound close to alike, though, so I am no judge.

I think the best form is "getys.", supplying the usual consonant and unrounding
the oe to e.

>> > For people where this isn't possible (dead ones, etc.)
>> > and for places of uncertain pronunciation,
>> 
>> What "places of uncertain pronunciation"?  There are places whose names
>> have more than one pronunciation, certainly (New Orleans, e.g.), and there
>> are names of whose pronunciation some Lojbanist may not be certain (I only
>> found out the other day how to pronounce "Otranto", the name of a city in
>> Italy), but these are two separate issues.
> 
> For an uninformed Chinese Lojbanist, Otranto and New Orleans are the same
> problem, except that we will have one (at least) in the dictionary for
> the latter.

I picked N.O. specifically because two pronunciations were given in the
latest list.

> But when we turn to pronunciation of dead languages, we REALLY are uncertain.
> Moreover, we are uncertain what standard to use (do we use Vulgar Latin or
> book Latin, and from what era, to determine the pronunciation of various
> Roman provinces)?

Yes, that is a third case, when nobody knows what the right answer is.

> I don't intend to require every Lojbanist to become a master of the world's
> phonologies, past and present.  So when in doubt relying on spelling is not
> a bad idea for dealing with a name you don't know.  And recognizing that the
> average Lojbanist will do that, the knowledgeable Lojbanist who is devising
> a Lojban name based on multiple 'legitimate' pronunciations, probably should
> choose the one that will match most closely with what the unknowledgeable
> Lojbanist will choose - after all, they might end up talking to each other %^)

This is related to And's point.  A name means what the speaker wants it to mean.
But equally, as Mark says, you can't just make these things up and be
understood.  Several people assumed that Pierre, N.D. was "pi,er.", but it's
"pir."

>> >  and usage or input
>> > from a local native speaker will decide which one sticks for the long ter
>> 
>> The latter, we hope, not the former.  Yet native speakers aren't
>> infallible guides to their own phonology, either, witness the various
>> Chinese-speakers on sci.lang who insist that English /b/ = Chinese /b/,
>> despite the fact that the former is usually voiced and the latter usually
>> isn't.
> 
> Fine, but who am >I< to argue with that native speaker.  If he likes the sound
> of his name transliterated with 'b' instead of 'p', then I would say that he
> is correct.  When Ken Shan the other day tried to Lojbanize his native
> Taipei and did not know whether to use 'b' or 'p', then I have to say: use 'p'
> if they sound more-or-less equal to you, since that is the spelling, otherwise,
> he as a native speaker should feel free to choose what sounds best -EVEN IF
> what he chooses may be less close to the linguist-evaluated sound..

"Taipei" is not "the spelling" of Taipei.  It's the Wade-Giles romanization of
it, which has no official standing anywhere (as far as I know).  "Taibei"
would be the Pinyin, and in case of unification (on whatever political terms)
would undoubtedly become the standard spelling.

Since we map the sound written in Pinyin with "b" and in WG with "p" 
elsewhere consistently to Lojban "b", we should do so here also:
"taibeis.".

>> Except that by that time it may be too late, and we will be stuck with
>> monstrosities like Eng. "Leghorn" = It. "Livorno".
>> 
> 
> But we won't. at least for languages with written Romanizations.  The worst
> probalme we will have is likely to be with native English, when we try to
> Lojbanize "Worchestershire" %^)

Too right.

> Would we be offensively wrong to pronounce "Otranto" as "la otranton."?

I don't know about "offensive".  It's wrong, anyway: "la Otranton." is more
like it, mapping both Italian "o" vowels ([o] and [O]) onto Lojban "o".
The stress is the important point here.

> Or is it as bad as Leghorn?

No.

What I really want to avoid is rendering "Ougadougou" (a city in Africa)
as "la .o'ugado'ugo'us." rather than "la ugadugus." through failure to
understand an elementary French spelling rule:  "ou" = [u], "u" = [y].
Consider the mish-mosh that JCB made out of "Khrushchev".

-- 
John Cowan		sharing account <lojbab@access.digex.net> for now
		e'osai ko sarji la lojban.