From cbmvax!uunet!CUVMA.BITNET!LOJBAN Sat Mar 14 10:04:01 1992 Return-Path: Received: by snark.thyrsus.com (/\==/\ Smail3.1.21.1 #21.19) id ; Sat, 14 Mar 92 10:03 EST Received: by cbmvax.cbm.commodore.com (5.57/UUCP-Project/Commodore 2/8/91) id AA14300; Sat, 14 Mar 92 09:58:54 EST Received: from pucc.Princeton.EDU by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA09223; Sat, 14 Mar 92 09:34:19 -0500 Message-Id: <9203141434.AA09223@relay1.UU.NET> Received: from PUCC.PRINCETON.EDU by pucc.Princeton.EDU (IBM VM SMTP V2R2) with BSMTP id 6308; Sat, 14 Mar 92 09:18:47 EST Received: by PUCC (Mailer R2.08 PTF011) id 2766; Sat, 14 Mar 92 09:18:21 EST Date: Sat, 14 Mar 1992 11:02:46 GMT Reply-To: Ivan A Derzhanski Sender: Lojban list From: Ivan A Derzhanski Subject: Place names To: John Cowan , Eric Raymond , Eric Tiedemann In-Reply-To: Logical Language Group's message of Fri, 13 Mar 1992 19:58:24 -0500 <3085.9203140200@cogsci.ed.ac.uk> Status: RO > Date: Fri, 13 Mar 1992 19:58:24 -0500 > From: Logical Language Group > > Now I'm hoping the proposals for latitude/longitude place names aren't > serious. If so, this effort is getting out of hand in a totally > negative way. Have no fear - when I made this suggestion, I was very well aware that it was not acceptable for anyone. But I really wish it were. > The 'obvious' answer is to use the country and city names that the people > themselves use WHEN TALKING TO FOREIGNERS. There is NO SUCH ANIMAL as _the_ name used when talking to foreigners. When a Chinese is speaking English, he calls his country {tcaina}. When he is speaking Russian, he calls her {kiTAI}. When he is speaking Mandarin, to a foreigner or a countryman alike, he calls her {tconkuo}. Names are determined by the language, not by the setting. > (I have no > problem with giving multiple names for Switzerland reflecting the multiple > names they themselves use in official capacities.) Sure. The existence of multiple names for a country with more than one major language is quite natural. > We also have to deal with the problem of how names look on paper as well as > in sound. If exactly matching the sound AS pronounced by the natives gives > a textual name that no one will recognize without reading aloud, <...> I addressed this point in another article. Since these considerations make for different treatment of languages using a Roman-based script, I insist that we must forget how a name is written before lojbanising it. > It might be nice to have some absolutely rigid way to encode each natural > language's phonology into Lojban, but we don't - Then we're bound to be inconsistent on many occasions. Nick still hasn't answered my question why he mapped _Oesterreich_ to {.e-} but _Tu"rkiye_ to {-u-}, given that the two vowels differ only in height. > Preservation of the written form is desirable, and can resolve such > problems. Yes. That's what {la'o} is for. > For country and city names, this > means that you can't go making a name into something the local speakers who > would use the name are going to have problems with. Thus when we tried > to put Nihon/Nippon for Japan into a gismu, we quickly decided that > nixno or nipno would not be acceptable, since Japanese speakers would not be > abale to pronounce the words without heavily buffering them. We chose to > avoid the 'x' entirely and made the gismu "ponjo", with the 'J' from Japan > for those people who want the hook, while providing a consonant cluster > that the japanese speaker may not need to buffer. Excuse me, but this makes no sense whatsoever. Cultural gismu are words of Lojban. No Japanese needs to be able to pronounce any Lojban word, including {ponjo}, unless he is a lojbo se bangu, in which case he will have to learn to pronounce all Lojban consonant clusters anyway. Did you also check whether dogs and cats are comfortable with the words {gerku} and {mlatu}, respectively? I would also very much like to learn how exactly you carried out your research which showed that Japanese speakers would find it easier to learn to pronounce the consonant {j} than syllable-final {p}. > For Hungary, the official name is "Magyar". We are respecting the people > of that country by calling it la magiar. You are insulting the people of that country by claiming that her official name is "Magyar", whereas in reality it is "Magyarorsza1g" ("a1" is a-acute), pronounced {m@#@rorsa:g}, where {@} can be equally fairly rendered as {a} or {o}, {#} as {gi}, {di} or {dj}, {:} denotes length, and the stress is on the first syllable. > Albania has that 'q' that has no Lojban counterpart. It could become any of {ki}, {ti} or {tc}. > But if you look at > the name of the country on a postage stamp (where it is all caps), it is > easy to not notice the cross on the 'q', Certainly. Fools always have an easier time. > giving what seems to me the obvious la shopeiras. Surely you mean "snopeiras". Since Lojban has no "h", I expect you to substitute "n" for it. It looks pretty much the same, doesn't it? Now I'm hoping this proposal isn't serious. If so, this effort is getting out of hand in a totally negative way. > This isn't going to be how the natives pronounce it, > but they would probably recognize it if a Lojbanist said it without too much > difficulty, as being an attempt to respect their self-identity, as opposed > to la .albeinias. I doubt that they would recognise {snopeiras} as _Shqipe"ri(a)_, but I can't begin to doubt that they will have a good laugh at this idiotic "attempt to respect their self-identity". > la moskvas*. is technically illegal since it has an impermissible medial. So these restrictions apply to cmene as well? This does it. As of now, I won't be caught using anything but {la'o rus. Moskva rus.}, and I don't care what anyone else uses. > Lojban assumes that voiced/unvoiced combinations will degenerate to either > voiced or unvoiced. Which is a wrong assumption. > <...> the Russian capital should be changed, and > reflect thepreference most likely to be acceptable to the Russian speakers. > If we don't know that preference, the I would say to go with la moskfas since > that preserves the highly recognized first syllable in print unchanged. > But I could live with la mosgvas. if Ivan thinks it would be preferred, > or any other minor change. Is {sg} permissible? I'd say {myskFA} is much less than tolerable, but {myzgVA} is even worse. You may try {myskUA} if you wish. Russian as spoken in Moscow does NOT tolerate {o} in unstressed syllables. > The important thing to me is to <...> make the pronunciations reasonable > for botht the native speakers and for Lojbanists in general, while optimally > preserving some visual recognition for those people who don't speak the > language as much as they write it. Because, of course, all new Lojbanists are already familiar with the way names look in English. So much about cultural neutrality. > (A good point to remember is that the > peoplee of the country will be most likely to speak the name. <...>) Not really. First, they will have no use for the lojbanised name. I would always prefer to pronounce a name from my country in the natural way, in quotes, rather than make it into a cmene, which in a huge number of cases will mean distorting it beyond recognition. Second, one can always call his own country `my country', and get away with it. co'omi'e la'o bylg. Ivan bylg.