[915] in tlhIngan-Hol
Re: Re[4]: "movie"mey, etc. (was: RE: RESENT: Bounced Mail
daemon@ATHENA.MIT.EDU (daemon@ATHENA.MIT.EDU)
Fri May 21 05:56:05 1993
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Errors-To: tlhIngan-Hol-request@village.boston.ma.us
Reply-To: "Klingon Language List" <tlhIngan-Hol@village.boston.ma.us>
From: Ken_Beesley.PARC@xerox.com
To: "Klingon Language List" <tlhIngan-Hol@village.boston.ma.us>
Cc: tlhIngan-Hol@village.boston.ma.us
Date: Fri, 21 May 1993 01:47:26 PDT
In-Reply-To: "Mark_Nudelman@go:com:Xerox's message of Thu, 20 May 1993 10:12:00
Mark Nudelman:
>>However, as I thought about this issue, I
realized that to my mind, "I sleep him" (in English) doesn't
SOUND ungrammatical. At least, not like "I him hit".<<
>>The sentence doesn't
sound like bad grammar, but like some non-standard *meaning*
of "sleep".<<
>>There seems to be a fuzzy line here, and most,
if not all, speakers not linguistically trained would
probably not make a qualitative distinction between certain
utterances considered "ungrammatical" by linguists and
others which are "grammatical" but nonsensical.<<
Mark's points are perfectly valid. Judgments of grammaticality even by
natives, and especially by linguists themselves (who have typically lost all
their intuitions), are always somewhat suspect. But in the end, all one can do
is ask a native, and preferably several natives, if a given word or sentence is
acceptable or unacceptable. They are the only possible authorities. If you
can invent a story, however silly, in which a native will accept a suspect word
or sentence, then most formal linguists will call it grammatical. Outright
ungrammaticality is often described as causing a funny feeling in the pit of
your stomach. Not too scientific. But there is general agreement that rules
of some sort do exist--you can't say just anything.
Mark's comment about "I sleep him" feeling like a possible non-standard meaning
of sleep is especially astute. We can certainly imagine finding this sentence
in a text and trying to make sense of it. A linguist would probably describe
this as the coining of a new transitive verb "sleep" that happens to look like
the intransitive verb "sleep" and probably has a related meaning. Something
very like this has happened in the last decade or so with the Spanish verb
"desaparecer" (to disappear), which originally was intransitive as in English
(John disappeared. = Juan desaparecio/.). From a new slang usage, there is now
a new verb "desaparecer" that is transitive--you can now hear and see sentences
parallel to the following, even in formal Spanish journalism:
The junta disappeared 3000 people in the state.
Juan was disappeared in December of 1985.
This transitive "desaparecer" means, of course, something like "murder
illegally and secretly", usually by a government in a political civil war; it
is related to the original intransitive verb in the sense that the murderers
typically maintain that the victims have simply disappeared.
So in a computer lexicon, say for a machine-translation system, I would now
postulate two separate verbs, with separate codings, syntactic behavior and
translations.
desaparecer Vintr
desaparecer Vtr
Similar things might happen with English "sleep" and Klingon "Qong." Slang
often expands the language with such coinings. Poetry as well often explores
the edges of grammaticality, purposely breaking the rules and coining new
words. The coinings that catch on and become generally accepted are
occasional, non-productive and idiosyncratic in meaning. About all you can do
is look out for them and add them to your lexicon on a case-by-case basis.
>>I knew when I wrote that ("it doesn't matter much ...")
that any linguist would object strongly. <<
Your points are valid and you won't get too much argument from a linguist,
certainly not any angry argument. My motivation, however, is to write a
morphological analysis of Klingon that neither underrecognizes nor
overrecognizes, as complete and as tight as possible. Some fuzzy areas may
remain. The hypotheses I state and the questions I ask have concrete
implications for my program. In that sense at least, they do matter. If, as I
have postulated, there is a transitive/intransitive distinction in Klingon, and
if forms like ?vIvem and ?vIpub are identified by Okrand as being ill-formed or
ungrammatical, then I want the analyzer to reject them automatically. If not,
then the program has to allow them. These are big IFs, but with Okrand
answering questions they could easily be tested in the ways I have described.
Ken Beesley
