[532] in tlhIngan-Hol
De-verbal nouns
daemon@ATHENA.MIT.EDU (daemon@ATHENA.MIT.EDU)
Fri Apr 9 20:38:33 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: Ken_Beesley.PARC@xerox.com
Date: Fri, 9 Apr 1993 15:45:30 PDT
In-Reply-To: "krankor@codex.prds.cdx.mot:com:Xerox's message of Fri, 9 Apr 1993
Krankor:
>>It also says that many simple verbs have the identical word as the noun
form. Absolutely no rule is given to guide us in when to assume that such a
noun exists. <<
>>You really don't
want to presume a noun form but give the reader no clue that that's what you
did, cuz when they go to look it up, all they'll find is the verb.<<
Until a rule is revealed to us, all such presumption is dangerous. As usual,
our problem is the scarcity of informants and the fact that Dr. Okrand has
apparently monopolized the only one available. We also lack any extended texts
in Klingon that we could analyze (a few epics on the line of the Odyssey, the
Kalevala, Norse Sagas, Sinbad, etc would be very welcome).
Unfortunately, we really don't have a complete enough grammar or lexicon to
support creative composition, so the pressure to go beyond the published
description, improvising and presuming, is very strong.
The Klingon analyzer program currently aims to reject any word that is not
sanctioned by the available rules and lexicon. So any word involving a
"presumed" de-verbal nominal root will not be accepted. This is the
conservative approach.
Until a better dictionary and reference grammar are available, I think that I
can do the following:
1. Define a nominalizing suffix that comes directly after verb roots and turns
them into noun roots. This would be a "null" suffix, meaning that it appears
at the abstract "lexical" level but does not actually appear in surface
strings. If we let the pound sign # be that null suffix, then Quj (N "game")
would actually be Quj+# at the lexical level, the nominalized form of Quj (V
"play a game"). These Quj words are presently listed separately in the
lexicon. I currently find "Qub" for "to think," but no simple entry for the
noun "thought." If I make # a productive suffix that can appear on any verb
root, turning it into a noun root, then a surface ?QubmeywIj would analyze as a
lexical Qub+#+mey+wIj with a gloss something like think-NOMINALIZED-Plural-my
(i.e. "my thoughts").
2. To reflect the conservative approach to the lexicon, I can write a rule
that effectively kills all such nominalized verbs. That rule would normally
suppress all analyses such as Qub+#+mey+wIj. But the user could consciously
choose to turn off this rule, allowing such analyses to succeed. (I have
taken a similar approach to "clipped" Klingon analyses, which are normally
suppressed by a rule. This reflects the fact that clipped Klingon is slightly
substandard and might often be inappropriate, especially in written texts.
Here again the user can turn off the rule and allow the clipped analyses to
succeed.)
This solution should allow us to have it both ways until we get better
descriptions.
We should not be too surprised if the semantic relationship between the verb
and the deverbal nominal is not completely transparent. We should also not be
surprised to find that some verbs nominalize in this way and other just don't
(i.e. native Klingon speakers may simply reject many forms like *QubmeywIj).
Derivational morphology is typically idiosyncratic, and in the end only careful
fieldwork can resolve the questions.
Comments welcome.
Ken
