From newshub.ccs.yorku.ca!torn!cs.utexas.edu!sun-barr!ames!elroy.jpl.nasa.gov!sdd.hp.com!spool.mu.edu!hri.com!know!mips2!news.bbn.com!ingria Wed Sep 23 16:54:12 EDT 1992
Article 6957 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca sci.bio:3967 sci.skeptic:20177 soc.men:5155 soc.women:4849 comp.ai.philosophy:6957
Path: newshub.ccs.yorku.ca!torn!cs.utexas.edu!sun-barr!ames!elroy.jpl.nasa.gov!sdd.hp.com!spool.mu.edu!hri.com!know!mips2!news.bbn.com!ingria
>From: ingria@bbn.com (Bob Ingria)
Newsgroups: sci.bio,sci.skeptic,soc.men,soc.women,comp.ai.philosophy
Subject: Re: missing verbs
Message-ID: <lbhhjvINNone@news.bbn.com>
Date: 17 Sep 92 17:55:11 GMT
References: <1992Sep9.230021.5182@news.media.mit.edu> <BILL.92Sep9232609@ca3.nsma.arizona.edu> <1992Sep11.120124.15227@techbook.com> <1992Sep16.220726.17031@usl.edu>
Reply-To: ingria@BBN.COM
Lines: 69
NNTP-Posting-Host: bbn.com
In-reply-to: mhf4421@usl.edu's message of 16 Sep 92 22:07:26 GMT

In article <1992Sep16.220726.17031@usl.edu> mhf4421@usl.edu (Flynn Matthew H) writes:
   In article <1992Sep11.120124.15227@techbook.com> szabo@techbook.com (Nick Szabo) writes:
   >
   >Dr. Minsky has pointed out a curious assymetry in the English language,
   >but does any language _not_ have such an assymetry?

   Going back to the discussion of "A beats B" vs. "B loses to A", the problem
   does seem to be fairly cross cultural.  "To Beat" is and active verb, and (I
   suppose) so is to lose.  However you cannot say A lost B unless you mean B is
   something a once posessed and then (voluntary or no) released posession of.
   Lose in the active sense has nothing to do with competition, beating, et. al.

   When we say A lost to B we are saying that A lost (a potential victory in a 
   competion) to B.  B is an indirect object, which I believe someone has already
   mentioned.  The action is indirect: A performed no action directly on B.  In
   fact A has done nothing to B except perhaps giving her the satisfaction of
   winning.

   The alternative form of course is A was beaten by B.  This, the passive form of
   beat, is stated as two words in English, but in other languages "was beaten" is represented by one word, generally with the same root as "to beat" but with a
   different ending.  The problem is stil not solved in those languages, I suppose,as we would still have an extra word: A <verb+passive ending> by B. B is now theobject of the preposition "by"!!!  

Actually, I think the number of words is peripheral.  The main point
of the original observation is that there is a canonical form for
tranistive, active sentences in every language.  While there is a way
of expressing the victory of the subject over the object in this
canonical form seemingly universally (at least, nobody has pointed out
a language that doesn't have active equivalents of ``win'', ``beat'',
etc.), there seems to be no verb in any language that expresses the
defeat of the subject by the object (NB this formulation; does the
purported generalization extends to nouns/nominalizations, too?) in
the canonical transitive, active form.  In English, the deviations
noted have been two:

(1) use of passive voice
(2) use of a prepositional verb (i.e. not a strictly transitive verb)

I suggested that in Case-marking languages, one might discover, in
addition to or instead of (2):

(3) use of oblique marking on ``object'' (i.e. in a
Nominative/Accusative Case marking language like Latin or Greek, the
canonical active subject is marked with Nominative Case, and the
Direct Object is marked with Accusative Case.  One might hypothesize
that, in such a language, parallel to verbs like ``lose to'' in
English, there might be verbs in which the ``Object'' appears in a
Case other than Accusative, such as Dative, Instrumental, Ablative,
etc.)  Here the number of words would be the same as in a transitive,
acitve sentence, but the Case-marking would be different.  There are
languages in which passive morphology appears as an ending on the verb
and the agent phrase is expressed by an oblique Case, so again, the
number of words would be the same, but the sentence would again not
fall into the canonical transitive, active form.

The difference in Case-marking isn't just splitting hairs, because
oblique marked ``Objects'' in languages like Latin and Greek typically
do not behave like full Direct Objects; e.g. they do not passivize.
Along these lines, note that the prepositional object of ``lose to''
doesn't seem to be subject to the so-called pseudo-passive, in which
the object of a preposition is passivized.  Thus, while one can say:
``They lied to me'' or ``I've been lied to'' (pseudo-passive), it
seems pretty bad to say ``*Yale's been lost to''.  So the object of
the preposition in such examples seems to be incapable of taking on
Direct Object like properties, even though it is sometimes possible
for the object of a preposition to behave like a Direct Object of the
main verb in English.

-30-
Bob


