Sanskrit

[Elliot Stern]

>>Let us now look at the Sanskrit examples Sandra van der Geer
>>provided (in a message Re: Sanskrit Fri, 3 Jul 1998 09:37:14 +0200)
>>
>>1) rAmo nalAya pustaka.m dadAti
>>2) rAme.na nalAya pustaka.m dIyate
>>3) nalo rAme.na pustaka.m dApyate
>>
>>1) means "R. gives N. a book"/"A book is given to N. by R."/"N. is given a
>>book by R.".
>>2) means "R. gives N. a book"/"A book is given to N. by R."/"N. is given a
>>book by R.".
>>3) does not mean the same thing as 1) and 2). I believe it should be
>>paraphrased as "raamo nalaM pustakaM daapayati", and mean "R. makes N. give
>>a book"/"N. is made by R. to give a book" (N.B., if Apte explains an
>>utterance like this, I missed it. So be cautioned that the source of this
>>interpretation (Elliot) is a lesser, and fallible, authority!)*
>>
>>If Sandra meant that 3) is semantically equivalent to 1) and 2), I too
>>would like to know how that can be so.
>>
>>Elliot
>
>        Perhaps just a problem with 'semantically'.
>        At low-level we can consider passive and causative being reciprocal
>functions:
>1) le chat mange la souris
>2) * le chat fait être mangee la souris (= le chat fait que la souris est
>mangee [par le chat])
>3) * le chat est fait manger la souris (= que le chat mange la souris est
>fait [par le chat])
>        Evidently, 1) = 2) = 3) is just a low-level equivalence, not a
>fully semantical equivalence. The same occurs with the double negation: not
>just an assertion but a strong assertion.
>        Regards,
>Dominique

I think Sandra was thinking about 'semantic equivalence', not 'low-level
equivalence'. I infer that from her reply message  re: Sanskrit Wed, 8 Jul
1998 21:37:39 (my local time) beginning "Eliot Stern wrote".

When I said 'semantically equivalent', I meant that each utterance said to
be 'semantically equivalent' to another utterance may be considered as a
paraphrase of that other utterance. The given utterances may then be said
to express the same meaning. Evidently, at least two previously unstated
assumptions apply: first, the examples should be utterances that one might
speak or hear in the course of ordinary, everyday discourse; second, all
the examples in a given set of utterances use only the same bases (here,
for example, raama-, nala-, pustaka- and daa-). There seems also to be a
third assumption such that each base is used the same number of times in
each example (and there may be other assumptions). I think that the
paaNinIya and other commentators are asserting such 'semantic equivalence'
when they say that two utterances like

1) rAmo nalAya pustaka.m dadAti
2) rAme.na nalAya pustaka.m dIyate

are equivalent. They would *not* assert that

3) nalo rAme.na pustaka.m dApyate

is in this way equivalent to either 1) or 2), or that 1) or 2) is
equivalent to 3). While 3) satisfies seems to satisfy all of the
assumptions that I just stated, it falls outside of the set simply because
speakers would understand something rather different from what they
understand when they hear 1) or 2), when they hear 3). Speakers would,
however, understand much the same thing upon hearing either 1) or 2).

Dominique's French examples 1), 2), and 3), which may be rendered in
English as:

1') the cat eats the mouse
2') the cat causes the mouse to be eaten [by the cat]
3') the cat is caused to eat the mouse [by the cat]

make a strikingly different sort of set from Sandra's Sanskrit set. The
Sanskrit set consists of three utterances that we could imagine the average
devadatta uttering to the average yajJadatta on the road. The French and
English sets consist of one utterance that the average Jean, Jeanne, John
or Jane might utter (le chat mange la souris / the cat eats the mouse). on
the street. But the second and third both in the French and English are
most assuredly not utterances of the ordinary man/woman in the street, but
are most probably the utterances in some grammatical metalanguage.






Elliot M. Stern
552 South 48th Street
Philadelphia, PA 19143-2029
USA

telephone: 215 747 6204