double accusative (was Re: passive of causatives)

[Boris Oguibenine]

A 13:06 10/07/98 +0200, vous avez écrit :
>Jacob Baltuch wrote:
>>Dominique Thillaud wrote:
>>
>>>Sandra van der Geer wrote
>>>>I agree with you in considering one accusative the direct object, and the
>>>>other accusative the indirect object, or the dative. As we all know, one
>>>>and the same function or kAraka can be denoted by several cases.
>>>
>>>        I can't agree: to take one of both accusatives as a dative is an
>>>anchronism, projecting ancient syntax to our own and above all forgiving
>>>the pragmatical shift of some verbs. I believe that, in ancient time, to
>>>give, to spell and some other ones were direct actions on the receiver,
>>>constraining him strongly.
>>
>>We were talking about an example with the verb 'darzayati' I believe in the
>>meaning of 'show'.
>>
>>That the 2nd accusative of the (theoretical) double accusative construction
>>of darzayati was considered an indirect object by the users of the language
>>themselves is made plain by the fact that that accusative is much more
>>commonly
>>replaced by a dative or a genitive.
>
>Pace Panini I feel rather uneasy about the use of meta-grammatical
>arguments, because they invite to pressing linguistic data into
>preconceived schemes of interpretation. In this case: "tam pustakaM
>darzayati" is felt to be equivalent to English "he shows him a book"
>(German "er zeigt ihm ein Buch"), and since "him" in this case can be taken
>(in analogy to the German "ihm") as an indirect object, we feel entitled to
>interprete the accusative "tam", too, as an indirect object. But is it not
>more sound to take case endings seriously and to understand the sanskrit
>sentence as "he makes him see a book", where we feel no doubt about the
>direct object character of both "him" and "the book"? And the fact that
>"darzayati" could (since when?) be used with a dative or genitive shows
>only that its semantics gradually shifted into the direction of "to show"
>(i.e. "to give someone something to see") so that it could be used in
>analogy to verbs like "dA". But that does not entitle us in my opinon to
>call the accusative "tam" an "indirect object".
> Georg v. Simson
>
>


It is however useful to recall that "tam pustakaM darzayati/dadAti" can be
analyzed as possessing three semantic roles.The agent "he" acts upon "a
book" which is an affected object or patient and upon "him" which is the
recipient. On the other hand a current distinction is between the high
transitivity (a book is given or given to see) and the low transitivity
("he" is only secondarily affected by the agent). The point is that
"pustakam darzayati/dadAti" is a complete utterance, whereas"*tam
darzayati/dadAti" is an ellipsed utterance which may be understood
adequately, only if an additional information about the object of high
transitive verb (which information may be recovered from a foregoing
sentence or the general context or the accompanying gestures), is provided.
Thus the two verbs are semantically akin, but their semantic features are
distributed differently depending upon the semantic roles involved.

Boris Oguibenine