`real' Sanskrit vs `conversational' Sanskrit

[Madhav Deshpande]

	In response to a discussion on modern varieties of Sanskrit, the
following comment was recently made:

>Ten-day HSP Sanskrit has not gone quite as far as Latino sine flexione,
>but by the same criteria we can say that it is no longer "real" Sanskrit
>(in the sense that it is far from what Madhav Deshpande aptly termed the
>"full variety"). 

	While I would support the use of the term "full variety" to refer
to the high-end variety of modern Sanskrit seen in scholarly Sanskrit
works, Sanskrit research journals like Saarasvatii-Su.samaa, and the high
quality Sanskrit poetry of poets like Bhaaskara Varnekar, we need to take
a more studied look at the kind of Sanskrit taught by the HSP.  While it
is clearly not "full" in terms of its relatively small inventory, I would
not call it "unreal".  It is, to use a more neutral expression, a "subset"
of the full variety.  It sticks to a narrower set of choices which are
indeed available in the full variety.  It is possible, as suggested in
several email messages, that a usage like raama.h graamam gatavaan may be
preferred in the south because of its similarity with some constructions
in Kannada and Tamil.  However, we should not lose sight of the fact that
it is a construction fully allowed by the Paninian grammar.  Increased use
of past participal constructions, in the place of finite verb
constructions, in the medieval narrative works is in all likelihood
related to the ergative construction in the mother-toungues of the users
of Sanskrit.  What seems most interesting is that different local
varieties of simplified Sanskrit make different choices for their subsets.
To the extent these subsets do not violate the rules of the full variety,
these subsets could serve as elementary steps toward a later acquisition
of the full variety.  However, if the subset violates the rules of the
full variety, for example by substituting the plural for the dual, then
this is indeed not a subset of the full variety, but a compromised form of
Sanskrit.  Such compromises were suggested by some innovators in Pune when
I was growing up.  One of these was to have all verb roots moved to the
first conjugation, and of course to get rid of the dual.  I would be very
careful to avoid such compromised varieties.  But, otherwise, a subset
variety may be useful for certain pedagogical purposes.  I remember that
when we were learning Sanskrit at the Tilak Maharashtra Vidyapith in Pune,
we did not go beyond the past imperfect for quite some time.  It was only
later that we were introduced to the forms of past perfect and aorist.
The last one was particularly felt to be a difficult acquisition, and
being able to use aorist was considered to be a great accomplishment.
Here, the decision to start with a smaller subset of grammar was indeed a
pedagogically sound decision.  If one were to stop before acquiring the
past perfect and the aorist, one still had a legitimate subset of Sanskrit
grammar.  Thus, each variety of Sanskrit needs to be carefully analysed to
see whether it is a legitimate subset of the full variety, or whether it
involves compromises which violate the rules of the full variety.  
					Madhav Deshpande