Sanskrit syntax problem: Double bind constructions

[Hartmut Buescher]

Dear VyAkaraNa-savants,

now and then we meet with ambivalent syntactic constructions in Sanskrit,
that is, with cases where a particular part of a sentence -- say, a genitive
construction -- is strategically placed so as to be connected to both the
preceding and the subsequent elements with equal strength in terms of
semantic possibilities. 
Recently I saw that, for example, Jan E. Houben (in: The saMbandha-samuddeza
[Chapter on Relation] and BhartRhari's Philosophy of Language, Groningen
1995: 6 n. 8) shortly referred to this phenomenon. 

In the context of Buddhist studies it has been the phrase "ekasmin samaye"
in the opening formular of sUtras ("Thus I have heard [: ?] once [: ?] the
Buddha was staying .....") that has triggered a fairly large scholarship
providing interpretations, references to which have been conveniently
collected by Daniel Boucher (on p. 90 n. 1 of the contribution "The
NagaropamasUtra .... " in: Sanskrit Texte aus dem buddhistischen Kanon:
Neuentdeckungen und Neueditionen, Dritte Folge, Göttingen 1996). With
particular reference to Paul Harrison, Boucher's solution is to take
"ekasmin samaye" as what I am here designating as "double bind construction"
(being unaware of any existing technical terms to designate this
phenomenon).

The problem becomes more complex when we have to try to reconstruct the
original Sanskrit wording of a philosophically significant passage, e.g.,
from the Tibetan. 
It is -- just to come up with an example without wishing to discuss the
passage itself -- rather problematic, when Lamotte reconstructs the Sanskrit
original of the term kun gzhi as a verb ["gîte (Aliiyate)"] in the
MahAyAnasaMgraha I.2 passage
chos kun sa bon thams cad pa'i / rnam par shes pa kun gzhi ste //,
because Aliiyate has regularly (though, admittedly, not always) been
translated by sbyor ba (just as in the immediately subsequent passage MSg
I.3).
On the basis of contextually related passages, I regard it as quite probable
that the original Sanskrit read:
*sarvabIjakaM vijJAnaM sarvadharmANAm AlayaH //
whereby sarvadharmANAm would also semantically be closely related to both
the preceding and the subsequent elements. 

Now, in view of such problems I would like to become more familiar with this
type of double bind constructions in Sanskrit. The title of Prof.
Deshpande's book "Ellipsis and Syntactic Overlapping ... " sounded
attractive, but I had to realise that it was not pertinent to the present
concern. The Speijer centennial volume "Studies in Sanskrit Syntax", ed. by
H. H. Hock contains a long bibliography; yet, from the listed titles
themselves I could not immediately spot those that might eventually help
further.

I am interested in all sorts of references drawing attention to instances of
double bind constructions either in connection with more or less extended
theoretical reflections, or just in form of (foot-)notes merely saying:
"here occurs this phenomenon; I deal with it in such a way".

For any pertinent information I would be grateful.

Best regards,

Hartmut Buescher