Re: New Project

I want to announce that I'm temporarily putting aside my "abbreviated
reference grammar" of Hwendaaru (which unfortunately is still less than half
finished) to do some preliminary work on a new project that I'm very excited
about, one which brings together a number of ideas that have been knocking
around in my head for several months now.

The idea that precipitated this, the central new idea around which the old
ideas are coalescing, is that I think I've figured out how to use
proximate/obviative marking on 3rd-person noun phrases and direct/inverse
marking on verbs not only to distinguish between agents and patients of
transitive verbs (as in the Algonquian languages) but also, in combination
with word order, to mark topic and focus.  The basic idea is to start with
the same basic word order as Tokana, which I've always found very appealing
(underlying verb-initial, with both topic- and focus-fronting in main
clauses), and add to that the rule that fronted proximates are interpreted
as topical and fronted obviatives as focal.  It gets more complicated for
topics and foci that are 1st- or 2nd-person (and thus neither proximate nor
obviative) or aren't NPs at all (prepositional phrases, adverbs, etc.), and
for subordinate clauses (which must remain verb-initial), but I've already
got a rough idea how to deal with these problems.  (In other words, I've
given it enough thought to satisfy myself that it's complicated enough to be
interesting but not too complicated to be doable.)

For those of you who may be wondering what proximate/obviative and
direct/inverse marking are, the basic idea is that there's a hierarchy of NP
types, with (in the Algonquian languages) 2nd person at the top, then 1st
person, then 3rd person at the bottom.  (In some other language families
that use this kind of system, 1st person is above 2nd person, but they're
both always higher than 3rd person.)  If there's more than one 3rd-person
argument in a clause, one of them is proximate (roughly, the one that's most
topical) and the others are obviative, with proximate being higher in the
hierarchy.  (In the Algonquian family, this distinction is explicitly marked
in the noun morphology; in some others it's determined by word order and/or
context.)  A transitive verb is morphologically marked as either direct or
inverse.  If it's direct, of the two core arguments, the one higher on the
hierarchy is interpreted as the subject or agent, and the lower one as the
object or patient; if it's inverse, it's the other way around.  There's
somebody on this list (I think Laurie Gerholz) who could tell you more about
how this works in the Algonquian family; there's also at least one other
conlang, Dirk Elzinga's Tepa, that uses this kind of system.

One potential problem with this system is that, for a transitive verb with
two 3rd-person arguments, the system only works if at least one of the
arguments (the subject or the direct object) is proximate.  I'm not sure how
Algonquian languages handle the situation where some other NP (an oblique
object) is more topical than either of the two core arguments, but I'm
borrowing something from the Bantu languages: a set of verbal derivational
processes called applicatives, that turn indirect or oblique (prepositional)
objects into direct objects.  I'm also using applicatives for
relative-clause formation, because I'm using a method of relativizing that
only works for subjects and direct objects.

I'll tell you about this in more detail sometime soon (I hope).

-------------------------------------------------
Tim Smith
timsmith@global2000.net

The human mind is inherently fallible.  It sees patterns where there is only
random clustering, overestimates and underestimates odds depending on
emotional need, ignores obvious facts that contradict already established
conclusions.  Hopes and fears become detailed memories.  And absolutely
correct conclusions are drawn from completely inadequate evidence.
        - Alexander Jablokov, _Deepdrive_ (Avon Books, 1998, p. 269)