TECH.ADV - non-specific SE

["Mark E. Shoulson" <shoulson@CTR.COLUMBIA.EDU>]

Hmm.  At first glance, Colin's unspecified SE looks very useful and
promising.  It should be noted, though, that its semantics are a little
more powerful than any simple conversion, otherwise I might recommend doing
it via {jai do'e} or some such.  The proposed SE word indicates not just
*an* unspecified conversion, but rather *any number* of unspecified
conversions.

Another important thing I notice is that it's actually even more than that.
Not only do we have arbitrary scrambling of all the sumti places, but even
actual *merging* of sumti places, something which isn't otherwise allowed
(I recall a discussion involving {bai je va'u} vs. {fe ja fi} and such, but
this is conversion, not FA marking).  That is, Colin proposes that {le xe'e
broda} mean not "the filler of some unspecified place of broda" -- that
would be {le jai do'e broda} (hmm.  Is there an unspecified FA marker, as
distinct from unspecified BAI marker?), but rather "the filler(s) of
unspecified place(s) of broda", a rather different affair (see his first
example, le xe'e tanru).

Note that this makes its semantics very confusing, not so simple as they
were before.  {mi la mrvn. xe'e patfu} means more than just "Mervyn and I
are in some father relationship, each of us filling one role", but rather
"Mervyn and I are in some father relationship, each of us filling one or
more roles", meaning not just that we may be father and child, or child and
father, or mother and father, or mother and child, etc, but also that I may
be Mervyn's mother *and* father (nothing terrible about that, recalling
that {mi} may be plural), or that I/we may be Mervyn's child and mother of
that child, and so on.

This may not be a bad thing, but it should be considered.

~mark