Message-Id: <m0qEjFG-00001WC@xiron.pc.helsinki.fi>
Date:         Fri, 17 Jun 1994 16:54:30 BST
Reply-To:     C.J.Fine@BRADFORD.AC.UK
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         Colin Fine <C.J.Fine@BRADFORD.AC.UK>
Subject:      Re: sumti categories
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 2323
Lines: 67

Jorge: Yes -- and no


I started thinking about (sub-)categorisation a while ago -
in fact I rather think I posted something on it.

We have already two strong instances of sub-categorisation of
tergi'u - individual/set/mass and abstraction.

Where Jorge goes much too far, is in assuming that everything has
just one category.  This simply doesn't work.

My suggestion is that we subcategorise on a number of features, which
need not always be specified.

COLLECTIVES and GADRI
______________________

Thus for type of collectivity we have features +/-set and +/-mass

Most terbri are -set but unspecified for mass
gunma is +mass
klesi and te cnano are +set

Then gadri also may have markings
(le/lo/la are unmarked; lei/loi/lai are +mass, le'i/lo'i/la'i are +set)
and these may interact with the relevant terbri of the selgadri:

thus I suggest
le gunma        inherits +mass from gunma
lo klesi        inherits +set from klesi
lei remna       is +mass by virtue of the gadri
lo'i badna      is +set by virtue of the gadri.

When the gadri and the te bridi disagree on the marking, there is a sort
of semantic construction implied:

lei te cnano    sort of constructs a mass (+mass) out of the +set elements
                specified by te cnano, and by the nature of a mass is itself
                +set.
On the other hand
le'i gunma      constructs a set (+set) out of the mass(es), but is not itself
                +mass because that is a property of the members of the set.

ABSTRACTIONS and SUCMA'O
_________________________
Then there is a quite separate feature +abstract. (The x1 of) all sucma'o
(i.e. NU words) are +abstract, but so are many other terbri, eg
se djica
te djuno
fasnu.

I am certain that there are further subdivisions of +abstract (eg +/-event,
+/-concept) but I have not worked out in detail what they are. I doubt
very much though whether even they are mutually exclusive.

tu'a of course is a specific operator which delivers +abstract, and
unspecified for all dependent features.

I think there are probably other features not in either of these categories
(eg +/- animate, +/-personal) but I'm not sure.

Once all terbri, gadri, sucma'o, sumga'ima'o (eg LUhI) and a few others
have their features specified, it will be possible to check a sentence for
category consistency.

        Colin