That is, I know, the kind of subject line to make those who lived
through the gadri wars shudder.
But my question is simple and hopefully simply resolved.
From the gadri BPFK section:
"""
An individual can be anything, including a group, a set, a
substance, a number, etc. {lo broda} can refer to one or
more individuals. {lo'i broda} can refer only to those
individuals that are sets. {loi broda} can refer only to
those individuals that are groups ('masses').
"""
From the Indirect Referers section
"""
lu'a (LAhE)
Individual. (Member.) 1. It converts a sumti into
another sumti. The converted sumti points to the
referents of the unconverted sumti, removing any
indication of collectivization if there was any.
"""
These seem to indicate conflicting semantics for handling of masses.
(I'll use 'mass' in the CLL sense, intended to be synonymous with
'gunma'/'group'/'collective'/'plurality')
The first appears to indicate that masses are still first-class
entities, such that e.g. in {lo tadni cu sruri le dinju}, the referents
of {lo tadni} are masses of students, not individual students. In
particular, it is reasonable for that {lo tadni} to have just one
referent.
The second appears to indicate that sumti can sometimes be 'flagged' as
being interpreted collectively - the referents are the same whether it
is so flagged or not, but if it is so flagged then a bridi involving it
is understood to hold of the mass consisting of the referents, rather
than distributively of the referents themselves.
Admittedly, these two interpretations are not literally inconsistent - you
*could* have first-class masses *and* mass-flagging, it would just be
very confusing. Is this really what was intended?
Things are confused even furtherly by the example given on the gadri
page of:
lo tadni cu sruri le dinju gi'e krixa
Students are surrounding the building and yelling.
, which seems (in the context of the use of this kind of example in the
lingustics literature) to suggest that the referents of {lo tadni} are
acting as a mass in the first bridi and distributively in the second.
Which would need the distributivity flag to have third value of
"ambiguous", or something like that...
Personally, I think the first interpretation (first-class masses which
gadri can return) fits best with the rest of lojban - although it leaves
open the question of how to specify that you *don't* want masses as the
referents when using gadri... {ro lo tadni} is no good, as it could be
interpreted as quantifying over some (perhaps just 1) masses which are
the referents of {lo tadni}. {lo tadni poi na gunma su'o tadni} is the
best I can come up with.
Hoping for clarification,
Martin
Attachment:
pgp4fFzQGfzn_.pgp
Description: PGP signature