[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
la pycyn cusku di'e
Namely riders with {lo broda} apparently retain the implicit plurality of
{lo
broda}, which is why {da poi broda} is useful occasionally (even when not
shoving negations around).
I see no difference between:
(1) ko'a goi lo broda cu brode
and:
(2) ko'a goi da poi broda cu brode
To me they make identical assignments to {ko'a}. I don't think
there is any stronger hint of plurality in (1) than in (2).
This can lead to some confusion down the pike, if
a {ko'a} introduced for {lo broda} starts being used in a way that implies
that it is singular (without retroactively declaring that {lo broda} was on
this occasion).
Using {ko'a} down the pike should be like using {da}, with the
advantage that you don't have to use ge-gi or similars to make
sure everything stays within the scope of {su'o}. In other words:
ko'a goi lo broda cu brode
...............
... ko'a ......
is equivalent to:
ko'a goi da poi broda cu brode
...............
... ko'a ......
and also to:
da poi broda zo'u ge da broda
gi ge ..........
gi ... da ......
I'm surprised that {prenex tu'e ... tu'u} is not grammatical!
I wanted to write:
da poi broda zo'u tu'e da broda
.............
... da ...... tu'u
But it is not grammatical
Singularity in such cases comes by quantifying (with {pa}
presumably) on {ko'a} ({pa} is the second easiest way to get an individual
in Lojban).
{pa} as quantifier is more complex than singular terms. The
only fool-proof assignments of {ko'a} are with singular terms.
Assignments with quantified terms are always messy.
Matters get worse when what is {goi}d is something where theindividuals are buried but may want to act independently: {loi}, for example,which requires demassing loi broda (or ko'a) to get back to individuals.
If the assignment is a piro-mass, there is no problem, as it is
a singular term and so it is transparent to other quantifiers and
negations. (To get the individuals you can then just use {lu'a
ko'a}). Assignments of pisu'o-masses, on the other hand, have the
same problems as assignments of su'o-quantified terms.
In short, anaphora keeps the character of what is anaphorized.
In terms of individual/mass yes. The problem is that when there is a quantifier, the anaphor must remain forever under its scope, creating horribly extended scopes for these quantifiers and therefore for any other quantifier/negation/etc with scope over them.
I suppose the buried {ko'a} in your example could be dealt with by some ruleto avoid madness -- subscripting by the various instances of {le ci nanmu}seems as likely as any, but the assignment would be neater if made after theseparation of the three cases.
I don't see how subscripting would help, and in any case it is
easy to create even more horrible cases:
le ci nanmu na kansa da poi ninmu zi'e goi ko'a
It is not the case that each of the men was with a woman, ko'a.
No subscripting will help here. {goi}ing quantified terms is tricky.
It is important to keep in mind that ko'a must remain always under
the scope of the quantifier, and extending scopes indefinitely
without creating great confusion can only be done in very special
cases. Probably an outermost {su'o} is safe, but not much more.
mu'o mi'e xorxes
_________________________________________________________________
Get a speedy connection with MSN Broadband.� Join now!
http://resourcecenter.msn.com/access/plans/freeactivation.asp