Theiling Online    Sitemap    Conlang Mailing List HQ    Attic   

Jaibi Distributives, Reflexives, and Reciprocals

From:Christopher Bates <chrisdb@...>
Date:Saturday, September 13, 2008, 11:31
Hi all,
         I posted this on the ZBB too, but any thoughts would be
appreciated.

I've been recently considering how to form reflexives and reciprocals in
Jaibi, and how this interacts with the marking of distributivity. For
those who haven't read any of the stuff I've posted before, a brief
description of the semantics of Jaibi distributives can be found below:

distributive If the referents of a noun are regarded as being internally
divided in a way that is relevant to the discourse, the noun normally
takes the distributive marker. The stem+distributive form is neutral
with regard to the size of the divisions: the minimum subset could
consist of one individual or a number of individuals.

Jaibi does have offer methods of forming pseudo-reflexives, including a
middle voice which covers some of that semantic ground, but here I'm
going to focus on the use of a separate reflexive/reciprocal pronoun.

Now, what I have been thinking is as following. In a clause, this tends
to be used to assert that a particular event occurred / property holds
separately for each subset of the greater set, so it has an effect
similar to the following:

A-dist went

->

A1 went
A2 went
A3 went
....
An went

Where A1 U A2 U .... U An = A (in other words, where the union of all
the A1..An subsets is A). In some ways, this is similar to English
"each", although with English each the subsets are strictly of
cardinality one:

each of the men went

->

man1 went
man2 went
...
manN went

where "the men" refers to man1 + man2 + ... + manN

Notice that in English any reflexive or reciprocal pronoun referring
back refers to the individual and not the whole set:

each of the men hit himself
*each of the men hit themselves

Now, what marking should NPs in Jaibi reflexive constructions take? The
following are possible:


ANTECEDENT REFLEXIVE PRONOUN
coll coll
coll dist
dist coll
dist dist

Here coll = collective/non-distributive, and dist = distributive. I
think it's clear that the antecedent needs to be able to vary. Consider
the following examples:

(1) the man hit himself
(2) each of the men hit himself

Since in (1) the antecedent is singular, distributive marking is not a
possibility, whereas in (2) the kind of quantification implied by each
would normally mean a distributive in Jaibi.

The question, then, is what marking the reflexive pronoun should take.
It seems to me that it depends exactly what the antecedent of the
pronoun is. I think there are two possibilities:

(1)If the antecedent is the entire set of referents "the men", then only
a subset is involved in any one hitting act, so you would expect
distributive marking to agree with the antecedent.

(2)If the quantification of the "the men" (and its splitting into
subsets) has scope over the reflexive pronoun, and the antecedent of the
reflexive pronoun is a particular subset only, then you would expect
collective marking.

Now, I am leaning towards (2) as the most natural solution, but this
obviously has implications for the exact nature of distributivity in
Jaibi, and how it alters the semantics of a clause. What do you think?
Does (2) make sense?

Another question: can plural reflexives always be reduced to a set of
singular reflexives, or not? For example:

"The men hit themselves"

->

Man1 hit himself
Man2 hit himself
...
ManN hit himself

Does it make semantic sense to have plural reflexives which cannot break
down in this way? Can anyone provide an example? If plural reflexives do
always reduce in this way, then in Jaibi distributive marking on the
antecedent should be obligatory when the referent is a set of
cardinality strictly greater than one.

This leads on to reciprocal constructions. I want to use the same
pronoun for both reflexive and reciprocal constructions, but again there
is the issue of marking. The possibilities, as before, are:

ANTECEDENT RECIPROCAL PRONOUN
coll coll
coll dist
dist coll
dist dist

Now, I am in the opposite situation for reciprocals as I was for
reflexives. Since a set of referents must be internally divided in order
to act against each other, I think the reciprocal pronoun should always
be marked as distributive. The issue is how the antecedent should be
marked. I can see a few arguments:

(1) Since the set is inherently divided, the antecedent should also be
marked as distributive.

(2) Since the set of referents is united in performing the action, but
divided in receiving it, perhaps the antecedent shouldn't be marked as
distributive?

(3) As a variant on two, perhaps there should be a semantic difference?

The men hit each other (collective antecedent)
Each group of men hit each other (distributive antecedent)

I am less sure about what option to each here. I am leaning towards
answer (1), that for reciprocal meanings both the antecedent and the
reciprocal pronoun should be marked as distributive.

Note that, whatever the answer, it seems likely that the primary marking
of reflexivity vs distributivity will be whether distributive marking on
the reflexive/reciprocal pronoun occurs or not.

Anyway, does any of this make sense? What options do you think make sense?

Chris.