Cowan weighs in #1: specific, definite

[Logical Language Group <lojbab>]

Specific/definite as I understand them (these terms are part of linguistics
jargon, and neither linguistics nor logic is my profession):

SPECIFIC (sometimes spelled "+specific"):  a reference is specific if the
speaker's intention fixes the referent.  English pronouns are specific,
most English uses of "the" are specific (but note American English "I am in
the hospital" which is non-specific), Lojban "le" is specific.

NON-SPECIFIC (sometimes spelled "-specific"): a reference is non-specific if
it is not specific.  English "a" is sometimes non-specific, Lojban "da"
is non-specific.

DEFINITE (sometimes spelled "+definite"): a reference is definite if the
listener knows the referent.  English "the" is usually definite.

INDEFINITE (sometimes spelled "-definite"): a referent is indefinite if the
listener does not know the referent.

Except in pathological cases, a non-specific reference is always indefinite:
if the speaker does not fix the reference, the listener cannot know it.

For completeness, I will add:

VERIDICAL: a description (not any reference, but one which contains what
purports to be a property of the referent) is veridical if the referent
must have (se ckaji) the property.  Lojban "lo" is veridical.

NON-VERIDICAL: a description is non-veridical if the referent may or may not
actually have the property.  Lojban "le" is non-veridical.

The open question is whether "lo" is specific or non-specific; or perhaps it
is vague wrt specificity.

	mi klama le zarci	specific
	mi klama da poi zarci	non-specific
	mi klama lo zarci	???

As far as I know, Lojban doesn't mark definiteness in any way.  There has
been some talk of an attitudinal to mark it, but none has ever been proposed.

And holds that "a certain man" is specific; Jorge held that it was not.
I believe that "a certain man" is indeed specific, as the speaker's intention
fixes which man is meant, but it is not definite, because the listener has
no way of knowing.  On this view, the "normalness" of "Which man?" is not a
test of specificity but of definiteness: a listener who says "Which?" to
an indefinite reference is legitimately asking for a referent, whereas the
listener who says "Which?" to a definite reference is expressing his confusion.
(But then again, perhaps I am merely expressing my confusion.)

-- 
John Cowan		sharing account <lojbab@access.digex.net> for now
		e'osai ko sarji la lojban.