WikiDiscuss

---

> On 5/16/05, John E Clifford
> <clifford-j@sbcglobal.net> wrote:
> > My point would be that BAI that don't have
> usage
> > are not BAI at all but only potentially BAI.
>
> I'd be very happy to eliminate many, perhaps
> most, BAIs
> from the language. I doubt very much anything
> like that
> will happen though.

Relevance? I was ovbjecting to BAI that are not
iin the language to begin with (except in these
barely justifiable examples).

> > They have no place in a dictionary, or, at
> most,
> > should be mentioned as potential forms with
> some
> > indication of where there usage might be
> expected
> > to lie — but obviously not exact
> specification
> > because we do not yet know what that is (if
> ever
> > anything).
>
> Given that they will all appear in the
> dictionary (we may
> not like it but that's how it's going to be)
> that's exactly
> what we are doing, i.e. providing indications
> of how we
> expect them to be used. A lojban sentence is a
> much
> better way of doing this than out of context
> English
> keywords, in my opinion.

Well, I agree that keywords are generally not
very useful — even misleading in many cases --
but my protest is exqactly to including them in
the dictionary when they have never occrred in
the language. Will all the 100000 or so gismu be
included? What about all the cmavo in /x/? Or
the as yet unused CV'V'VV? And so on. It isn't
even that there are plausible cases that these
BAI forms ought to be used, in most instances.
And, of course, saying that the English sentence
gives some indication of how they are to be used
is only partly (and not the most useful part)
true; it tends in fact to fix the meaning rather
than open the range.

> > That is, the way to build a
> > dictionary — at this point in the language
> — is
> > to find and explain the forms actually used,
> not
> > to create and explain every possible form (an
> > impossible task anyhow, so why even mess with
> a
> > half-assed job).
>
> That assumes that the language as actually used
> is
> already good enough to be worth solidifying. I
> disagree
> with that, I think most current usages are of
> relatively
> poor quality and so a good deal of prescription
> is still
> needed.

I have no objection to clarifying — even
prescribing to some extent — the existing usage.
It does seem to me that some expressions have
been used inconsistently (or at least unclearly)
and inappropriately for their intended roles (as
subordinate clauses, for example, rather than
added places in the case of BAIs). But that is
very different from creating NEW usages out of
whole cloth (and based on nothing real at all).

> > Notice that, had the present program been
> carried
> > out from the beginning, a large number of the
> > established forms would not have the meaning
> they
> > do and a number of useful (and used) notions
> > would lack forms of the simple BAI type. I
> would
> > expect this pattern to continue when the
> program
> > is put in place starting after the beginning.
>
Well, we have discussed half a dozen cases over
the last little while: two that stick in mind
without searching are {du'u}, whose meaning is
not related by the "rules" to {djuno} (more to
{jinvi} or {krici} — which pair of words needs
some work, come to think of it) and {ri'anai},
whose meaning violates both {rinka} and {nai}, as
typically understood in this game of BAI
creation, and is not easily constructed in this
way from other brivla.