Re: brain fart metaphor

[ucleaar <ucleaar@UCL.AC.UK>]

> One of the basic rules of tanru making IS that the essential character is
> not sufficiently changed to render the final term places meaningless or
> substantially unlike their original meaning. Otherwise you have NO rules
> for tanru-making, because "rokci cinfo" could be defined to mean the same
> as "bridi" mapping the places in some weird way, and ignoring those that
> do not fit.  We did NOT attempt to make tanru-making have NO rules.

The system you advocate makes sense only if the extension of a tanru is
always a subset of the extension of the tertanru; i.e. Ax if x is <tanru>
then x is <tertanru>. This is the only way in which you could argue that
the meaning of the tertanru is preserved.

As I understand the present system, a tanru has no defined meaning. Thus
the logical form of a bridi containing a tanru is incomplete. The
addressee must choose something to complete the bridi; and as always,
the presumption is that the seltanru and tertanru are relevant to the
choice the addressee must make.
In sum: it's all done by pragmatics.

> >> >{zio} is not metalinguistic. It in effect derives new lexemes. It
> >> >yields a "literal" meaning.
> >> zi'o is defined as a metalinguistic device.
> >Where? It's in KOhA. What is metalinguistic about it?
> We seem to have different ideas as to what metalinguistic means. Anything
> that consciously manipulates or refers to the meaning of an utterance
> is to me metalinguistic.

I recommend you use a different, perhaps lojban, term for this, since
metalinguistic (in the sense that is relevant here) standardly means
a comment on the current utterance such that the comment is truth
conditionally outside the scope of the bridi it pertains to. That's
not put very well. If you don't grock what I mean, say so and I'll
rephrase.

In your sense of metalinguistic, {zio} is - I see that.

> >> It yields a bridi with a "literal" meaning but I cannot envision
> >> anything other than a figurative meaning to, say "mamta be zi'o".
> >It's figurative only in relation to {mamta be da}, but since the literal
> >meaning of {mamta be zio} needn't be {mamta be da}, {mamta be zio}
> >needn't be figurative.
> I totally fail to understand, partly because I cannot conceive of a
> meaning to mamta be zi'o that is not referring to a se mamta.  All of
> the mother-metaphors in English that do not require an offspring are
> figurative, and indeed that type of usage comprises the largest bulk of
> figurative expressions that I am familiar with - they ignore the
> applicability of some place that is essential to the meaning of one
> component word.

The figurativeness of zio constructions is the same as that for tanru
[not taking into account what John says, below]. The relationship between
the meaning of the source brivla on the one hand, and the literal selbri
meaning on the other hand is indeed figurative, but utterances containing
tanru are no more or less likely than any other utterance to be figurative
(i.e. to have a figurative relationship between the literal meaning of the
sentence and the idea that is being communicated).

John:
> There is a constraint that whatever satisfies "mamta be da" must
> satisfy "mamta be zi'o", but not necessarily vice versa.

It's a very stringent constraint, and one I'd forgotten about. It
would limit the places where zio could profitably be used. Is that
the intention?

> la lojbab. pu cusku di'e
> > > - I think the rule about touching your piece requiring it be moved,
> > > absent a French expression, is part of the rules of chess. Is that rule
> > > "grammar" or convention. I was always taught it as a rule.
> la .and. cusku di'e
> > It's a rule of usage. It's not part of the grammar. A chess computer
> > needn't be taught it.
> If "chess computer" is taken to mean "computer that plays the game of
> chess", then "chess" is an ambiguous word: there are two games of chess.
> Most chess programs play what I will call "Virtual Reality chess",
> wherein a certain uninterpreted formal state is manipulated in accordance
> with uninterpreted formal rules until a victory (or stalemate) state is
> reached. But in "Real Life chess", there must be a physical board and
> physical pieces. This is the kind defined by the FIDE regulations and
> played in (FIDE) chess tournaments.  A computer program that plays RL
> chess must have a robot arm or a person simulating such an arm, and the
> total system (not just the program per se) must understand the piece
> touchee rule, or it will play suboptimally.
> (I make this point about chess because I think there is a direct analogy
> between chess and language: what plays in VR language, which is formal,
> may be insufficient in RL language, which is not merely formal.)

I agree with all this. We must still recognize that if we ignore the
distinction between VR and RL rules of chess and language we are missing
important (and in some sense real) insights. I'm not opposed to people
who study RL rules; I'm opposed only to the failure to make the VR/RL
distinction.

coo, mie and