[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >


[Date Prev] [Date Next] [Thread Prev] [Thread Next]

preparing for BF (was: Re: xorxes's simple solution (was: Re: a minimalist fantasy



I propose to post this to the BF, unless there are further comments.

> le                 = Spec
> PA le              = members of Spec
> lo                 = Kind
> PA lo              = avatar of Kind 
> PA lu'a lo         = members of Kind
> PA LAhE-subkind lo = subtype of Kind [Can also be applied to {le}]

({PA lo broda} continues to be equivalent to {PA broda}, with no change
in meaning.)

> It requires only the following changes:
> (1) Bare lo = tu'o lo instead of su'o lo
> (2) Bare le = tu'o le instead of ro le
> (3) {lo PA} != {lo PAro}
> (4) a new LAhE (which Nick can hardly veto, since he's proposing new LAhE
> himself)
> 
> Hardly any usage at all would be invalidated by these changes, and 
> probably a substantial amount of previous usage would become validated.
> 
> Making these pretty minimal changes allows us to say everything we need.
> The remaining existing gadri can then be defined according to whatever
> compromise between usefulness, logical coherence and CLL-conformity
> is deemed best.
> 
> So in all seriousness I say to Nick: accept (1-3) and we will at a single
> bound have solved virtually the whole gadri problem. The only challenge
> left will be to make the remaining gadri logically coherent, but this
> task will be much easier because it wouldn't be freighted with the
> additional requirement of having to afford us ways to say what would
> otherwise not be sayable.

I propose the following addenda:

(a) Define {ro} as whatever number is the cardinality of the set being 
quantified over. There was a massive debate over this last year, and
everybody supported it except John.
(b) At least when the gadri is lo or le:
(b.1) Treat nonfractional quantifiers as fractional/rational expressions.
(b.1.1) For outer quantifiers there is an implicit denominator ro. 
(b.1.2) For inner quantifiers there is an implicit denominator pa (= a 
single broda).
(b.2) When there is an overt denominator PA (as opposed to roPA), the
quantifier expression means "x out of every PA", "x per PA". 
(b.3) As an optional extra for brevity's sake: make fi'u elidable before
ro in these expressions.

I think these involve no further changes. 

CLL {(PA) lo PA broda} becomes {PA (fi'u) roPA lo broda}.
CLL {ro lo PA broda} becomes {roPA lo broda}.

--And.