Re: mut

[Colin Fine <colin@kindness.demon.co.uk>]

vecu'u le notci po'u <01be8b8d$85f30b20$LocalHost@jorge> la "=?us-
ascii?Q?Jorge_J._Llamb=EDas?=" <jorge@intermedia.com.ar> cu cusku di'e
>
>This brings me to a recent comment by Colin about the meaning
>of Michael's {pa lei karce}, which was intended to mean "one of
>the cars" and Colin took it to mean "the one mass of cars". I tend
>to prefer the first meaning because it is so much more useful and
>cannot, as far as I can tell, cause any ambiguity. I would tend to
>interpret a quantifier of individuals (pa, re, ci, su'o, ro, so'i, etc) as
>itself converting from mass to individual bypassing the need to
>use {lu'a}. (Of course pisu'o, piro, piso'i, etc still work for masses.)
>If that is acceptable, then in this case we could also say:
>{e'u ro ma'a tugni}. Another example (used by several people) is
>{coi ro do}, "Hello to each of you". If not interpreted like this, {ro}
>is pretty meaningless there since there is only one "mass you".

I like this suggestion (which I hadn't thought of). But I'm not entirely
sure it works in general. The problem is that (in the feature analysis I
am still toying with) I see +/-mass (kamgunma) as a feature of sumti and
terbri, which must then match for a sentence to be semantically well-
formed. I have no problem with an operator (some cmavo) explicitly
altering this feature - 'lei' explicitly sets +kamgunma, and "lu'a" (if
I've got the right one) sets -kamgunma. But your suggestion means that a
na'uvla may or may not change the feature, depending on its numerical
value, and I'm not happy with this. 
I need to think about this further - it may depend simply on the
presence of 'pi', in which case we can in principle analyse the features
without having to determine the meaning; but I'm not sure.

On further thought, I don't like it. I think that your suggestion, while
probably workable, is unnecessary, and pandering to those that think
'lei' is plural.

I don't agree that 'do' is a mass: it seems to me much more reasonable
to individuate it.

I suggest that almost all uses of your terkancu to unmassify will be
followed by a 'lei' or 'loi', in which case the appropriate question is,
why has the utterer chosen to use a massifier and then immediately
counteracted it? Is there a difference between 're lei prenu' and 're le
prenu'? 
I would like to encourage precision (in the use of masses/individuals)
by requiring an explicit unmassifier (or fractional quantifier)

I see this as a similar kind of argument to the sumti raising question.
We realised that we were fudging an issue (in my terms, the feature
+/-kamsucta), and actually changed the language to allow precision, and
starting teaching people to use that precision.
I claim to have found a similar imprecision in the use of another of
Lojban's unique features, and have proposed (without changing the
language) a way of being precise. Your solution introduces an *implicit*
conversion which allows unwary speakers to fudge the issue again. It is
implicit because (I think) it will in general be necessary to evaluate
the quantifier in order to determine whether it is converting or not
(and what if the quantifier evaluates to .99999?).

You say it is more useful to interpret it your way - I disagree. It
would be more useful only in the sense that allowing 'mi gleki lemi
bersa' would be more useful: finding a way to assign a meaning to an
inherently imprecise construction simply to allow speakers the luxury of
not having to think about how they are using the language. 

But then I have never approved of omitting 'lo' in 'ci prenu' .u'i
-- 
-----------------------------------------------------------------------
|     Colin Fine    66 High Ash, Shipley, W Yorks. BD18 1NE, UK       |
|  Tel: 01274 592696/0976 635354  e-mail:  colin@kindness.demon.co.uk |
|        "Don't just do something! Stand there!"                      |
|              - from 'Behold the Spirit' (workshop)                  |
-----------------------------------------------------------------------