Message-Id: <m0sUUwf-0000ZIC@xiron.pc.helsinki.fi>
Date:         Sat, 8 Jul 1995 03:54:41 -0400
Reply-To:     Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender:       Lojban list <LOJBAN@CUVMB.BITNET>
From:         Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Subject:      rismi
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 9284
Lines: 187

Steven Belknap writes:
>I would argue that it should be possible to describe these concepts
>either discretely or continuously.

You can.

>  The clearest demonstration of
>inconsistency is demonstrated by the dembi/rismi definitions.  If I go
>to the store and buy either navy beans or basmati rice, I pay by the
>pound.  I do not buy individual beans or rice grains.  Why then is dembi
>defined discretely (an instance of one or more) & rismi defined
>continuously (a quantity)?

And later:
>Why should /dembi/ be thought of as individuals and /rismi/ thought of
>as groups?  Note that in English the predicate /beans/, as a food, is
>customarily pluralized.  Sometimes we use the singular, when talking
>about a specific instance of Mr. Bean.  In English rice implies a group.
>/Rices/ implies a group of individual cultivars or varieties of rice.

Why?  English idiom, which we are forced to use for the English
translations.  In the specific example, size is also a factor.  Bean was
listed in the vocabulary in contrast to seed, and except for farmers, we
(English speakers) generally tend to identify seeds as collections of
units, rather than as quantities.  You cook a cup of rice, but you plant
seeds individually.

Where a word could be interpreted either as a mass noun or a count noun,
I chose based on the words I was contrasting it with, for style of
definition.  Bean matched seed, the various grains matched each other,
etc.  Deer, a classic individual mass distinction since the plural is
identical to the singular, is written as a count noun because other
animals are also count nouns.

There is no way to clearly talk about English mass nouns as if they were
count nouns, that does not dictate the unit size - and as we have seen
with djacu, there can be clear disagreement with some of these as to
what the unit size is.

You can argue that the unit of djacu should be molecules, but some
specific properties of water do not apply to small numbers of molecules,
some cultures don't know what a molecule is (Lojban needs to be able to
translate Ancient Egyptian discussions of water, and they certainly
weren't talking about molecules.)  And there isn't much sense in a
"molecule of food" (and there are some things like festi/waste, poisons
and germ-caused diseases where individual molecules don't have the
wastage, poison or disease-causing effect that a larger collection of
units might.

>I disagree.  "Useful" to me requires that it be accurate!  Talking about
>water as an infinitesimally divisible mass is not accurate, at least as
>far as science can tell us.

What is "accurate"?  Are the Trobriand Islanders inaccurate when they
view all rabbits as the same thing?  Accurate means that it reflects the
speaker's concept.  I rarely think of molecules when I talk of water, so
to make my words mean or imply molecules is NOT accurate.

The scientific world-view is a metaphysical bias just as others are.  We
"lo dembi/rismi/djacu/canre/cidja all refer to some kind of countable
portions of the respective substances.  And loi
dembi/rismi/djacu/canre/cidja all refer to mass portions of the same
substances.  The countable units of some substances, like festi, may
themselves be mass units "lei festi", or in the case of water, lei
djacu.  If you go into a restaurant with 3 people and ask for 3 waters,
you don't want 3 molecules of water, nor can you specify the number of
molecules you want.

We need to enable scientific thought, but we must not force it.  By the
way, can someone accurately define a molecule/unit of thought?

Another problem is that very seldom are statements about macroscopic
things really statements about particular atoms or molecules.  I see
shades of your fuzzy logic ship here:  when I drink a glass of water, it
is quite safe to say that the molecules I ingest are not precisely the
molecules that were in the glass when I picked it up.  Liquid/gas
dynamics would have cause some exchange.  So how many molecules did I
actually drink?  Do we include the ones that had evaporated off the
surface and mixed with nearby air, but were inhaled when I swallowed?
But then, do we also include those molecules that were in the same air
before evaporation?

The atomic model may explain something about what water is, but it is
not necessarily true that statements about water are statements about
specific (or definite) countable molecules.  At least not if you really
want "accuracy".

>The atomic theory seems rather well accepted.  Shouldn't we build its
>contexts into the elementary structure of the language?

No.  Because not everything is necessarily atomic/quanta, and we cannot
clearly say we know what things are or are not.  The atomic theory is
merely another way of looking at things.

The quest for scientific precision almost destroyed Loglan when it first
went public.  Early articles in The Loglanist went on endlessly in great
analytical detail about color words having to be specified as additive,
subtractive, etc.  The places of "color" were specified as saturation
and hue - that sort of thing.  But when I talk about what color
something is, I haven't the vaguest idea how that color might be
described analytically - the analytic scientific concept is NOT the
human concept.

>Notice how person, human, and people all map to /prenu/ (Seems
>reasonable) This suggests that the cmavo determine whether we mean a
>group or an individual.  Notice that /ninmu/ & /nanmu/ explicitly in
>their definition acknowledge group or individual meanings!

This is one case where English does have words that clearly distinguish
the different senses.  The gismu English-definitions weren't trying to
come up with strained English forms that cover the individual and the
mass.

>One form of context could
>be inclusion of units (moles, grams, milliliters) as a selbri as part of
>the definition.  And I haven't even begun to talk about isotopes,
>contaminants, temperature, ambient & vapor pressure etc.  All this
>complexity is not necessary in every discourse, but IMHO it ought to be
>built into the structure of the predicate, available for use when the
>cmavo tease out *exactly* what flavor of djacu the speaker intends.

But if you avoid English or other metaphysical bias, every single
predicate needs to be describable in units or quanta.  So if a units
place is necessary on any given predicate, you can specify it with a BAI
place built from ci'u or tela'i.  Or you can make a tanru/lujvo using
selci/dirgo etc., as appropriate.

>>But lei/loi already makes a different distinction. In fact, you can
>>use {lei} as you propose, because {lei djacu} is "the mass of those
>>I have in mind that I'm calling quantities of water", and you can think
>>of the quantities as being molecules. The question is whether anyone
>>who listens will understand your intent. If the context makes it clear
>>that you are talking about molecules then there's no problem.
>
>This is very interesting to me.  Jorge has clarified what the problem
>is.  (As he often does.)  Jorge & I apparently interpret the meaning of
>/djacu/ in the dictionary differently.  Now the question is (shades of
>interprative literature class 301!)  "What did the author (lojbab?)
>intend?"

No.  I think the problem here is that you were talking about lei and loi
which are both cmavo used for masses, but implying that you thought of
lei in terms of molecules.  I think that since you were talking about
masses/individual molecules, your probably meant lei/le or loi/lo as the
distinction.

Jorge:
>> From the discussion about trobriand islander masses I inferred (perhaps
>> mistakenly) that the particular mass designation of /ractu/ depends on the
>> preceding cmavo:
>> le la lo, lei, loi etc.
>
>The discussion about Trobrianders, in my opinion, does not apply to
>{loi}.  It may apply to {lo'e}, I'm not sure.  Unfortunately, almost all
>explanations of {loi} are made in English with very little Lojban
>examples.  I very, very much doubt that the Lojbanistani would say {mi
>viska loi ractu} upon seeing a single rabbit, instead of {mi viska lo
>ractu}.

On the other hand, there was discussion about calling for a taxi a while
back (before the "any" discussion got started, though it is probably
related).  And both we and JCB's group independently decided the same
thing - that this was "loi {taksi}".

>> This suggests that the cmavo determine whether we mean a group or an
>> individual.
>
>Right, but based on the meaning of the gismu.  For a given {broda}:  {le
>broda} and {lo broda} refer to individuals that fit the x1 of
>relationship {broda}; {lei broda} and {loi broda} refer to entities
>whose components (but not necessarily themselves) fit the x1 of {broda}.

It is also possible that (a portion of) the mass fits the x1 and yet
none the components do individally.

>> I think the (minor) inconsistencies in the dictionary suggest you are
>> putting far too much weight on the literal english used to translate the
>> lojban predicate /djacu/ into English.
>
>Believe me, I tend to do the opposite.  I disagree with lots of things
>in the dictionary, but in this case I agree with it that {pa djacu} is
>not one molecule of water.

Well, it COULD be a molecule of water, but only in some pretty
constrained circumstances.

lojbab