Re: re loi smani

[bob@GNU.AI.MIT.EDU]

First a discussion of the kind of categorization done by {loi}, then
use of numbers, then a prediction of Lojban style.

    loi  part of the mass of those which really are

{loi} is the Lojban derivative of JCB's `lo', which he based on his
understanding (or possibly, misunderstanding) of a grammatical usage
of Trobriand Islanders. He attempted to describe `lo' many years ago in
`The Creatures of LO'.

In that article, JCB talked of an islander coming upon a monkey and
perceiving it as a manifestation of `monkey'.  In English one might
say: `I see Mr. Monkey.'

A Platonist might consider {loi} to refer to the shadow that we
perceive of an ideal Platonic Form, except that the Form itself may
not exist.

An object oriented computer programmer might consider {loi} to refer
to the instance of a class, except that the class itself may not
exist.

Certainly, {loi mlatu} is part of the mass of those which really are
cats; {loi matne} is part of the mass of that which really is butter.

In each case, {loi} refers to a manifestation of something.

JCB often used examples in which the entity manifested is an
individual of a biological genus:

    .i mi ca ca'a viska loi mlatu
    I now, actually see a cat.
    I now, actually see a manifestation of Cat.
    I now, actually see a part of the mass of all cats.

    .i viska loi cmani
    I see Mr. Monkey.

(That in context, the expressions may or may not refer to a particular
entity specific in the mind of the speaker and definite in the mind of
the listener; I am skipping past this issue.)

All parts or manifestations of the mass must be fairly similar in some
important way.  Thus, it makes sense to speak of {loi mlatu}, but it
makes less sense, nowadays, to speak of {loi jubme} since tables vary
so much.

However, you could speak of {loi jubme}, especially if your context
consists mostly of four legged tables for human use.  (Six of the
seven entities which `really are' tables in the room in which I am now
sitting fit this context.  Also there are two more entities that I
often `designate as' tables although I also speak of them as low
shelves.)

Similarly, you can speak of a part of the mass of those which really
are boxes, {loi tanxe}.

The default quantification of {loi} is {pisu'o}, `part of'.  As I
remember, most of JCB's examples suggested `one' as the practical size
of the part.  This is not inconsistent with {pisu'o}.

In addition to speaking of individuals of a biological genus, {loi} is
used to refer to entities that in English require mass nouns, like
butter and water.

    .i mi viska loi matne
    I see some butter.
    I see a manifestation of Butter.


Incidentally, {loi matne} is *not* a mass consisting of *all* butter
as someone said in a recent posting; nor is {loi tanxe} a mass
consisting of all boxes.  Both are parts of the mass of all (as
defined by the current universe of the discourse).

Now let's return to Mr. Cat:

    .i mi viska loi mlatu

Suppose I see another cat!?  I say again,

    .i bi'u mi viska loi mlatu
    [New information] I see part of the mass of those which really are cats.

Now I have seen two manifestations of Mr. Cat.

Surely, it makes sense to say:

    .i mi viska re loi mlatu
    I see two manifestations of Mr. Cat.
    I see two parts of the mass of all cats.

Similarly, I see some water (in a pond, out the window):

    .i mi viska loi djacu

Then I see the second pond, lower down the valley, behind the first

    .i ji'a mi viska loi djacu
     Additionally, I see some water.

Again, it makes sense to me that I can say:

    .i mi viska re loi djacu
    I see two manifestations of Water.

Now, for a prediction of Lojban style.

A while back, I predicted that fluent Lojban speakers will use `lo'
frequently.  Now I think that they will also use `loi' frequently.
However, I suspect that two kinds of entity will absorb most uses:

  * Entities like water and butter that are considered mass nouns in
    English.

  * Entities like cat and monkey in which all are members of a
    biological genus.  Entities like this do not differ much from each
    other (except in ways not relevant to biological classification,
    such as age, sex, or whether alive).

Entities like buildings, which vary greatly from one another will not
be categorized using {loi} except in exceptional cases, as when an
explorer comes upon a vast, unlikely building, and wishes to emphasize
that what he sees really does belong to the genus `building':

    .i .ua .ue .u'e mi ca ca'a zgana loi dinju
    Discovery! Surprise! Wonder! I behold a manifestation of Mr. Building.
    ... a part of the mass of those which really are buildings

    Robert J. Chassell               bob@grackle.stockbridge.ma.us
    25 Rattlesnake Mountain Road     bob@gnu.ai.mit.edu
    Stockbridge, MA 01262-0693 USA   (413) 298-4725