Re: Dvorak (& Lojban)

[Logical Language Group <lojbab@ACCESS.DIGEX.NET>]

>If what we're trying to translate here is the meaning "of greater
>magnitude in one physical dimension" (colloquially "longer"), then I
>have to go with {tremau}, not {clanymau}.

I disagree, since "tremau" has an implicit expansion based on zmadu with
an x3 of le ni/ka mitre, and we have not come to much consensus as to
what leni mitre would mean.  I agree that anything that se ckaji leka mitre
has length - it is characterized by being measurable in metres.  But I do not
think that something longer is zmadu fi le ka mitre (more in measureability
in metres), but rather is more in le se mitre.

>{clanymau} also maps to
>the English word "longer", but with the a different meaning--which is
>not very lojbanic: greater in the degree of being subjectively "long"
>in a certain context

Why subjectively?  If you use an objective standard of measure, the result is
objective.

>which has the same lack of precision as {melbi}

melbi?  How did "melbi" get into this?  Are you claiming that "melbi" is
unLojbanic because we cannot agree on an objective te ni  melbi?

>The "more measurable in meters" concept you list above is a meaning
>that makes sense, but isn't very useful.

Whether it is useful or not, that is what "tremau" would mean to me using the
"obvious" place structure expansion used for most other -mau lujvo.

>It makes far more sense to
>me to use /all/ the measurement words like {mitre} in a consistent
>and simple way, to have meanings not specifically tied to their units
>when abstracted.

I would agree with this, but ...
>After all, isn't that what the cognitive process of
>"abstraction" is, to omit specific referents and measurements?

Yes, but the Lojban abstractors are not(in my mind) defined differently for
different kinds of brivla.  I start from the basic assumption that all
brivla are semantically alike in some basic sense of stating a relationship.
ka brivla is then a reference to the nature or applicability of that
relationship.  I am not entitrely sure that for all brivla, one can talk
about zmadu fi lo ka brivla.  In fact I am pretty sure that zmadu doesn't
work very often with ka, except when there is an implied measurement in the
statement of a bridi based on brivla as to the relationship.  With mitre,
the quantification is not on the relationship "mitre" but in the x2 of
mitre.

ni brivla implicitly assumes that there is a (possibly objective) scale by
which one could measure brivlaness, and indeed requires that scale to be
present at least implicitly in the x3 of the abstraction.

jei brivla is similar expect that the scale is explicitly 0 to 1 based on a
degree of truth of the brivla relationship.

So your statement as to the cognitive process of abstraction does not seem to
tell us anything useful about Lojban abstractors, which reach some kind of
common endpoint, which may have new measurements that are not present in the
original brivla.

>Secondly, let me retract part of my earlier statement about {clani}.
>{le na clani} doesn't actually fail to have {ni clani} (in the same
>context, as Jorge correctly emphasizes), it just has a {ni clani} of
>zero, which is as valid a quantity as any other, while it /does/
>fail to have {ka clani}.  Here I depart from Jorge a bit: something
>that is {na blabi} has no {ka blabi}, but it /does/ have {ni blabi}
>--More--
>of zero.  That helps the case where you're comparing two things.

le na blabi might have a "jei blabi" of zero.  Then again, it might simply
have a jei blabi which is less than one.  In English we generally say that
something is false if it is even a little bit false.  Similarly, something is
not long if its length is less than some minimum amount, which does not mean
that it has no length, or that it has a length of zero.
>Electrons and baseballs have {ka mitre}.

I'll leave it to someone else to discuss the length of an electron %^)

>Things like,
>say, the company I work for, do not (though the building we're in
>does).
Agreed.  In which case  le kagni cu na se ckaji le ka clani
and na'i (not "no") is the answer to "xo ni le kagni cu clani"
In other words, you cannot decide on a measurement standard to measure
the length of the company.

>My company may not have the property of length, but it might
>have the property of duration (having been founded on a certain date)

Uh oh.  Then in relativistic space, your company DOES have length in the time
dimension.  You just don't know what that length is, since you don't know
if/when the end-in-time of the company exists.

>Another meaning of "length" is the specific amount of space a thing
>takes up--{ni mitre}.
We aren't agreed as to what ni mitre is, but I would not think that it has
anything to do with amount of space taken up.  ni mitre would be some
scale for measuring mitre-ness, which is measurability in metres to some
 specific value.  Hmm.  If an electron can be said to have length, than it is
 safe
to say that its ni mitre would be less than ni mitre for some object less
affected by its quantum nature.  Schroedinger's cat has a greater ni mitre
than his electron %^).

>The only distinction I can see between {se
>mitre} and {ni mitre} is that {se mitre} seems to imply a specific
>referent, while {ni mitre} does not.  One can think of "3 meters"
>as {ni mitre li ci} without any specific 3-meter-long thing involved.

Ah, when you put a specific value into the x2 you do get somewhere.
"ni mitre li ci" is the degree to which some object is measureable in metres
to be "3", which hopefully is greater for some object that is 3 metres long
than it is for something 2 metres long, but it is also greater than for
something which is 4 metres long.  Thus "leni mitre li ci" is greater for
a meterstick than it is for an electron or a (typical) locomotive.

lojbab
----
lojbab                                                lojbab@access.digex.net
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                        703-385-0273
Artificial language Loglan/Lojban: ftp.access.digex.net /pub/access/lojbab
    or see Lojban WWW Server: href="http://xiron.pc.helsinki.fi/lojban/";
    Order _The Complete Lojban Language_ - see our Web pages or ask me.