Re: Fuzzy Fallacies

["Steven M. Belknap" <sbelknap@UIC.EDU>]

la pitr cusku di'e
>There seems to be a continuing misunderstanding, which perhaps I can
>correct here.

Perhaps not. The professorial tone is amusing, of course. I hope your
students don't laugh at you as hard as they laugh at me when I get all
pompous and speak ex cathedra. (Which unfortunately I still do sometimes.)
:-)

>Tall is not defined with respect to "not-tall".  It is defined against the
>>speaker's criteria for tallness.  This criteria are essentially the
>"ideal" that >you refer to.

Defined by who? The Peter Schuerman Dictionary of Idiosyncratic American
English? Well according to The Steven Belknap Dictionary of Idiosyncratic
American English, the definition of tall is not limited to subjective
impressions. Hmmm. Seems we have an impasse here. No doubt you consider the
dictionary you walk around with in your head to be very fine indeed,
perhaps because it contains definitions which conveniently mututate so as
to win arguments with your learned colleagues. Of course I consider my own
dictionary to be superior, perhaps for much the same reason. So we're
stuck.

But, wait, what's this? External authoritative sources! Compiled by
experts! With attention to etymology, common usage, multiple meanings, and
historical context! Hurrah!  I have more than 30 such dictionaries at hand.
The definitions of tall do not differ much among these sources.

As a comprehensive source, the Oxford English Dictionary is hands down the
best choice. (I have a CD-ROM version, which I highly recommend and also
get get at the OED online through the University of Illinois.) For American
English, I think the American Heritage Dictionary is an excellent choice.
(I also happen to have a copy on my Mac laptop.)

The Oxford English Dictionary entry for <tall> is quite lengthy, so I won't
quote it all; if anyone really wants the whole thing, I'll be happy to
email to them. The appropraite subentry for <tall>in the OED would seem to
be section II 6 a:


Of a person: High of stature; of more than average height. Usually
appreciative. Also of animals, as a giraffe, stag, or the like. (Cf.
elegant a. 2 b =3D tall of stature.)


The American Heritage Dictionary Defines <tall> & <ordinary> as follows:

tall (t=F4l) adj. tall0er, tall0est.
1.a. Having greater than ordinary height: a tall woman. b. Having
considerable height, especially in relation to width; lofty: tall trees.
See Synonyms at high.
2. Having a specified height: a plant three feet tall.
3. Informal. Fanciful or exaggerated; boastful: tall tales of heroic exploit=
s.
4. Impressively great or difficult: a tall order to fill.
5. Archaic. Excellent; fine.

or0di0nar0y (=F4r2dn-Rr1T) adj.
1. Commonly encountered; usual. See Synonyms at common.
2.a. Of no exceptional ability, degree, or quality; average. b. Of inferior
quality; second-rate.
3. Having immediate rather than delegated jurisdiction, as a judge.
4. Mathematics. Designating a differential equation containing no more than
two variables and derivatives of one with respect to the other.

So the OED suggests using an external standard (comparison to average
height) while AHD suggests using common experience as a guide. (greater
than commonly encountered height) As I've repeatedly corrected Peter,
dictionaries are neutral as to the appropriate choice of logic (fuzzy or
Aristotlean), scale (nominal, ordinal, interval or ratio), and are
purposely vague as to criteria. My science dictionary describes the
analogous term <length> in terms of procedures by which it can be measured.

>Note that different cultures
>have different criteria, and even different people have different
>criteria.  This may be very annoying, but it's the reality.  Fortunately,
>most people within a culture have *similar* criteria for most words we
>use... when our criteria are very different than the common criteria, it
>makes us use words incorrectly; other people correct our usage, and we
>adjust our criteria accordingly.  But the process does not result in an
>exact duplication of criteria in all people.  We just tend to think
>similarly.
>

Two cultures I exist in are the cultures of medicine and science. Citizens
of these two cultures actually do manage to do a fair job of calibrating
their separate criteria, logics, and scales. In fact, a fair amount of
discourse in these cultures is directed precisely at agreeing on common
definitions, operating procedures, criteria,logics, and scales. As Robert
Chassel points out in regard to scales, much progress in science results
when scientists move to a more sophisticated scale. I would respectfully
suggest that attention to what exactly we mean when we use a word might be
of some use in other cultures as well.

>> In that sense Steve is right.  For any person A that everyone agrees IS
>> "tall", we can envision the possibility of someone B who is "more" tall.
>> It is not clear whether we would mark the statement "A is tall" as being
>> less than perfectly true MERELY because B exists.
>
>The statement "A is tall" is a semantic shorthand for "A seems tall to me
>based on the criteria I use to judge tallness."  If B is taller than A,
>the statement "B is tall" does not in any way contradict or support "A is
>tall" because it also means "B seems tall to me based on the criteria I
>use to judge tallness."  This is true even if A is taller than B.
>
>> It is possible to define tallness as compared to something smaller.  But
>> I dare say that unless there is agreement as to what the reference is,
>> then it will be hard to get any consistency in values.
>
>Which is why it is a mistake to use "tall" as a way of communicating
>objective information about height.  That is why systems of measurement
>were created in the first place!  You are right, that there must be
>agreement as to what the reference is.  That's why we create references,
>such as sticks with incremental marks on them and standard weights for
>callibrating scales.
>
I again challenge Peter's insistence that he is the one true source of
information about the proper use of language. I do not see any hint of a
consensus among speakers of English supporting Peter's assertion that
<tall> is purely subjective, although Peter certainly believes that this is
how he uses language. As I've written before there *is* published work in
this area supporting the view that objective data can be extracted from
utterances which are not explicit on matters numerical. Apparently the
human comedy is a fairly consistent experience.

I would be most interested in Peter's comments regarding the Guttman
scales. Does Peter think there might be value in making scales explicit,
whether speaking English or lojban? This would seem to address his
concerns. If one wishes to be understood as being totally subjective, then
simply avoid using a scale in one's utterances.

Of course, people tend to use short-hand, so a given speaker-listener pair
may drop such scale references if both understand that a given scale,
logic, and criteria are implied and inferred in context. Perhaps wearing
all-black, hanging out in Berkeley coffee-houses, and being painfully thin
could be a society-wide red flag that the speaker intends all his
utterances to be totally subjective.:)

>"A is tall" is not useful as accurate communication unless the criteria of
>the speaker are known.  For the *same* reason, "A is 5-tall" is *also*
>not useful, unless the criteria of the speaker are known.
>
>Of course, once the speaker and listener know what the criteria for these
>descriptive words are, there is no longer a problem.
>So, even "5-tall" can be meaningful if the listener knows that it means
>'between the heights of 5'8" and 5'11"' (for example).

We agree.

>
>The only problem with using numbers as descriptors in this way is that
>they cease to function as numbers.  This in turn is not a problem, as
>long as it is recognized that the numbers no longer have the same
>semantic meaning.  When you do mathematical operations on numbers, you
>get meaningful results.

Please see my discussion of Guttman scales. I think you may agree that they
are useful. Actually, this discussion and the manner posts I've gotten have
been quite helpful in clarifying my thinking as to how things numerical
ought be expressed in lojban. Those interested ought to read mex.tex on the
ftp site. There is a lot of work that needs to be done in this area. Dylan?
Any comments?

co'o mi'e. stivn.


Steven M. Belknap, M.D.
Assistant Professor of Clinical Pharmacology and Medicine
University of Illinois College of Medicine at Peoria

email: sbelknap@uic.edu
Voice: 309/671-3403
=46ax:   309/671-8413