[Martin Taylor 931102 16:30]
(Bill Powers 931102.1230)
[LOGICAL OR]
/ | | | | | \
PERCEPTIONS THAT BELONG TO CATEGORY
Replace [Logical Or] with [Logical Function], and it works
better, doesn't it?
I hesitate to do that, for reasons you mentioned yesterday: some
things that look (verbally) like categories may actually be
logical combinations of categories. When that's the case, the
logic level gets into it and we're not talking about just the
category level. The logical OR is a simple fixed type of function
that would serve to create simple ("pure") categories, with or
without sharp edges. If we allow general logical functions, I
think we collapse too much into the category level.
Perhaps it slipped by, but I did split the category level into two: an
interface level at which natural contrasts create the categories, and
a constructed category level that I also labelled "logical expressions"
or something like that. I differentiated the perceptions into "natural
kinds" and "constructed categories," which may be unfortunate choices
of name, but which do, I think, represent a real difference in the kind
of thing perceived.
The interface level is just that: an interface between analogic and logic.
Your idea of using copies of the analogue perceptual signals in a flip-flop
arrangement makes the level so simple that Rick's question can be answered
explicitly:
What is the form of the PIF in the category level? A mutually inhibitory
connection with the PIF of a contrasting category. A category level
PIF needs no more than this. Otherwise it is a straight-through feed.
The important point about interface categories is that they lead to logic.
Whether or not you buy the idea of contrast as process, the hysteresis
boundary is, I think, an important aspect of the category level. This is
true whether the logic levels are off to the side of a fully analogue
hierarchy or above it. The notion of "categories" without sharp edges
is to me an oxymoron. The position of the edge in the underlying perceptual
space may shift moment by moment, but it is always abrupt, so long as a
smooth change in the analogue perception can lead to a sudden change in
category.
Another thought is that at the logical level, we seem able to
construct logical propositions (like "(right-angled AND six-inch
handle) OR (flexible shaft)" in real time, rather than having to
learn a new logical form permanently in order to use it.
Sure. I don't have a good handle on how this reshuffling works,
either. It applies to the logical levels generally, doesn't it?
Or maybe it applies only to the level of logical expressions.
Either way, it smacks of imagination and reorganization on a very
different time scale than main hierarchy reorganization (as you
said of the example of lending the guest the umbrella). Even in
the umbrella example, if reorganization is happening, which I questioned,
the reorganization is in the conditional expression "If P," rather
than in the program itself.
And what about temporal logical functions (i.e. "sequence" perceptions)?
Should they be part of the "logical expression" system or should they be
fed by it? "After" and "not after" (as opposed to "before") sound as if
they interact in a normal logical-expression way. Maybe one should
think of the "logical expression" level as a substitute for the sequence
level, allowing as constructed categories such forms as
"A before ((B or C) at the same time as D)".
Interesting.
···
===================
Rick Marken (931102.0800) --
There must be some way to refer to people who use empirically-
discovered relationships as the basis for treating problems
without calling them witch doctors.
Is "witch doctor" a pejorative term still? I thought they were
gaining respect for the success of a lot of their cures. Much
medical success is largely due to the patient's belief in the
power of the doctor, and some folk medicines work, as we are
beginning to learn. Does it matter whether the doctor dresses up
in feathers and beads and dances in a smoky hut, or dresses in a
white coat and speaks authoritatively in an antiseptic office? A
lot of the good he/she does is in the authority. The patient does
the curing. It's all a matter of perceptual control, isn't it?
============
Incidentally, a week or two ago I saw a headline on an article about
health from, I think, the UN. The authors claimed that a large part
of the health problems sometimes attributed to poverty and the like
should actually be attributed to the inability of people to control
their own lives, whether they be junior bureaucrats or terrorized
peasants. It was the lack of control that mattered. (No, I can't say
whether this is a "fact." I suppose it isn't. But it is interesting
that a major report says it is.)
Martin