categorically speaking

[bnhpct]

[From Bruce Nevin (980128.2239)]

Bill Powers (980128.0835 MST) --

[was: Re: The MOL and Therapy]

If one habitually sees the world in terms of categories, these categories
just seem to exist in all the lower-level perceptions, and even outside in
the environment. There's a chair, there's a person, there's a sensation, and
so on. We categorize automatically without sensing this as something we're
doing. To become aware _of categorizing_, we must somehow move to a higher
level, so that categorizing becomes an object of attention -- not the
categories themselves, but the process of creating them. Only then can we
see that categories impose a lumpy structure on a world that is actually
continuously variable. Things that are alike categorically are not alike at
any lower level.

I thought that the "subjective reality" levels (suggested to be programs,
principles, system concepts) lived in a universe of categories ("symbols").
Is there reason to believe that they can traffic directly in "physical
reality" perceptions (suggested to be relationships, sequences and events,
transitions, configurations, sensations, and intensities)?

we must somehow move to a higher
level, so that categorizing becomes an object of attention -- not the
categories themselves, but the process of creating them.

There are two processes called "categorizing":

  Creating a nonce-category
  Perceiving something as an instance of an established category

You have slipped from one to the other here. Commonly, we don't create a
nonce-category (the objects on that end of that shelf), we perceive
something as an instance of an already established category, typically with
words associated. Your examples are all of this sort: There's a chair,
there's a person, there's a sensation, and so on.

The process of *creating* a category is something else. I'm not convinced
of it; it may be an appearance, an effect of using language. Even if I put
all the things on that end of the shelf in a box, or do something else with
them, that doesn't mean that I created a category comprising those objects.
The objects may disturb my control of a perception of an empty space on
that shelf that matters to me for some reason. Their seeming membership in
a nonce-category is an effect of their disturbing my control of that
perception. The nonce-ness of the category reflects the immediacy of that
contingency.

And I can't think of any pre-established and enduring (non-nonce-)
categories that are not straightforwardly named. This may just be failure
of imagination or memory on my part -- maybe I need to revisit the
anthropologists and sociologists; but as I said at the beginning of the
preceding paragraph, what they *describe*, like rings and arcs in crowds,
may be side effects, not controlled perceptions of categories.

For (and finally), merely describing a congeries of things in a unitary way
forges of them a nonce-category, or appears to. Are there any examples of
nonce-categories that are not called into a semblance of categoryhood by
the act of describing them, perhaps supported by their together disturbing
control of otherwise unrelated perceptions, like the clutter on that shelf?
This, it seems to me, is not the stuff of a category level!

Our dependence on language obscures the process of categorizing (both
senses) for the very reasons you say. Even if one does not habitually see
the world in terms of categories, one is doing so at the time of using
words to describe the process. And even when one has indeed "gone
elsewhere" than the universe of categories, it may not appear so when one
reports back using language.

How-to directions can be more convincing, if someone first follows them
successfully. But the success must be made known, which brings us back to
reporting experience with language.

Perhaps people who appear to be stuck in a category level are so because of
infatuation with the ramifications of verbal and logical constructions, not
noticing that the limb they are on outreaches nonverbal perception. Such
constructions seem like models, and are even called models, but they can't
be tested as models, only checked for logical properties like consistency
and completeness. It's old fashioned, but I do believe that operational
verification is important.

  Bruce Nevin

[Bill_Powers1]

[From Bill Powers (980129.1458 MST)]

Bruce Nevin (980128.2239)--

I thought that the "subjective reality" levels (suggested to be programs,
principles, system concepts) lived in a universe of categories ("symbols").
Is there reason to believe that they can traffic directly in "physical
reality" perceptions (suggested to be relationships, sequences and events,
transitions, configurations, sensations, and intensities)?

No, I don't think they can. They traffic in symbols such as
"relationships," "events", "transitions", and so on, which are names
assigned to perceptions of lower levels. We have to deal in names to
communicate with each other, unless we're in a position to directly
demonstrate what we mean. For example, if we were in the same room, I
could say "This is called 'red'" and show you something which produces the
sensation I call red. Or I could move my hand in a circle and say "This is
what I mean by 'continuous transition.'" I could hold my hands level with
each other and say "This is what I mean by 'hands level with each other.'"

Where this gets tricky is in seeing that categories themselves are
different from the names of categories. I show you three objects, and say
"This is what I mean by the category 'three.'" The word "three" is only the
name of the category; it's a vocal or written symbol used to refer to the
nonverbal sense of categoriness, which is just a neural signal. What makes
the signal a category signal is not anything about the signal; it's the
input function that generates this signal when any of a set of lower-level
perceptual signals is present (even though they represent very different
things). So a horse and a car and an airplane all generate the same signal,
at the category level, giving us the sense that they are all alike in some
way. We can explain why we categorize them together, by saying they are all
means of transportation. But when we perceive them, the experience is that
they all seem to have a common character (if we happen to be using that
input function). Similarly, all sets of three things seem to have a common
character, namely their threeness.

I'm sure anyone can poke holes in this picture. It's just my way of trying
to convey the difference between perceptions as direct experiences and
_talking about_ perceptions, which is a different kind of experience.

There are two processes called "categorizing":

       Creating a nonce-category
       Perceiving something as an instance of an established category

You have slipped from one to the other here. Commonly, we don't create a
nonce-category (the objects on that end of that shelf), we perceive
something as an instance of an already established category, typically with
words associated. Your examples are all of this sort: There's a chair,
there's a person, there's a sensation, and so on.

That's an interesting point. I think there is more to what I call the
category level than can be contained in a single level. I also worry about
categories of categories, or in general the sort of fine structure that we
find in taxonomies. My concepts of the higher levels are painted with a
pretty broad brush.

The process of *creating* a category is something else. I'm not convinced
of it; it may be an appearance, an effect of using language. Even if I put
all the things on that end of the shelf in a box, or do something else with
them, that doesn't mean that I created a category comprising those objects.

Categories are subjective, so why not?

For (and finally), merely describing a congeries of things in a unitary way
forges of them a nonce-category, or appears to. Are there any examples of
nonce-categories that are not called into a semblance of categoryhood by
the act of describing them, perhaps supported by their together disturbing
control of otherwise unrelated perceptions, like the clutter on that shelf?
This, it seems to me, is not the stuff of a category level!

I don't have any convincing answers to such questions. One idea strikes me
-- in a "nonce-category," it doesn't matter what the individual items are
-- I could substitute different objects in the same places, and they'd
still be "the things on the other end of the shelf." So the category-ness
probably doesn't extend to the objects in the same way as "the screwdrivers
on the other end of the shelf." "On the other end of the shelf" is a
relationship, or the name of one, which requires an element to be in the
named place. So maybe this isn't really an example of categorizing. I don't
know, and I can't think of any way to know, for now.

Our dependence on language obscures the process of categorizing (both
senses) for the very reasons you say. Even if one does not habitually see
the world in terms of categories, one is doing so at the time of using
words to describe the process. And even when one has indeed "gone
elsewhere" than the universe of categories, it may not appear so when one
reports back using language.

Nail on the head. Exploring things like categories is something you pretty
much have to do alone, without the verbalizations. It's like the saying,
"Whatever you say the Tao is, it is not."

Perhaps people who appear to be stuck in a category level are so because of
infatuation with the ramifications of verbal and logical constructions, not
noticing that the limb they are on outreaches nonverbal perception. Such
constructions seem like models, and are even called models, but they can't
be tested as models, only checked for logical properties like consistency
and completeness. It's old fashioned, but I do believe that operational
verification is important.

Well said. I think that being stuck in a category level comes across as a
certain kind of abstractness; the nouns are all plural or collective.
Language is not very good at describing a single lower-level experience.

Best,

Bill P.