affordance

CSGnet Archive CSG1998
1

[From Bruce Nevin (980312.1611)]

Some things that we believe are "out there" in our environment have been
called affordances by Mr. Gibson. He seems to have believed that
affordances inform us and direct our activities. Perhaps the concept is
useful nonetheless. Here's a start at a definition.

Affordance: A controllable perception by means of which one can control
another perception.

A horizontal surface is an affordance for setting things down, or getting
under, etc. An upright surface is an affordance for leaning something
against, or sheltering behind, etc.

A construction with a horizontal surface at the height of hands above the
ground can be a table. A horizontal surface at the height of the knees can
be a bench. In combination with an upright surface it can be a chair.

Furniture: Conventionalized affordances for living arrangments.

Tool: A conventionalized affordance for changing things like furniture.

The fit of a phillips screwdriver to the head of a phillips screw as means
for turning it is not merely conventional, it is standardized. A saw is not
standardized.

Universals of language: some physiological and acoustic characteristics of
the human vocal tract are affordances. In certain regions of the oral
cavity, differences of articulation with the tongue make little difference
in sound. These stable regions are separated by volatile regions in which
small differences in articulation make larger acoustic differences.
Speakers of languages use the stable regions as means for making
distinctions between words. For some languages, more of these regions are
used. Some are more favored than others, such that you find languages with
distinctions made in region A, and languages with distinctions made both in
region A and region B, but no languages with distinctions made in region B
but not in region A.

2

[From Bill Powers (980312.20o38 MST)]

Bruce Nevin (980312.1611)--

Some things that we believe are "out there" in our environment have been
called affordances by Mr. Gibson. He seems to have believed that
affordances inform us and direct our activities. Perhaps the concept is
useful nonetheless. Here's a start at a definition.

Affordance: A controllable perception by means of which one can control
another perception.

A horizontal surface is an affordance for setting things down, or getting
under, etc. An upright surface is an affordance for leaning something
against, or sheltering behind, etc.

This has the same disadvantage as defining purpose as the use to which
something can be put. In fact, we can use slanted surfaces to lean things
against, or big round objects like boulders, or anything else that will
serve. Neither the environment or its properties can tell us how to use
them. Since you can use almost anything for almost any purpose, it's futile
to try to identify properties of things that have any special relationships
to behavior.

What is it that affords the perception of verticalness or uprightness, or
leaning-against-ness, or hiding-ness, and so on? Gibson's whole scheme is
based on unconscious acceptance of some perceptions as being "just there,"
givens of the environment. In HPCT as little as possible is taken as given,
down the the level of intensity perceptions. I prefer to take as little for
granted as possible, which is why I have never been impressed by Gibson.

Best,

Bill P.

3

[From Bruce Nevin (980312.2341)]

Bill Powers (980312.20o38 MST)--

What is it that affords the perception of verticalness or uprightness, or
leaning-against-ness, or hiding-ness, and so on? Gibson's whole scheme is
based on unconscious acceptance of some perceptions as being "just there,"
givens of the environment. In HPCT as little as possible is taken as given,
down the the level of intensity perceptions. I prefer to take as little for
granted as possible, which is why I have never been impressed by Gibson.

I agree, as I said. What I'm trying to get at is that we do recognize
furniture and tools. A rock can be used as a hammer but that does not make
it a hammer, a boulder can be used for sitting or for shelter from the wind
but that does not make it a bench or a wall. There is something different
about artifacts, partly convention, partly design for fitness.

The physiological basis of linguistic universals is a bit harder case
because it is not a designed artifact.

The only way the term affordance could be used would be with a disavowal of
Gibson's attributions of intelligence to the environment. (That's what I
did.) To avoid provocation we could just talk about furniture, tools,
universals, etc. without calling them affordances. But this group is not
remarkable for avoiding provocation.

  Bruce Nevin

4

i.kurtzer (980313)

[From Bruce Nevin (980312.2341)]

Bill Powers (980312.20o38 MST)--

>What is it that affords the perception of verticalness or uprightness, or

>>leaning-against-ness, or hiding-ness, and so on? Gibson's whole scheme is
>>based on unconscious acceptance of some perceptions as being "just there,"
>>givens of the environment. In HPCT as little as possible is taken as given,
>>down the the level of intensity perceptions. I prefer to take as little for
>>granted as possible, which is why I have never been impressed by Gibson.

>I agree, as I said. What I'm trying to get at is that we do recognize
>furniture and tools.

some people.

A rock can be used as a hammer but that does not make

>it a hammer, a boulder can be used for sitting or for shelter from the wind

but that does not make it a bench or a wall.

is the above an empirically derived conclusion? if not, on what basis is this
made?

There is something different
about artifacts, partly convention, partly design for fitness.

same query as previous.

>The only way the term affordance could be used would be with a disavowal of
>Gibson's attributions of intelligence to the environment. (That's what I
>did.) To avoid provocation we could just talk about furniture, tools,
>universals, etc. without calling them affordances. But this group is not
>remarkable for avoiding provocation.

what is gained by calling them affordances?

i.

5

[FRom Bill Powers (980313.0201 MST)]

Bruce Nevin (980312.2341)--

I agree, as I said. What I'm trying to get at is that we do recognize
furniture and tools. A rock can be used as a hammer but that does not make
it a hammer, a boulder can be used for sitting or for shelter from the wind
but that does not make it a bench or a wall. There is something different
about artifacts, partly convention, partly design for fitness.

If we get into this more deeply, I'm afraid we'll end up in that same
disagreement about categories.

A rock and a hammer can be used for some similar purposes, but whether a
rock "is" a hammer depends on how we have agreed to form the respective
categories. If you shaped a rock so it had a flat pounding surface that
would not leave marks on smooth wood, and a claw that could be used for
removing nails, and a handle that was of an optimal length and shape, and a
weight that was neither too much nor too little, would it not be a hammer?
If you found a rock that had all these attributes, rather than shaping it
yourself, would it be any less a hammer?

The answers are "yes" or "no" depending on what you decide. If you decide
that no artifact is to be called by the same name as a naturally-occurring
object, it will be "no." Otherwise it will be "yes." The question is very
much like asking what is the proper pronunciation of a word, or the proper
form of a grammatical sentence. The answer is not a matter of fact, but of
definition and agreement.

The physiological basis of linguistic universals is a bit harder case
because it is not a designed artifact.

The only way the term affordance could be used would be with a disavowal of
Gibson's attributions of intelligence to the environment. (That's what I
did.) To avoid provocation we could just talk about furniture, tools,
universals, etc. without calling them affordances. But this group is not
remarkable for avoiding provocation.

Is "furniture" a universal? Is "universal" a universal? Or are these terms
merely the names of categories of words? If you don't believe, as I don't,
that there is anything more to words than a function as pointers to
experiences (which can be other words), then asking about "linguistic
universals" is pointless -- the properties implied by such a term probably
don't exist. Many words and ways of speaking are empty of experiential
meaning; we use them because others use them, but whatever meaning we think
we see in them is purely imaginary and arbitrary. We assume that everybody
else knows what they mean, but in fact nobody does because they have no
meaning except the other words to which they point. "Linguistic universal"
sounds to me as if it ought to have a meaning, but I have to confess that
for me, it doesn't have one.

Best,

Bill P.

6

[From Bruce Gregory (980313.0645 EST)]

Bill Powers (980312.20o38 MST)

What is it that affords the perception of verticalness or uprightness, or
leaning-against-ness, or hiding-ness, and so on? Gibson's whole scheme is
based on unconscious acceptance of some perceptions as being "just there,"
givens of the environment. In HPCT as little as possible is taken as given,
down the the level of intensity perceptions. I prefer to take as little for
granted as possible, which is why I have never been impressed by Gibson.

I think of an affordance as a possible perception (perceptual organization)
at the level of concept. The perceptual organization allows something to
show up as a tool for example.

Bruce

7

[From Bruce gregory (980313.1130 EST)]

Bill Powers (980313.0837 MST)]

Bruce Gregory (980313.0645 EST)--

>I think of an affordance as a possible perception (perceptual organization)
>at the level of concept. The perceptual organization allows something to
>show up as a tool for example.

Why isn't "tool" simply a category -- a "means of control?"

It is.

And is it the
environment that "affords" seeing something as a tool, or the perceptual
system?

The perceptual system.

What about the person who grabs a handy rock and uses it to
prop a door open? Does that make the rock a tool?

Yes. When you are looking for a tool, a rock can show up as a
tool.

Unconvinced,

Not trying to convince, only to explain.

Bruce

8

[From Bruce Nevin (980313.1150 EST)]

Bill Powers (980313.0837 MST)--
(responding to Bruce Gregory (980313.0645 EST))

Why isn't "tool" simply a category -- a "means of control?" And is it the
environment that "affords" seeing something as a tool, or the perceptual
system? What about the person who grabs a handy rock and uses it to prop a
door open? Does that make the rock a tool?

The things categorized are in the universe of perceptions. Their status in
the environment is not specified with any more (or less) particularity than
the usual relation of the perceptual hierarchy to the environment.

  Bruce Nevin

9

[Martin Taylor 980314 19:50]

I seem to have got to Friday messages now! Only 40 or so to go.

Bill Powers (980313.0837 MST)]

Bruce Gregory (980313.0645 EST)--

I think of an affordance as a possible perception (perceptual organization)
at the level of concept. The perceptual organization allows something to
show up as a tool for example.

Why isn't "tool" simply a category -- a "means of control?" And is it the
environment that "affords" seeing something as a tool, or the perceptual
system? What about the person who grabs a handy rock and uses it to prop a
door open? Does that make the rock a tool?

Surely an "affordance" is simply a perception by one control system of
one of many environmental feedback possibilities available to other
control systems?

It ties in with the notion of program-level control as working largely
through the imagination connection (and of being possibly the only level
for which this is true). The affordances of things perceived in the
environment are available to the program-level imagination loop. Whether
they "things perceived in the environment" act the way they were imagined
when put upon (to quote Fortinbras) is a separate issue. And whether a rock
is a "hammer" or a "doorstop" depends more on the environmental feedback
function it fulfils than of any "is" about it.

Martin

10

[From Bill Powers (980315.0402 MST)]

Martin Taylor 980314 19:50--

I think of an affordance as a possible perception (perceptual organization)
at the level of concept. The perceptual organization allows something to
show up as a tool for example.

Is there something missing from PCT that we need the term affordance to
describe? If you can translate the term so easily into PCT terms, why not
just use the PCT terms?

Your interpretation is actually a correction of Gibbons' usage. He assumed
that affordances were properties of the environment. You are saying they
are properties of perceptual organization (with which I would agree, if I
saw any usefulness in the term at all, which I don't).

Best,

Bill P.

11

[From Bruce Nevin (980314.1121)]

Bill Powers (980315.0402 MST)--

Martin Taylor 980314 19:50--

I think of an affordance as a possible perception (perceptual

organization)

at the level of concept. The perceptual organization allows something to
show up as a tool for example.

Is there something missing from PCT that we need the term affordance to
describe? If you can translate the term so easily into PCT terms, why not
just use the PCT terms?

Your interpretation is actually a correction of Gibbons' usage. He assumed
that affordances were properties of the environment. You are saying they
are properties of perceptual organization (with which I would agree, if I
saw any usefulness in the term at all, which I don't).

Gee willikers, guys. I'm sorry I brought it up.

[From Bruce Nevin (980312.1611)]

Some things that we believe are "out there" in our environment have been
called affordances by Mr. Gibson. He seems to have believed that
affordances inform us and direct our activities. Perhaps the concept is
useful nonetheless. Here's a start at a definition.

Affordance: A controllable perception by means of which one can control
another perception.

So let's just talk about controllable perceptions by means of which one can
control another perception. Or concept perceptions or perceptual
organizations at the level of concept. Let's not say these are what people
think of as furniture or tools or affordances.

Instead of talking about a screwdriver as a controllable perception that
can be used to control other perceptions, we only need to talk about the
many sequence and program perceptions in which the use of the screwdriver
is a step.

It almost looks like word association is a valid instance of S-R
psychology. "Behavior is the control of perception? Sure, I've always
believed that perception governs behavior." "Affordance? You must be
claiming that properties of the environment limit and direct our behavior."

Guess I'm feeling grumpy.

  Bruce Nevin

12

[From Bill Powers (980315.1256 MST)]

Bruce Nevin (980314.1121)--

Gee willikers, guys. I'm sorry I brought it up.

Good.

So let's just talk about controllable perceptions by means of which one can
control another perception.

So a vertical surface is a controllable perception that affords us
something to lean things against? How does controlling the perception of a
vertical surface help us lean things against it? If the surface is already
vertical, why does it need to be controlled? If it's not vertical, why do
you call it a "vertical surface?"

Let's not say these are what people
think of as furniture or tools or affordances.

Furniture or tools, fine. Affordances, not fine.

Instead of talking about a screwdriver as a controllable perception that
can be used to control other perceptions, we only need to talk about the
many sequence and program perceptions in which the use of the screwdriver
is a step.

A THING is not a controllable perception. You can't control a screwdriver
without mashing it into something else. You control _variable attributes_
of things: length, color, weight, position, orientation, price, dirtiness,
and so forth.

It almost looks like word association is a valid instance of S-R
psychology. "Behavior is the control of perception? Sure, I've always
believed that perception governs behavior." "Affordance? You must be
claiming that properties of the environment limit and direct our behavior."

Guess I'm feeling grumpy.

No, sloppy. Affordance is a sloppy term that invites loose word
associations. It's part of a completely different model of perception.

Best,

Bill P.

13

[From Bruce Gregory (980315.1652 EST)]

Bill Powers (980315.1256 MST)

Affordance is a sloppy term that invites loose word
associations. It's part of a completely different model of perception.

Henceforth if any person on CSGnet uses the term "affordance" LET HIM BE
ANATHEMA! (I've always wanted to have an occasion to say that...)

Bruce

14

[Martin Taylor 980318 17:40

Lots of messages backed up again :frowning:

And this one was apparently not sent out by my mailer:-((

Bill Powers (980315.0402 MST)]

Martin Taylor 980314 19:50--

I think of an affordance as a possible perception (perceptual organization)
at the level of concept. The perceptual organization allows something to
show up as a tool for example.

Is there something missing from PCT that we need the term affordance to
describe? If you can translate the term so easily into PCT terms, why not
just use the PCT terms?

I don't think anything is missing from PCT that needs the term "affordance."
But people do use it, and it has a simple interpretation in PCT. So why
make a new PCT-specific term?

Incidentally, what _is_ the PCT-approved term to describe all the different
environmental feedback functions of which a particular perceived part of
the environment can form part? I don't think it has come up in my time
on CSGnet.

Martin

15

[From Bill Powers (980324.1206 MST)]

Martin Taylor 980318 17:40--

And this one was apparently not sent out by my mailer:-((

Yes, it was.

Best,

Bill P.