fairness and meaning ACT

CSGnet Archive CSG2003
1

Quoting Bill Powers <powers_w@EARTHLINK.NET>:

[From Bill Powers (2003.10.08.1415 MDT)]

Any practicable solution would require not a

meaning-detector, but a meaning-indicator. You'd have to analyse the sounds
or marks to recognize patterns, and then you'd have to have a store of
possible meanings which could be indicated after any pattern had been
recognized.

While it may not fit all of the demands you suggest, does not Affect Control
Theory endevour to do as much. I am not saying it does it, but it does make a
good prtactical attempt. It may be at too basic a level for where you guys are
talking but thought I'D bring it up.

The paterns are on the three EPA dimensions of affective meaning. Then the
store relates to how these patterns are understood or meaning interpreted,
particularly in relation to controlled perception in an event.

The patterns of meaning for a word are then "controlled", we control the
meaning we have for a word.

Cheers Rohan

The meanings would have to be in the system already, before the

ยทยทยท

meaning-indicator could select the correct one that corresponds to the
detected symbol. And the meanings would be of a nature quite different from
the physical nature of a word.

I haven't even raised the question of what form these stored meanings would
take.

The point is that meaning doesn't arise in the manner of a response to a
stimulus, but through a very active process of perception and synthesis.
Turning words into meanings, or meanings into words, is a lot of work. You
have to have a device with a lot of functions in it, and without that
device, you might be able to hear or see words but that's all they would
be: sounds and marks. We won't be able to understand meaning until we can
say what this device would have to do, in enough detail to allow building
one.

I don't think we're there yet.

Best,

Bill P.

-------------------------------------------------
This mail sent through IMP: www-mail.usyd.edu.au

2

[From Bill Powers (2003.10.09.0820 MDT)]

Rohan Lulham *2003.10.08) --

> [From Bill Powers (2003.10.08.1415 MDT)]

> Any practicable solution would require not a
> meaning-detector, but a meaning-indicator. You'd have to analyse the sounds
> or marks to recognize patterns, and then you'd have to have a store of
> possible meanings which could be indicated after any pattern had been
> recognized.

While it may not fit all of the demands you suggest, does not Affect Control
Theory endevour to do as much. I am not saying it does it, but it does make a
good prtactical attempt. It may be at too basic a level.

You're going to have to explain what affect (emotion?) has to do with word
meanings. As you undoubtedly know, my theory of emotion makes affect simply
a normal part of every control process, to one degree or another. It's not
a separate thing.

The paterns are on the three EPA dimensions of affective meaning. Then the
store relates to how these patterns are understood or meaning interpreted,
particularly in relation to controlled perception in an event.

I've forgotten what those dimensions are -- if I remember right, it seems
to me that they're just classification names with no particular theoretical
significance in themselves, like the axes in a factor analysis. The
correlations in the kinds of studies involved aren't high enough to excite
my interest.

More charitably, any affect involved might be the meaning of a word -- for
example, "fear" is a word that I, at least, interpret as a combination of
sensations in my body and the goal of getting away from, or pushing away,
whatever it is I fear. But that's not what I was talking about, exactly.

The patterns of meaning for a word are then "controlled", we control the
meaning we have for a word.

I wasn't referring to patterns of meaning, but patterns of marks or sounds.
We have to recognize the patterns of marks that make up words (WORD, for
example), before we can even begin to enquire about any meanings they may
indicate.

I don't think we spend much time "controlling the meanings we have for a
word." We establish the experiences that we will be reminded of by a word,
and from then on pretty much use the same word to indicate the same
experiences (though as Bruce N. has pointed out, words have to do multiple
duty, often indicating different experiences in different contexts, as for
the word "chicken").

The control process I have been referring to goes like this. You wish to
communicate an experience to someone else, say the experience of seeing a
lion loose in your front yard and desiring very strongly that it or you be
elsewhere. You start with the image of the lion, the image of your front
yard, other images, the visual relationships among the images, and the set
of bodily sensations and cognitive goals you experienced at the time. These
are all experienced without words. You then search for words that, when you
hear them, remind you of these various experiences and their relationships
to each other through time. You say, "There is (or was) a lion in my front
yard and it's scaring (or it scared) the hell out of me." Actually it
wouldn't take long to find these words and construct this sentence, since
nothing in it is ambiguous and the individual word meanings are all very
familiar. You may hear yourself starting to utter the sentence before you
have formulated its end, and if something doesn't fit the experience
correctly you may have to edit it -- "No, I meant it's a tiger." A longer
or more complex communication may have to be edited numerous times before
you're convinced (correctly or not) that your words lead to the meanings
you intend, so others will grasp what you mean. The process is easiest to
see when you have to make adjustments to the meaning of a communication
before you're satisfied that it means what you intend it to mean.

Best,

Bill P.