Devil's biblio;Belief;Good $uggestion;Ref signals and "no"

[From Bill Powers (930525.1930)]

Ray Allis (930524.10:30 PDT) --

This is exactly the thesis of Julian Jaynes' "The Origin of
Consciousness ...

Be careful when you use the word "exactly." I might allow
"somewhat" or "in a general way through a sympathetic stretch of
interpretation." I prefer to reserve "exactly" to mean exactly.


Chuck Tucker (930525) --

RE: Devil's bibliography

I think it is very easy for any of us (some do it quicker that
others I have noticed; I can't account for it!) to dismiss
these statements as coming from persons who are ignorant,
foolish, stupid or just plain sinful BUT I would suggest that
we might all learn something if the statements are treated as
serious problems with CT (or PCT or HPCT) and point out how we
would deal with them.

As in Rick's post concerning detailed errors in Locke et al,
treating most of these statements seriously just turns into
another basic lesson in elementary control theory, which we can
see as a refutation but to which those committing the errors will
pay no attention at all. Most of the errors would never have been
committed if the person had taken any time at all to learn about

There is an enormous difference between a person who is trying to
grasp the meaning of control theory and has failed to understand
some subtle aspect of it and a person who is interpreting control
theory incorrectly for the specific purpose of showing its errors
or lacks.

If the first person doesn't see how control theory applies in a
specific case, that person will try to work out a CT explanation
and ask someone else if it works. When an error is found and a
more correct use is suggested, that person will go away again,
ponder the suggestion, and come back either with a big AHA or an
even better way of doing the analysis. Maybe no good answer will
be forthcoming; in that case, a research problem has been

The second person doesn't work that way. What you find, as in
Locke, is interpretation and after interpretation that clearly
violates the principles of CT or the models built on it, that is
loaded with assumptions about what CT says or means, and a use of
these assumptions and interpretations in order to make one point
after another proving that CT is not applicable or doesn't handle
a specific situation. There is not the slightest effort to find a
viable CT interpretation. The whole point, decided long before
the arguments were selected, is to refute CT. When that is the
clear intention, it is a complete waste of time trying to
overcome the arguments. There is no interest in learning how a
given problem might be approached using PCT. Just look at Carol
Fowler's response to an invitation to defend against my comments.
Even to enter into a conversation of this sort would be to grant
some legitimacy to PCT -- and the whole point of attacking it is
to take away that legitimacy.

The most difficult critics to answer are those who pick some
subject like consciousness, and challenge PCT to come up with an
explanation of it. There are plenty of questions that PCT can't
answer, because nobody has done the work, building up a knowledge
base from which even to attack such problems. Generally, such
critics already have their own answers with which they are
evidently satistifed. The only grounds for rejecting their ideas
would inevitably be interpreted as a counterattack -- pointing
out that their answer is shot full of holes, based on sloppy
thinking, and untestable, for example.

While I don't think we should stop listening to the critics, I
don't see that we have much to gain by spending a lot of time on
what they say. Remember, I've been listening to these same
criticisms for more than three decades. There simply aren't any
new ones. All that comes through to me now is "I hate this theory
because it doesn't agree with mine, and I'm going to do my best
to make it go away."
Rick Marken (930525.0800) --

Thanks for taking the time to go through the details. Saves me
the trouble.
Joel Judd (930524) --

In any truly entrenched belief system there is one key hypothesis
on which everything else depends. For those who take a literalist
view of the Bible, the key hypothesis is in the Bible itself,
where it says that THIS IS THE WORD OF GOD. If you accept that
one statement as true, it follows that everything else in the
Bible has to be true, assuming that God does not lie.

If there is a criterion by which you can recognize true
revelations, that criterion is clearly the key hypothesis. You
can forget about anything that follows from it; everything stands
or falls on that one hypothesis. I don't know what this criterion
is for the LDS, but in other religions I have heard about, it is
a special kind of mystical experience that strips away all doubt
and leave only a feeling of peace and absolute conviction. In my
view this kind of experience is attendant upon resolving some
fundamental internal conflict. There is a sense of the Ego giving
up and turning the problem over to a higher power -- in terms of
my science-fiction story, to a higher level of control within the

I don't dispute or discount phenomenological accounts of
religious experiences. It's only the explanation of them that I
find unconvincing.
Tom Bourbon (930524.1057) --
RE: science fiction

"Any less plausible?" you ask. Certainly not! ("Any less
profitable?" you did not ask. Certainly! Unless the Powers'
mountain-top PCT meditative shrine has acquired a golden dome
since the last CSG conference.)

I knew there was something missing. Golden domes, however, are
expensive. Now if everyone out there were to donate a mere 10% of
their (gross) income to me, from now on, this deficiency could be
quickly corrected (after all, PCT was revealed to me, wasn't it?
And if I hadn't been nice enough to tell everyone about it, you
would all still be living in darkness, ignorance, stupidity,
sloth, avarice, greed, adultery, behaviorism, etc.. Show a little
Bruce Nevin (930524.1414) --

So the word and syllable detectors specify the reference
signals for the phoneme detectors. The reference signal for
producing /n/ is probably the same for "no", "on", "paining",
and "painting". That the behavioral outputs differ in these
four kinds of cases is probably a matter of the physics of
moving the tongue around rather than variation of the reference

You're right. However, there's a subtle point here that depends
on the current concept of perception in HPCT. A given perception
is simply a signal indicating THAT the perceptual function has
found something to report. Assume that there's a control system
dedicated to controlling for the word "no", as heard/spoken. When
this system receives a reference signal indicating that this word
is to be experienced, the reference signal, plotted against time,
looks like this:

                          ******************** ...
(0) ***********************
As long as the reference signal remains high, the control system
will produce the perception "no" over and over: nononononono...

The perceptual signal representing the first "no" will look like
this on the same time scale:

                                   * *
                                  * *
                                 * *
                                * *
                               * *
(0) ************************** *****

In other words, a plot of the perceptual signal for the WORD "no"
neither looks like nor sounds like the sensations or phonemes
received by the perceptual function that recognizes the word as a
word. What comes out is simply a signal like any other neural
signal. That it functions as a word, and specifically as the word
"no", is determined by the higher systems that receive this
signal as well as others. In order to function socially as the
word "no", the acts that produce the phonemes that lead to this
signal must produce a signal in other people similarly used by
their higher-level systems.

The phonemes, of course, are also represented as signals: the
word "no" is recognized when certain signals from phoneme-
perceivers reach the system that produces the above signal. Those
phoneme signals would look just like the "no" signal, although
their time course might be different.

My point is that the reference signal that calls for producing
the perception of "no" is not itself the word "no." It is just a
signal, which specifies saying no not because of anything special
about the reference signal, but because it enters a certain
comparator. Only its destination makes it a "no" reference
signal. And all that makes the corresponding perceptual signal
represent the word "no" is the way it is derived from perceptual
signals indicating that certain phonemes have been perceived. The
only way to get the "no" signal to appear is to supply signals
from phoneme detectors within a certain range of temporal and
amplitude limits. It's this concept that made Mark Olsen's
remarks on "place" coding ring a bell with me -- assuming this is
anything like what he meant. As you noticed.

The big objection to this whole picture is that this doesn't seem
at all like the way we experience "no" or phonemes or anything
else. Some quality seems to be missing. But what is missing, I
think, is the context of all the other perceptual signals that
are in awareness at the same time.

There is a Zen exercise (mantras being an example) in which one
focusses on some perception, at any level, for so long and so
narrowly that it becomes the whole universe of awareness. When
this experience finally occurs, whatever the perception was
completely loses its distinctive character. It is simply
something happening HERE instead of THERE, and it is
indistinguishable from any other perception except for its place
in the world of experience. Its special unique qualities don't
return until the concentration relaxes and broadens to include
other perceptions again. You can do this with any perception: a
word, a taste, a face, an emotion. It seems, as Paul Churchland
(sp?) suggested in his "network theory of knowledge," that
perceptions have conscious significance and character only as
part of a whole network of perceptual signals.

I'm only about 30% sure why I'm bringing this up in connection
with your comments about saying "no," but perhaps you can see a
glimmer of pertinence. Higher systems DO control their own
perceptions by VARYING lower-level perceptions (through
manipulation of reference signals) -- but don't forget that the
higher level systems are perceiving different aspects of the
world. A system that tells a lower system to produce "no" does
not perceive the word "no". It does not know it is telling the
lower system to produce a word. Only the lower system can do that
or know that; there are no words at the higher levels. The higher
level system perceives something that is derived from the
presence of the "no" signal and other signals, and in this
derived perception there is no longer any trace of "no". To
awareness, which can span several levels at a given time,
everything from the intensities and sensations on up coexists in
the experienced world, but a single level in the hierarchy always
perceives only its own specialized aspect of the situation.

This, at least, is how I would give a strict reading to the HPCT
model in this case.

Note that playing back a perceptual signal from memory reduces,
under this reading, simply to playing back a signal. Period.

When we model behavior, all this becomes self-evident. Whatever
we call a given signal in the model, it is nothing but a number
that varies with time and depends on other signals -- numbers. No
matter what attribute of experience you're modeling, when you
finally represent it in a model it has no character but magnitude
and behavior of magnitude through time. This could be seen as a
gross deficiency of modeling as applied to human experience -- or
as a fundamental insight into what it's all about. Take your
choice (after you send me your 10%).

Your remarks on attention, by the way, point to a great big
research area, suggested at least three years ago by David
McCord. We need to set up some control tasks requiring division
of attention, to see what happens to the parameters of control.