linguistic structure in our shared environment

[bnhpct]

[From Bruce Nevin (2003.11.17 16:54 EST]
I think maybe we’ve bottomed out on the reality issue that derailed us,
so I’d like to try to get this “fairness and meaning” thread
back on topic.
Recapitulating where we were as of my (2003.10.22 23:00 EDT) post:
What I am saying does not entail a claim of special direct knowledge of
the environment that somehow bypasses perception.
Here’s a scenario. I control a perception of a sentence in English being
audible to you, for example:
Linguistic structure is in our shared
environment.
A sentence is a structured arrangement of words, which are themselves
structured arrangements of morphemes, syllables, and phonemes or
phonological contrasts. So far we are entirely in the realm of my
perceptions as speaker.
We agree, I think, that I perceive structure in this, and that I
control those perceptions – for example, the words must come in that
particular order to be an English sentence, and to be this particular
sentence.
The input functions, references, and output functions for my control of
these perceptions accord with conventions established as common knowledge
among users of the English language. In the course of learning those
common-knowledge conventions I have become so organized as to be able
control these perceptions by means of various behavioral outputs, in this
example by saying the above sentence to you.
This is also true of you. In the processes of having also learned the
common-knowledge conventions of English, you, too, have become so
organized as to construct the same structure out of your
perceptual inputs from your environment while I am speaking.
You recognize the structures that constitute this sentence. These are of
course structure-perceptions which your input functions have constructed,
but you perceive them to be the words, etc. that I have put in the
environment. Remembering common-knowledge conventions as to what
intentions one has when producing such structures, you ‘read’ into them a
perception of my intentions; in particular, you infer meanings that you
perceive (correctly or incorrectly) that I intended you to understand
from it.
Among these meanings that you infer are meanings that I did not intend.
You say:
You’re asserting that you have direct knowledge
of the
environment, bypassing perception.
This is an inference that you have constructed. I did not actually say
“I know this because I have direct knowledge of the environment,
bypassing perception.”
But you could not have made this inference without first understanding
what is directly said in my sentence, namely:
Linguistic structure is in our shared
environment.
I did not say how I arrived at the conclusion stated in this sentence;
you inferred that I must be claiming to have arrived at it by some sort
of direct apprehension of the environment.
What I intended (but did not say) was that I arrived at this conclusion
about the environment in the same way that you arrived at the above
conclusion about my intended meaning, and indeed the same way that we
arrive at any conclusions about reality: by inference.
I infer this conclusion about reality – that there is linguistic
structure in the acoustic effects of my speaking – from the fact that
you construct the same perceptions of language structure that I do –
namely, the above sentence (prior to your associating any meanings with
it or deriving any inferences from it).
(Any fluent speaker of English will perceive the same linguistic
structures; but it is certainly possible, maybe even very probable, that
any two such fluent speakers of English will attribute somewhat different
meanings to it, and derive different inferences from it.)
My inference that there is linguistic structure in our acoustic
environment seems unavoidable because your only relevant inputs for
constructing these perceptions are the physical effects of my control
actions. There is no other channel by which that information can get from
my output functions to your input functions.
This does not mean that every structural feature that is perceived has a
direct physical counterpart in the environment. The information in the
speech signal is virtually always degenerate. But the physical effects of
my speaking include physical transforms or traces of enough aspects or
parts of that linguistic structure for you to construct the rest. Your
(re)construction of the linguistic information that I intended is based
on your control of those physical effects of my speaking, but could not
be accomplished without the redundancies in language. Language is highly
redundant, and we use that redundancy to help us construct perceptions
for which the evidence in the speech signal is incomplete, ambiguous, or
even absent.
There is redundancy at every level of language structure. Speakers may
reduce redundant or low-information forms. I’ve given various simple
examples: pigpen in a subordinate clause vs. in a freestanding
sentence (phonetics of unstressed or de-emphasized phonemes); can
not, cannot, can’t; sgwout for let’s go
out (phonemic shape of morphemes); John plays piano and Alice
violin (repeated words in syntactic context). That hearers undo these
reductions (Alice plays violin? I didn’t know that!) suggests
control of redundancy. However, for the present discussion it does not
matter whether these redundancies are controlled perceptions or effects
of control.
The differences of formant contours for the same consonant before
different vowels provides another case in point. First, the contrast is
with other consonants.

And the transient is the same for all consonants produced at that point
of articulation. The second formant comes up from below for b,
p, or m; it comes down a bit from above for d,
t, or n; and it comes down from much farther above for
ga, ka, or nga.
Even when you compare the various allophones of the same consonant before
different vowels, the direction from which the transient moves to the
location of the vowel formant is constant. For voiceless aspirated
consonants like p, t, and k, less of that transient
is clearly audible. (Compare the formants in the voiceless h at
the end of the spectrogram for ba.) For nasal consonants m,
n etc., which are fully voiced, a formant turns up at the origin
of the transient to the following vowel. In the course of babbling a baby
doubtless learns that those sound squigglies do that when you
close your lips and say mama, or baba, or papa, and
that when you touch your tongue tip just so and say dada,
or nana, or tata, and this when you touch that other
part of your tongue to the top of your mouth and say gaga or
nganga or kaka.

Surely, like any environment variable these can be disturbed. The noise
masking experiment that Rick mentioned is an example of such disturbance.

Bill Powers (2003.10.24.1226 MDT)–

Something is certainly in the environment, but
is it

literally shared? To create structures that we experience as
linquistic

structures, we must do things to the environment, but what is done to
the

environment need only evoke the proper perceptions; it need not itself
have

the same structure.

Whatever it is in the environment, it evokes the same perceptions
(phonemes, syllables, morphemes, words, in a certain sequence, under
certain intonation contours) in more than one person simultaneously.
These hearers each independently associate meanings with these language
perceptions and draw inferences from those meanings, but they perceive
the same linguistic structure as a common basis for these associations
and inferences. Repetition demonstrates that they control the identical
linguistic structure. That seems to me to be a pretty strong
demonstration that whatever it is in the environment when one of them
produces a repetition, it’s common input to all of them. That’s what I
mean by shared. If you mean something different by the word shared, then
suggest a technical term that won’t be a source of confusion for
you.

Does the environment control behavior in this scenario? Whatever the
common input may be which is shared in the environment, it is itself a
product of behavior. It is the means by which the speaker controls a
perception of having spoken as intended. But it is also the means by
which the speaker controls a perception of being heard as intended. If
your intention as listener is to hear the speaker as the speaker intends,
then yes, you are controlling for the speaker to control your perceptual
input. Does the speaker thereby control your behavior? Well, is
recognition behavior? Is verification that you have recognized correctly
behavior? Then so be it. Communication, understanding, and cooperation do
seem to require that we intentionally put ourselves at one another’s
disposal at least to this extent.

    /Bruce

Nevin

(Attachment 4e7d0ea1.jpg is missing)

···

At 01:38 PM 10/24/2003 -0600, Bill Powers wrote:

[Bill_Powers]

[From Bill Powers (2003.11.17.1934 MST)]

Bruce Nevin (2003.11.17 16:54 EST) --

I'm afraid that there are some points of difference between us that are
extremely hard to make clear or illustrate. although we unquestionably
agree on most basic concepts about reality, language, perception, and so on.

What I am having trouble formulating convincingly is the concept that two
people can agree on a perception even though the environmental means by
which they explore their agreement never repeat twice. Here is an allegory,
meaning an illustration of a principle in a different context (my
definition, not Webster's)

You say, "eight." I say, "Yes, 7 plus 1." And you agree with me: "that's
what I said, 6 plus 2." A third party comes along, and tells us we're
really talking about 5 plus 3, and both of us agree with him: you say,
"Yes, 9 minus 1" and I say, "I agree, too: 4 plus 4."

Of course for this allegory to work, none of us should be aware that when
we speak, we are using a different expression every time, so we are
perceiving only the outcome of the process as the same thing each time. All
we should know is that it seems to us that we are saying the same thing
every time. Since we have no way of checking to see what actually is
passing from one of us to the other, we will never discover these
variations. We will assume that if we hear the same phonemes, morphemes,
words, sentences, and other structures, the reality behind them must be the
same every time. But I am proposing that this assumption can't be
supported. My multiple control systems demo is an illustration of how the
allegory above can come to pass.

Of course we can infer what is going on in the reality between us, which
requires stipulating to our various understandings of physical reality. But
if we can do that, we can also infer differently, by using different
premises, which simply goes to show that anything we say about reality is
arbitrary. If, for some reason, you wish for linguistic structures to exist
in the world between brains, you can provide premises and logic that will
lead to that conclusion. And if you wish for them not to exist, that
conclusion can be accomodated too, using the same kinds of tools. My
position is that we can speak safely only about our own experiences. with
other people's experiences coming in a distant second under the assumption
that all people are organized if not identically then at least similarly.

There is a further problem, which is that when you and I agree that we are
hearing the same thing, there is no way at all for either of us to verify
that your experience is like my experience. This does not apply just to
sensations like red, but to all other levels of perception as well. When
you say "i before e except after c," you assume that "before" means the
same experience to me as it does to you. No amount of cross-checking will
ever establish the truth of that assumption. We simply have no way of
comparing our experiences to each other.

Somehow we have to find a way of speaking about communication, or more
generally, knowledge, that does not implicitly assume that we can verify
our perceptions by looking at the external reality. Either that, or we will
just have to give up on this whole problem and do what everyone else seems
to end up doing: speak like naive realists and forget that there is any
problem. I do that when I model, although having to model the environment
does serve as a reminder that even the environment is a perception.

It's possible that the basic problem here is not solvable by human beings.
It's possible that even if the solution were known to some superior being,
it could not be expressed in any way that human beings could understand. If
only we had one example of an independently evolved intelligence! But
perhaps even that would not suffice.

Best,

Bill P.

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.18.0614)]

Bill Powers (2003.11.17.1934 MST)

Indeed. As far as we are concerned, the world consists of experiences
(perceptions) and the stories we tell about those experiences.

Bruce Gregory

[Martin_Taylor2]

Martin Taylor [2003.11.18.0939 EST]

I've been lurking and not reading much of CSG this year because of
pressure of other work, but seeing the Subject line on Linguistic
structure and shared environment has tempted me to break silence.
Bruce suggested I write on this for the Festschrift. I would have
liked to, but I've been up to my ears (still am). Sorry about that.
Here's a quick summary of my thoughts on the matter. Take them as you
will.

[From Bill Powers (2003.11.17.1934 MST)]

Somehow we have to find a way of speaking about communication, or more
generally, knowledge, that does not implicitly assume that we can verify
our perceptions by looking at the external reality. Either that, or we will
just have to give up on this whole problem and do what everyone else seems
to end up doing: speak like naive realists and forget that there is any
problem. I do that when I model, although having to model the environment
does serve as a reminder that even the environment is a perception.

I'm tempted to suggest to those that have a video of the 1993 CSG
meeting that they check out what I said on the matter at that time.
The gist of it is this: that the naive infant acts in a variety of
ways, few of which actually serve to influence its perceptions in
"useful" ways. It reorganizes so that its actions actually do begin
to control some of its perceptions. Some of those actions are on the
inanimate world, some on other animals, and some on other humans. For
the purposes of this argument, we can largely ignore the
reorganizations affecting actions on the inanimate world.

Actions on other animals (including, initially, humans) result
sometimes in the animal doing something that brings the infant's
perceptions closer to their reference values, sometimes not. For
example, hitting is not likely to bring a "good" change in perception
from the subsequent actions of either human or pet. So, as with the
more predictable inanimate world, reorganization proceeds so that the
growing infant's actions are more likely to get it "what it wants".
In other worlds, it becomes increasingly socialized.

In my 1993 talk, I concentrated on why there is no full formal
grammar for a natural language, using the above as a basis. I started
with the unrealistic proposition that a group of infants would be
socialized with entities that did use a fully formalized syntax. Call
them "seniors". Each infant individually would perceive that when it
made certain noises--perceptibly similar to noises made by its local
seniors--certain things would be likely to happen, sometimes good
(i.e. bringing other perceptions nearer references), sometimes not.
When it made noises in patterns not perceptibly similar to those made
by seniors, more or less random things might happen.

The long result of this reorganization would be that the infant could
reorganize so that in some degree it could control its perceptions by
making noises, and the more those noises mimicked the noise patterns
made by its seniors, the better the control.

However, there would be little improvement in control between a
reorganization state in which the pattern match of the infant's
noises was "excellent" and a state in which the pattern match was
"perfect." Indeed, statistically, perfection would be impossible to
achieve, both because the perceptual consequences of any "patterned
noise action" would be context-dependent and not all of that context
would be perceptible to the infant (e.g. the infant could not know
when the senior had had a bad day at the office), and also because
perfection would require the infant to have observed all possible
patterns that could be emitted by the seniors.

The conclusion from the foregoing is that the infant's "language
structure" could not be identical to that of its seniors. Even if the
seniors' language structure could be exactly prescribed by some
formal device, the infant's will not be--at least not by the same
specifications in all details. Furthermore, each infant would be
exposed to different seniors in different curcumstances, so each
would devise a somewhat different language structure.

Now forget the "seniors" for a moment, and consider the group of
infants interacting together after they have reorganized to make
noises in patterns that are fairly effective in dealing with seniors.
The infants' structures differ from each other, perhaps only
slightly, but they will differ. Control of each infant's perceptions
through the emission of patterned noises will not be as effective as
it would have been through interaction with seniors (partly because
of the differences in the language structures of the individuals,
partly because of the relatively poor control the infants have over
the inanimate environment). So the infants' individual control
systems will tend to continue to reorganize to improve the control
that noises have when emitted in the presence of other infants.

The group of infants will tend to drift by individual reorganization
toward a state in which their individual language structures are more
similar than they would have been had their contact continued to be
only with their seniors. But their individual language structures
will not be identical. All the same, there must be a core of
approximately the same structures in all the infants who are able to
control their own perceptions by making patterned noises at each
other.

That core is a "shared language structure", and I argued in 1993 that
it can be treated as having the same status as any more concrete
object in the perceived world. It is presumptively "out there". It is
"my language", and it is independent of the way "I" idiosyncratically
use it.

Continuing this process to further generations of infants, to whom
the "seniors" are the previous generation's infants, it seems clear
that the core structure of a language shared by the successive
generations is likely to drift, not only in terms of the patterns of
noises, but also in terms of the noises themselves. An aspirate "p"
may tend to increase its aspiration (and here I deliberately refer to
the "out there" core language rather than to the individual
perception). Over time, some infants may find such a highly aspirated
"P" indistinguishable from what we might call a "hard F", and might
successfully make the noise with a slightly different mouth gesture
than the seniors used. (Successfully, here, means controlling
perceptions through the action of the language-using entities in the
neighbourhood).

Over a few generations, the central formal structure of the original
language is not likely to change very much, but so many variants are
likely to become established through this drift feedback mechanism
that full formal description of even the commonly used structures is
likely to become both very complex and insufficient to describe
completely what any individual actually does with language.

To summarize: Infants reorganize so that their actions somewhat
predictably influence their perceptions, thereby achieving some
degree of control. Some part of that control occurs when they make
certain noises in the presence of others who also make noises.
Control is best when the infant perceives its noises to be patterned
similarly to the patterning of noises made by those others. When
"those others" include other infants, the patterning used by each
infant in the group will tend through reorganization to converge to
some central "structure", which, for perceptual control purposes,
will have a status like that of any perceptible concrete object. That
structure will evolve and change over time and generations of
infants, and at no time will it be fully susceptible to a complete
formal analysis (though its basic principles may be).

I hope this is helpful, and that my breaking my long silence is not
perceived as intrusive.

Martin

[Bill_Powers]

Martin Taylor [2003.11.18.0939 EST]

I've been lurking and not reading much of CSG this year because of
pressure of other work, but seeing the Subject line on Linguistic
structure and shared environment has tempted me to break silence.
Bruce suggested I write on this for the Festschrift. I would have
liked to, but I've been up to my ears (still am). Sorry about that.
Here's a quick summary of my thoughts on the matter. Take them as you
will.

[From Bill Powers (2003.11.17.1934 MST)]

Somehow we have to find a way of speaking about communication, or more
generally, knowledge, that does not implicitly assume that we can verify
our perceptions by looking at the external reality. Either that, or we will
just have to give up on this whole problem and do what everyone else seems
to end up doing: speak like naive realists and forget that there is any
problem. I do that when I model, although having to model the environment
does serve as a reminder that even the environment is a perception.

I'm tempted to suggest to those that have a video of the 1993 CSG
meeting that they check out what I said on the matter at that time.
The gist of it is this: that the naive infant acts in a variety of
ways, few of which actually serve to influence its perceptions in
"useful" ways. It reorganizes so that its actions actually do begin
to control some of its perceptions. Some of those actions are on the
inanimate world, some on other animals, and some on other humans. For
the purposes of this argument, we can largely ignore the
reorganizations affecting actions on the inanimate world.

Actions on other animals (including, initially, humans) result
sometimes in the animal doing something that brings the infant's
perceptions closer to their reference values, sometimes not. For
example, hitting is not likely to bring a "good" change in perception
from the subsequent actions of either human or pet. So, as with the
more predictable inanimate world, reorganization proceeds so that the
growing infant's actions are more likely to get it "what it wants".
In other worlds, it becomes increasingly socialized.

In my 1993 talk, I concentrated on why there is no full formal
grammar for a natural language, using the above as a basis. I started
with the unrealistic proposition that a group of infants would be
socialized with entities that did use a fully formalized syntax. Call
them "seniors". Each infant individually would perceive that when it
made certain noises--perceptibly similar to noises made by its local
seniors--certain things would be likely to happen, sometimes good
(i.e. bringing other perceptions nearer references), sometimes not.
When it made noises in patterns not perceptibly similar to those made
by seniors, more or less random things might happen.

The long result of this reorganization would be that the infant could
reorganize so that in some degree it could control its perceptions by
making noises, and the more those noises mimicked the noise patterns
made by its seniors, the better the control.

However, there would be little improvement in control between a
reorganization state in which the pattern match of the infant's
noises was "excellent" and a state in which the pattern match was
"perfect." Indeed, statistically, perfection would be impossible to
achieve, both because the perceptual consequences of any "patterned
noise action" would be context-dependent and not all of that context
would be perceptible to the infant (e.g. the infant could not know
when the senior had had a bad day at the office), and also because
perfection would require the infant to have observed all possible
patterns that could be emitted by the seniors.

The conclusion from the foregoing is that the infant's "language
the "seniors" are the previous generation's infants, it seems clear
that the core structure of a language shared by the successive
generations is likely to drift, not only in terms of the patterns of
noises, but also in terms of the noises themselves. An aspirate "p"
may tend to increase its aspiration (and here I deliberately refer to
the "out there" core language rather than to the individual
perception). Over time, some infants may find such a highly aspirated
"P" indistinguishable from what we might call a "hard F", and might
successfully make the noise with a slightly different mouth gesture
than the seniors used. (Successfully, here, means controlling
perceptions through the action of the language-using entities in the
neighbourhood).

Over a few generations, the central formal structure of the original
language is not likely to change very much, but so many variants are
likely to become established through this drift feedback mechanism
that full formal description of even the commonly used structures is
likely to become both very complex and insufficient to describe
completely what any individual actually does with language.

To summarize: Infants reorganize so that their actions somewhat
predictably influence their perceptions, thereby achieving some
degree of control. Some part of that control occurs when they make
certain noises in the presence of others who also make noises.
Control is best when the infant perceives its noises to be patterned
similarly to the patterning of noises made by those others. When
"those others" include other infants, the patterning used by each
infant in the group will tend through reorganization to converge to
some central "structure", which, for perceptual control purposes,
will have a status like that of any perceptible concrete object. That
structure will evolve and change over time and generations of
infants, and at no time will it be fully susceptible to a complete
formal analysis (though its basic principles may be).

I hope this is helpful, and that my breaking my long silence is not
perceived as intrusive.

Martin

···

At 08:34 AM 11/18/03, you wrote:

[bnhpct]

[Bruce Nevin (2003.11. : EST)]

From Bill Powers (2003.11.17.1934 MST)–

I’m afraid that there are some points of difference between us that are

extremely hard to make clear or illustrate. […]

What I am having trouble formulating convincingly is the concept that two

people can agree on a perception even though the environmental means by

which they explore their agreement never repeat twice.

You have convinced me that this is possible.

Here is an allegory,

meaning an illustration of a principle in a different context (my

definition, not Webster’s)

You say, “eight.” I say, “Yes, 7 plus 1.” And you agree with me: "that’s

what I said, 6 plus 2." A third party comes along, and tells us we’re

really talking about 5 plus 3, and both of us agree with him: you say,

“Yes, 9 minus 1” and I say, “I agree, too: 4 plus 4.”

Of course for this allegory to work, none of us should be aware that when

we speak, we are using a different expression every time, so we are

perceiving only the outcome of the process as the same thing each time.

This puts the divergence inside the controllers. Weren’t you talking about variability in the environment?

All

we should know is that it seems to us that we are saying the same thing

every time. Since we have no way of checking to see what actually is

passing from one of us to the other, we will never discover these

variations. We will assume that if we hear the same phonemes, morphemes,

words, sentences, and other structures, the reality behind them must be the

same every time.

But I did not assume that the reality behind them is “the same”. In fact, I emphasized that there is a great deal of variability. Perceptual input functions make of a range of values for a given variable one perception. Output functions and effectors maintain the discreteness of these ranges of values in any given context (e.g. before /a/ or before /e/). Speakers of different dialects set the references for these ranges differently, and so evidently set input functions for others’ speech differently from input functions for their own speech. (There is other evidence for redundancy of input functions for speech, e.g. discriminating different speakers in a crowd.)
Phoneme perception is categorial. So long as the acoustic variables fall within certain discrete ranges, they are perceived as discrete elements – that is, not analogically but ‘digitally’. It does not matter what processes are used in producing an utterance. All that matters is that a value in one of two contrasted ranges for e.g. point of narrowest occlusion by the tongue, controlling tactile perceptions, has some regular physical transform which is (almost) simultaneously perceived as a value in one of two contrasted ranges for an acoustic variable, e.g. formant frequencies.
It does not matter whether you produce American English /r/ with retroflex tongue tip and your interlocutor produces it with a humped tongue body plus lip rounding. You each maintain the perceived acoustic results in contrast to /l/, /n/, /d/, etc. I gave the more extreme example of my colleague with the scar-stiffened face, and the ventriloquist. So yes, there can be variation in the means of producing the intended acoustic effects.
This has no bearing on the point:

• There is some physical transform of speech activities in the environment.
  (There is no claim here that it is the same physical transform every time, and we do not need to know what it really is in order to proceed.)

• This physical transform is (almost) simultaneously perceived by both the speaker and hearer as certain language perceptions.

• For the speaker, A, these are the intended language perceptions that she was controlling by means of speaking.

• The hearer, B, can in turn control these same language perceptions by speaking.

• There is some physical transform of B’s speech activities in the environment.

• Both B and A perceive this physical transform as language perceptions.

• B perceives it not only as the language perceptions that B is controlling by means of speaking, but also as identical to the language perceptions heard previously when A was speaking

• Concurrently A perceives this same physical transform of B’s speaking as the same as the language perceptions that she was previously controlling.

• This process of repeating and confirming for oneself can be iterated as often as you please.

• There is no claim here that it is the same physical transform every time.

• Whatever the physical reality may be for the first utterance, and however much it may vary in the repetition, it is such that perceptual input functions of the speaker and of all the hearers perceive the repetition as the identical language elements as the first utterance.

There is a great deal of variation in measurable properties of that physical transform. Different speakers have different voice pitch, different vocal tract length, different size and configuration of larynx, different configurations of resonant chambers in their heads, and so on and on. Aside from these differences of “voice quality”, in different dialects the acoustic ranges are set differently, but what is relevant is that they are contrasted, and contrasted in parallel to the contrasts in your dialect (with one or two mergers tolerable, especially of vowel contrasts).

But I am proposing that this assumption can’t be

supported. My multiple control systems demo is an illustration of how the

allegory above can come to pass.

It shows that it is possible. It does not show that it is plausible. Furthermore, it is of negative utility for modeling. Consider the task of modeling arm movement if you assume that a sensor for tendon stretch is ‘actually’ aggregating inputs from an incalculable number of subatomic particles at various distances, most of them unthinkably vast. But wait a minute – what tendon stretch sensor? What tendon? What stretch? None of these exist, except as our perceptions. By means of our visual analysis of anatomy and … Wait a minute! Whose perceptions?
But for what I am saying this does not matter. All that matters is that perceived acoustic variables be controllable in a reliable way. Set aside any question of how this is done. The fact that is getting buried under quasi-philosophical speculation here is the fact of repetition,

Of course we can infer what is going on in the reality between us, which

requires stipulating to our various understandings of physical reality. But

if we can do that, we can also infer differently, by using different

premises, which simply goes to show that anything we say about reality is

arbitrary. If, for some reason, you wish for linguistic structures to exist

in the world between brains, you can provide premises and logic that will

lead to that conclusion. And if you wish for them not to exist, that

conclusion can be accomodated too, using the same kinds of tools. My

position is that we can speak safely only about our own experiences. with

other people’s experiences coming in a distant second under the assumption

that all people are organized if not identically then at least similarly.

I say again: It does not matter to me whether or not there is actually linguistic structure in the environment.
I don’t see how there can fail to be traces thereof sufficient for A and B to construct the same perceptions from their respective inputs from A’s control actions, as demonstrated by A and B constructing the same perceptions from their respective inputs from B’s different control actions.
If A and B carry a heavy slab of stone, one at each end, they (and we) may fairly assume that there is something really there that each is perceiving. Whether their perceptions are the same, or whether their subjective experiences of those perceptions are the same, is completely irrelevant.
The movement of the stone as they carry it is a product of control actions by each of them concurrently with the other. For this to be analogous to language, the stone itself would have to be a product of control actions by one of them (the speaker), while concurrently being passively controlled (in recognition mode) by both of them. This passive control is the means by which the speaker verifies successful control of the utterance (correcting it if necessary by repeating all or part of it), and provides remembered perceptions by reference to which the hearer repeats what was said and the speaker verifies that it is indeed a repetition.
Just as with the table, we may fairly assume that there is something there that they both are perceiving. Furthermore, we can know something about it that we can’t know about the table, because its very existence is as a product of perceptual control.

···

At 08:18 PM 11/17/2003 -0700, Bill Powers wrote:

There is a further problem, which is that when you and I agree that we are

hearing the same thing, there is no way at all for either of us to verify

that your experience is like my experience. This does not apply just to

sensations like red, but to all other levels of perception as well. When

you say “i before e except after c,” you assume that “before” means the

same experience to me as it does to you. No amount of cross-checking will

ever establish the truth of that assumption. We simply have no way of

comparing our experiences to each other.

Somehow we have to find a way of speaking about communication, or more

generally, knowledge, that does not implicitly assume that we can verify

our perceptions by looking at the external reality. Either that, or we will

just have to give up on this whole problem and do what everyone else seems

to end up doing: speak like naive realists and forget that there is any

problem. I do that when I model, although having to model the environment

does serve as a reminder that even the environment is a perception.

It’s possible that the basic problem here is not solvable by human beings.

It’s possible that even if the solution were known to some superior being,

it could not be expressed in any way that human beings could understand. If

only we had one example of an independently evolved intelligence! But

perhaps even that would not suffice.

Best,

Bill P.

[Richard_Marken2]

[From Rick Marken (2003.10.18.1040)]

Bruce Gregory (2003.11.18.0614)

Indeed. As far as we are concerned, the world consists of experiences
(perceptions) and the stories we tell about those experiences.

The stories are perceptions, too, right? So all you really need to say is that
"the world consists of perceptions".

Best

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.18.1540)]

Rick Marken (2003.10.18.1040)

Bruce Gregory (2003.11.18.0614)

Indeed. As far as we are concerned, the world consists of experiences
(perceptions) and the stories we tell about those experiences.

The stories are perceptions, too, right? So all you really need to
say is that
"the world consists of perceptions".

Yes, but I still find the distinction useful. The class of perceptions
that we call stories play a very special role in being human.

Bruce Gregory

[Richard_Marken2]

[From Rick Marken (2003.11.18.1350)]

Bruce Gregory (2003.11.18.1540)--

> Rick Marken (2003.10.18.1040)

> The stories are perceptions, too, right? So all you really need to
> say is that "the world consists of perceptions".

Yes, but I still find the distinction useful. The class of perceptions
that we call stories play a very special role in being human.

Different people have different ideas about which perceptions play a very special
role in being human. For some it's stories, for others it's language, for still
others it's logic. For me it's music, PCT and one particularly lovely story
teller, who plays a more special role in my being human than even the wonderful
stories she reads to me at bedtime.

It seems confusing to distinguish perception in general from your favorite
perceptions (which are just a subset of perception). Even if certain perceptions,
like stories, are very important to you, they are still perceptions, not reality,
which is the point of the "it's all perception" story, I think.

Best regards

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[bnhpct]

[Bruce Nevin (2003.11.18 22:15 EST)]

From Bill Powers (2003.11.17.1934 MST)–

What I am having trouble formulating
convincingly is the concept that two

people can agree on a perception even though the environmental means
by

which they explore their agreement never repeat twice.

You have convinced me that this is possible. Your demo shows that it is
possible.
It does not show that it is plausible. And it seems to be of negative
utility for modeling. Consider the task of modeling arm movement if you
assume that a sensor for tendon stretch is ‘actually’ aggregating inputs
from an incalculable number of subatomic particles at various distances,
most of them unthinkably vast. But wait a minute – what tendon stretch
sensor? What tendon? What stretch? None of these exist, except as our
perceptions. Our perceptions by means of our visual analysis of anatomy
and … Wait a minute! Whose perceptions? Whose eyes? What are eyes?
But for what I am saying this does not matter. All that
matters is that perceptions of acoustic variables be controlled in a
reliable way. Set aside any question of how this is done.

What is getting buried under quasi-philosophical speculation here is the
fact of repetition. A speaks. B repeats the very same utterance. A
and all other observers affirm that it is the same words, albeit spoken
in B’s voice and dialect. A in turn repeats what B said. B and all other
observers affirm that it is the same words spoken again in A’s voice and
dialect. Regardless of the great variability of precise formant heights,
contours, durations, pitch, number of harmonics within a formant, etc.
etc., the sentence “Hello, ace!” was repeated instead of the
sentence “Hello, ice!” or the sentence “Hello,
ass!”

Somehow we have to find a way of speaking
about communication, or more

generally, knowledge, that does not implicitly assume that we can
verify

our perceptions by looking at the external reality.

In this exercise of repetition there is no verifying of perceptions by
some sort of direct reference to external reality. Nor is there any
assumption, either explicit or implicit, that we can verify our
perceptions by direct access to reality, bypassing perception.
As a matter of explanation it seems to me by far the more
plausible hypothesis that, as speakers and hearers control phonemic
distinctions, the control actions of speakers have effects on reality
that are in distinct ranges of variation, resulting in acoustic
perceptions that are in distinct ranges of variation.

Under conditions of redundancy or low information content speakers
control distinctions or contrasts with lower gain (e.g. unstressed vowels
merge into schwa). This is not counter-evidence, it actually demonstrates
control of contrast as means of transmitting linguistic information.

But this explanatory hypothesis is not required as an assumption. All we
need are the observational data of repetition and contrast. These yield
phonemes, and thence, from dependencies (departures from equiprobability)
of phonemes, all the rest of language structure. These are matters of
perceptual control, without privileged reference to ‘Real’ properties of
reality.

Communication and knowledge are much broader problems. Communication is
accomplished by many means in addition to language. Language is effective
for error-free transmission of information, but presents a restricted and
somewhat rigid channel for communication. Knowledge of certain kinds
appears to depend upon language – specifically, the theories and
‘stories’ we construct about reality. But most knowledge is embodied in
non-linguistic forms. For example, our organization as control systems,
the organization of our input and output functions, our reference
perceptions and their reference values, all are forms of knowledge.

    /Bruce

Nevin

···

At 08:18 PM 11/17/2003 -0700, Bill Powers wrote:

[bnhpct]

[From Bruce Nevin (2003.11.18 22:24 EST)]

Martin Taylor 2003.11.18.0939 EST --

I'm tempted to suggest to those that have a video of the 1993 CSG
meeting that they check out what I said on the matter at that time.
The gist of it is

The take-away for me from that video was the convergence of individual reference values in accord with emergent social conventions.

Of course language (and culture) changes, and for such reasons and in such ways as you suggest. But the issue at the time was whether social conventions or standards even exist. There was great alarm lest we suggest that some sort of 'social control system' was able to reach into people's heads and set their references. Fortunately, the level of discourse has matured somewhat since 1991, in no small part thanks to your contributions in 1993 and later. Currently, there is great alarm lest we suggest that the environment controls behavior.

I hope this is helpful, and that my breaking my long silence is not
perceived as intrusive.

Welcome back.

         /Bruce Nevin

···

At 10:34 AM 11/18/2003 -0500, Martin Taylor wrote:

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.19.0628)]

Rick Marken (2003.11.18.1350)

Bruce Gregory (2003.11.18.1540)--

Rick Marken (2003.10.18.1040)

The stories are perceptions, too, right? So all you really need to
say is that "the world consists of perceptions".

Yes, but I still find the distinction useful. The class of perceptions
that we call stories play a very special role in being human.

Different people have different ideas about which perceptions play a
very special
role in being human. For some it's stories, for others it's language,
for still
others it's logic.

Language and logic are both stories.

For me it's music, PCT and one particularly lovely story
teller, who plays a more special role in my being human than even the
wonderful
stories she reads to me at bedtime.

What a lovely story.

It seems confusing to distinguish perception in general from your
favorite
perceptions (which are just a subset of perception). Even if certain
perceptions,
like stories, are very important to you, they are still perceptions,
not reality,
which is the point of the "it's all perception" story, I think.

I beg to differ. Perceptions _are_ reality. In fact, the distinction
demonstrates a story about perceptions ("they are not reality").

Bruce Gregory

[Bill_Powers]

[From Bill Powers (2003.11.19.90554 MST)]

Bruce Nevin (2003.11.18 22:15 EST) –

You have convinced me that this is
possible. Your demo shows that it is possible.

It does not show that it is plausible. And it seems to be of negative
utility for modeling.

Absolutely right.When modeling, I decide on a model for the environment
and accept it as true, for the purpose at hand.

But for what I am saying this
does not matter. All that matters is that perceptions of acoustic
variables be controlled in a reliable way. Set aside any question of how
this is done.

What is getting buried under quasi-philosophical speculation here is the
fact of repetition. A speaks. B repeats the very same
utterance.

Oops. How do you know that B repeats the very same utterance? From your
hearing the very same thing twice? But I think I have shown that that
would not be sufficient evidence.This is the very crux of the issue –
how you know. Tell me the answer to this one, Bruce, and I will give you
a big wet smack on the cheek the next time we meet. I’m not arguing just
for fun here, you know, or merely playing amateur philosopher. I know
what I would like to believe – but that’s no reason to believe
it.I can get by on faith that I can observe the True Reality, but faith
is a feeble substitute for knowing. If you have the answer, don’t keep it
to yourself.

Best,

Bill P.

···

A and all other observers
affirm that it is the same words, albeit spoken in B’s voice and dialect.
A in turn repeats what B said. B and all other observers affirm that it
is the same words spoken again in A’s voice and dialect. Regardless of
the great variability of precise formant heights, contours, durations,
pitch, number of harmonics within a formant, etc. etc., the sentence
“Hello, ace!” was repeated instead of the sentence “Hello,
ice!” or the sentence “Hello, ass!”

Somehow we have to find a way of
speaking about communication, or more

generally, knowledge, that does not implicitly assume that we can
verify

our perceptions by looking at the external reality.

In this exercise of repetition there is no verifying of perceptions by
some sort of direct reference to external reality. Nor is there any
assumption, either explicit or implicit, that we can verify our
perceptions by direct access to reality, bypassing perception.
As a matter of explanation it seems to me by far the more
plausible hypothesis that, as speakers and hearers control phonemic
distinctions, the control actions of speakers have effects on reality
that are in distinct ranges of variation, resulting in acoustic
perceptions that are in distinct ranges of variation.

Under conditions of redundancy or low information content speakers
control distinctions or contrasts with lower gain (e.g. unstressed vowels
merge into schwa). This is not counter-evidence, it actually demonstrates
control of contrast as means of transmitting linguistic information.

But this explanatory hypothesis is not required as an assumption. All we
need are the observational data of repetition and contrast. These yield
phonemes, and thence, from dependencies (departures from equiprobability)
of phonemes, all the rest of language structure. These are matters of
perceptual control, without privileged reference to ‘Real’ properties of
reality.

Communication and knowledge are much broader problems. Communication is
accomplished by many means in addition to language. Language is effective
for error-free transmission of information, but presents a restricted and
somewhat rigid channel for communication. Knowledge of certain kinds
appears to depend upon language – specifically, the theories and
‘stories’ we construct about reality. But most knowledge is embodied in
non-linguistic forms. For example, our organization as control systems,
the organization of our input and output functions, our reference
perceptions and their reference values, all are forms of
knowledge.

    /Bruce

Nevin

[Richard_Marken2]

[From Rick Marken (2003.11.19.0820)]

Bruce Gregory (2003.11.19.0628)--

> Rick Marken (2003.11.18.1350)

> It seems confusing to distinguish perception in general from your
> favorite perceptions (which are just a subset of perception). Even if certain
> perceptions, like stories, are very important to you, they are still
perceptions,
> not reality, which is the point of the "it's all perception" story, I think.

I beg to differ. Perceptions _are_ reality.

Then what's "reality"?

My model of the situation says that perception is a representation of reality (or
what's called the "environment" in the PCT model). We have no direct access to
reality so we will never know, for sure, the actual nature of the reality "out
there". But I am persuaded that a very accurate guess about the nature of reality
comes from science in the form of models that precisely fit carefully made
observations.

In fact, the distinction demonstrates a story about perceptions ("they are not
reality").

Yes. That's what I think. Perceptions (including stories, of course) are _not_
reality; they are a representation of reality.

Best

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[bnhpct]

[From Bruce Nevin (2003.11.19 11:21 EST)]

Bill Powers (2003.11.19.90554 MST)

What is getting
buried under quasi-philosophical speculation here is the fact of
repetition. A speaks. B repeats the very same
utterance.

Oops. How do you know that B repeats the very same utterance? From your
hearing the very same thing twice?

No. From A having the same perceptions twice – the second time in
perceptions of B’s speech, and the first time in perceptions of A’s own
speech. And then having the same perceptions any number of times as A, B,
and perhaps others present, repeat the same exercise. Also, B having the
same language-perceptions twice, or any number of times; and the
observer(s) having the same language-perceptions as various participants
repeat the utterance.
“Hearing the very same thing twice” is ambiguous as to what
perceptions we are talking about. Words spoken with a variety of
pronunciations and voice qualities are perceived as the same words. The
variations of pronunciation and voice quality are not language
perceptions; the words are. (This could be said more precisely, but at
much greater length; can we let the detail be included by
reference?)
The utterance exists as language only in the perceptions of its hearers.
If you play a recording of the Gettysburg address and there’s nobody
there to hear it, there is no language present. This is not the same as
Bishop Berkeley’s idealism, which claimed that there are no sounds
present. I am simply saying that without an appropriately organized
perceiver there are no perceptions, and the utterance that is being
repeated exists as language only in the perceptions of its hearers.
We don’t have to suppose that the perturbations in the atmosphere are
repeated in order to know that we have recognized in them the same words.
In fact, we know that phoneme recognizers are categorial, that morpheme
and word pronunciations vary predictably, etc. (More detail included by
reference.)
B has no evidence that A has recognized the same language-perceptions
twice, other than A’s saying so. But A needs no other evidence than her
own perceptions. Likewise, A has no evidence that B has recognized a
repetition, but A needs no other evidence than his own perceptions. And
the observers need no other evidence than their own perceptions.
The corroboration is nice, and we make use of it, but it is secondary. If
A knows it wasn’t a repetition (“I said pick up the pan, not
pick up the pawn”), the fact that B says it was a repetition
is not convincing (“Sorry, you’re wrong” “Yeah, OK, I was
just checking”). Certainly there are disturbances to recognition,
but in our laboratory scenario we ruled these out.

When we work with an unknown language we must rely upon the report of a
native speaker assenting or denying that something is a repetition, but
once our references are appropriately set and the interconnections of
recognizers for concurrent phonemes, syllables, intonation contours, etc.
are appropriately organized, our own perceptions suffice.

    /Bruce

Nevin

···

At 06:08 AM 11/19/2003 -0700, Bill Powers wrote:

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.19.1305)]

Rick Marken (2003.11.19.0820)

My model of the situation says that perception is a representation of
reality (or
what's called the "environment" in the PCT model). We have no direct
access to
reality so we will never know, for sure, the actual nature of the
reality "out
there". But I am persuaded that a very accurate guess about the
nature of reality
comes from science in the form of models that precisely fit carefully
made
observations.

An excellent story. I like it, too.

In fact, the distinction demonstrates a story about perceptions
("they are not
reality").

Yes. That's what I think. Perceptions (including stories, of course)
are _not_
reality; they are a representation of reality.

Indeed. According to the story you tell, perceptions are not reality;
they are a representation of reality. If this is _not_ a story, exactly
what is it? (A model is, of course, a story.)

Bruce Gregory

[Bill_Powers]

[From Bill Powers (2003.11.19.1020 MST)]

Bruce Nevin (2003.11.19 11:21 EST)–

Oops.
How do you know that B repeats the very same utterance? From your hearing
the very same thing twice?

No. From A having the same perceptions twice – the second time in
perceptions of B’s speech, and the first time in perceptions of A’s own
speech.

But this does not establish that B is perceiving the same thing as A. It
says only that A is perceiving the same thing A perceived before, with
A’s perceptual input functions. A’s perceptual input functions ignore
changes in the input that A is not equipped to perceive. If, on every
trial, the actual input comes close enough to the form that A’s
perceptual input functions are tuned to, A will perceive exactly the same
thing every time, because A experiences only the output of this
input function. Different patterns appear the same to A,
not because they have something.in common, but because the input function
cannot distinguish betwen the different patterns.

And then having the same
perceptions any number of times as A, B, and perhaps others present,
repeat the same exercise. Also, B having the same language-perceptions
twice, or any number of times; and the observer(s) having the same
language-perceptions as various participants repeat the
utterance.

This argument founders on this little phrase “the same.” Who is
judging the sameness? All any single judge can look at is the output of
the judge’s own perceptual function. All this judge can say is
“the same” or “different” is this output signal: same
as the last time it appeared, different from the last time it
appeared.

“Hearing the very same thing
twice” is ambiguous as to what perceptions we are talking about.
Words spoken with a variety of pronunciations and voice qualities are
perceived as the same words. The variations of pronunciation and voice
quality are not language perceptions; the words are. (This could be said
more precisely, but at much greater length; can we let the detail be
included by reference?)

The words may be heard as the same words, at the same time one perceives
that their lower-order components are different. So here we have the
“reality” problem in a form where the observer can see
both the input and the output of the perceptual function. In PCT terms,
what this shows is that an input function can report as the same a set of
inputs that are actually different – in this case, observablydifferent. So clearly the sameness is in part an artifact of the
perceptual input function. It arises for precisely the same reason that
“2,6” can be seen as the same as “3,5”. These two
expressions are patently different, but if they are put through a
function that implements addition, the output of the function will be 8
in both cases (and a great many other cases beside). If, on the other
hand, they are fed into a function that implements multiplication, the
outputs will NOT be the same: they will be 12 and 15, respectively.
Whether the value of the function of the inputs is the same or different
depends on the nature of the function just as much as on the
inputs.

The utterance exists as language
only in the perceptions of its hearers. If you play a recording of the
Gettysburg address and there’s nobody there to hear it, there is no
language present. This is not the same as Bishop Berkeley’s idealism,
which claimed that there are no sounds present.

We agree completely on this. Language exists only in the perceptions of
its hearers.

I am simply saying that
without an appropriately organized perceiver there are no perceptions,
and the utterance that is being repeated exists as language only in the
perceptions of its hearers.

Again, complete agreement.

We don’t have to suppose that the
perturbations in the atmosphere are repeated in order to know that we
have recognized in them the same words.)

Same red flag again. We have no way of comparing your perceptual input
functions with my perceptual input functions. I think that our
difference, such as remains, lies in your statement above. The key word
is “recognized.” This implies that the word really is in the
atmospheric perturbations somehow, and that all the listener has to do is
respond perceptually to it. I contend that this confuses what the brain
does with what the atmosphere does. Again, my example of “2,6”.
Is an “8” really implicit in that pair of numbers presented
together? Or is there an infinity of numbers implicit there – every real
number, for that matter? If an infinity of numbers is implicit (as it is,
since an operation can be defined that will convert this pair of numbers
into any desired number), then no number is implicit, since all numerical
outcomes are equally possible.

In fact, we know that phoneme
recognizers are categorial, that morpheme and word pronunciations vary
predictably, etc. (More detail included by reference.

This is fine, but you’re talking about perceptual input functions and the
relation of higher-order signals to lower-order signals, the same thing I
am talking about. At each level of perceptual organization we have
many-to-one functions, meaning that the perceptual signals at any given
level can be the same even though the inputs to that level are all
different. Does this mean that the higher-order perceptual signal is
somehow “in” the lower-order perceptions? I don’t think so, any
more than we can say that a toy crane is “in” the Erector Set
parts from which it is made, or that Michaelangelo’s statues were
“in” the marble block as he claimed.

B has no evidence that A has
recognized the same language-perceptions twice, other than A’s saying so.
But A needs no other evidence than her own perceptions. Likewise, A has
no evidence that B has recognized a repetition, but A needs no other
evidence than his own perceptions. And the observers need no other
evidence than their own perceptions.

Again, complete agreement. All you have to do now is agree that A and B
are looking at the outputs of their own perceptual input functions, not
the inputs, and not at each other’s perceptual signals, and we should
arrive at least at a resting place.
Also, please understand that I am not closing down the argument. The open
question in my mind is whether there is some strategy by which it could
be established beyond reasonable doubt that my perceptual input functions
are or are not the same in their organization as yours. Or is
neurosurgery the only answer? As long as each of us has to judge sameness
(from one person to another) strictly in terms of our own
perceptual input functions, I see no way to arrive at a firm answer. But
maybe there is one that I just don’t see. It wouldn’t be the first time
such a thing has happened.

The corroboration is nice, and we
make use of it, but it is secondary. If A knows it wasn’t a repetition
(“I said pick up the pan, not pick up the pawn”),
the fact that B says it was a repetition is not convincing (“Sorry,
you’re wrong” “Yeah, OK, I was just checking”). Certainly
there are disturbances to recognition, but in our laboratory scenario we
ruled these out.

I think we can show in the laboratory that there are differences in
repetitions that the listeners simply do not recognize as differences.
That’s actually pretty easy. We can probably do this at every level of
organization. That is, we can show that when someone thinks he heard
“pick up the pan” twice in a row, the word “pan” was
actually pronounced differently. This is more informative than finding
that the person can detect obvious differences. Difference can reliably
indicate difference while at the same time sameness does not reliably
indicate sameness.

When we work with an unknown
language we must rely upon the report of a native speaker assenting or
denying that something is a repetition, but once our references are
appropriately set and the interconnections of recognizers for concurrent
phonemes, syllables, intonation contours, etc. are appropriately
organized, our own perceptions suffice.

Maybe the answer to my questions is somewhere in that paragraph. But
maybe not. I think it is still possible for the informant and the
linguist to have differences in their perceptual input functions that
neither one can detect. After all, I think this is also possible for
native speakers of the same language. If, every time you say
“qweep” I hear “qwoop”, consistently and in all
circumstances, how would we discover that we were not having the same
experience? “Give me two qweeps,” you say. I, hearing
“give me two qwoops,” will pick up two of the things you have
always called “qwoops” before, and give them to you. And you
will say what I hear as “Thank you for the qwoops,” while you
hear yourself say “thank you for the qweeps,” which you agree I
gave you two of. Furthermore, when I say “quoop”, I actually
emit sounds which to a third party sound like “flurge,” but of
course I hear then as “qwoop” and you hear them as
“queep”, while a spectrograph records the patterns that are
really there, which are neither qwoop, qweep, nor flurge. Repeat the
spectral sounds, and the three parties each hear their respective
versions, while agreeing that qwoops, qweeps, or flurges are delicious
with chocolate sauce.

Now is it possible that somewhat more subtle differences in what we hear
and experience could always be detected and rooted out? Is simple
agreement ever going to eliminate the possibility of inadvertent and
undetected disagreement? If you know, tell me how. Knowing how would save
me a lot of trouble.

Best,

Bill P.

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.19.1359)]

Bill Powers (2003.11.19.1020 MST)

The utterance exists as language only in the perceptions of its
hearers. If you play a recording of the Gettysburg address and there's
nobody there to hear it, there is no language present. This is not the
same as Bishop Berkeley's idealism, which claimed that there are no
sounds present.

We agree completely on this. Language exists only in the perceptions
of its hearers.

I hate to disturb such agreement, but it seems to me that you have only
agreed on a convention. "We won't call something language unless it is
associated with the perceptions of hearers." Bruce N's distinction
seems to lack an underlying difference. This message is written in
English whether or not anyone reads it.

Bruce Gregory

[Richard_Marken2]

[From Rick Marken (2003.11.19.1111)]

Bruce Gregory (2003.11.19.1305)

Indeed. According to the story you tell, perceptions are not reality;
they are a representation of reality. If this is _not_ a story, exactly
what is it? (A model is, of course, a story.)

Yes. Of course, it's a story. And Beethoven's 7th is a symphony. And the bubble
sort is an algorithm. And _Hamlet_ is a play. And blue is a color. And they are
all perceptions.

You had said that the world consists of experiences (perceptions) and the stories
we tell about them. To me, this is like saying "the world consists of perceptions
and symphonies" or "the world consists of perceptions and algorithms". It's just
saying the same thing twice, in one case using synecdoche.

But it is something more than redundancy that annoys me about this statement. I
may be reading too much into it but the statement seems to imply that stories have
some special epistemological status, that they are somehow more "real" or
"accessible" than perceptions; that sorties are different than perceptions.

To me stories (models), like everything else we experience, are perceptions. When
properly combined with careful observation and logical analysis (also
perceptions), models can provide a very powerful basis for helping us understand
the presumed reality behind our perceptions. So models are a very important
perception, but mainly to people whose goal is to understand the presumed reality
behind their perceptions. To people who want to write symphonies or develop
programming algorithms, other perceptions can be more important than models
(stories).

Best

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Richard_Marken2]

[From Rick Marken (2003.11.19.1140)]

Martin Taylor (2003.11.18.0939 EST)--

To summarize: Infants reorganize so that their actions somewhat
predictably influence their perceptions, thereby achieving some
degree of control.

If they are in control, then their actions have a _highly_ predictable influence
on their perceptions; actions keep perceptions in the reference state specified by
the actor, protected from disturbance.

Some part of that control occurs when they make
certain noises in the presence of others who also make noises.
Control is best when the infant perceives its noises to be patterned
similarly to the patterning of noises made by those others.

If the infant is controlling for imitation then their is control _only_ if the
infant is able to produce perceptions for itself that are similar to the
perceptions it wants to mimic.

When "those others" include other infants, the patterning used by each
infant in the group will tend through reorganization to converge to
some central "structure", which, for perceptual control purposes,
will have a status like that of any perceptible concrete object.

If the pattern you are talking about is the sound in the environment then you are
saying what Bruce N. is saying, and I would say that you are both demonstrably
wrong. In fact that actual "patterning" of the sound output produced by artifact
will have to be slightly different on each occasion (when it wants to produce the
same perception). if for no other reason than to compensate for disturbances in
the environment( such as the sounds produced by another infant).

That structure will evolve and change over time and generations of
infants, and at no time will it be fully susceptible to a complete
formal analysis (though its basic principles may be).

Are you talking about perceptual structure or physical structure? Physical
structure, as I mentioned above, will change each time the same perception if
produced, so I don't think it makes sense to talk about the physical structure of
the sound evolving over time. But I agree that the reference for the perception
being produced is likely to change over time. I think the great vowel shift is a
neat example of this. There we had a rather systematic shift in people's
references for the vowel perceptions that go into making word perceptions.

Best

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org