[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [xmca] The Ubiquity of Unicorns: conversation


I am very gratified to see that you grasp the full thrust of my work on
as it relates to word meaning. You are the first to do so!  I feel
vindicated and
validated! Thank you.

To be honest, when I first started down this path in 1981, I was working on
it felt right, but I was largely unconscious of the direction I was taking.
But in 2004
I had an epiphany, and since that moment, all of the pieces have been
falling into
place, and I have become able to consciously articulate the consequences of
taking this particular stance. For many years I have had to cope with a
fair degree
of disorientation due to the fact that private speech is at the center of
so many
philosophical knots. It was the shift of focus to conversation that finally
sense of so many issues, and that holds out the promise of making sense of
others (that I have yet to return to).

The epiphany came in the form of an old "Snow White" refrain: "Mirror,
mirror on
the wall, who's the fairest of them all?". Periodically, I would ask
myself: "Why do
children talk out loud to themselves?", and the answer that returned was
"Because they are learning how to think." But on this one, surprising
the answer that came back was: "Because they are learning how to talk." It
at that moment that I realized that LSV never made explicit the obvious
next step:
after thinking becomes verbal, speaking becomes intelligent--because inner
speech is then recruited for the purpose of talking! Once it became clear
inner speech--the mechanism for directing one's attention--was available
children to use in all realms of their lives, it was obvious that language
would become a major beneficiary of this new mental faculty. So Vygotsky
deserves TWO Nobel prizes in psychology--one for conceptualizing how
thinking becomes conscious, and the other for conceptualizing how children
master their native tongue. What a genius he was!

Far be it from me to reign you in!  Because I only became conscious of this
path a few years ago, I have yet to see its limitations or constraints--nor
do I think
about them, for I am still uncovering all of the possibilities that this
affords. If you would be so kind as to humor me, let me point out a few of
problem areas that I believe this new conversational focus sheds light on--
besides the conceptual integration it provides, as you have so delightfully

1) By looking at the development of word meaning from a conversational
point, Vygotsky's claims about the transformation of private speech take on
a new significance. I used to look at private speech as *causing* the
of thinking from unconscious and impulsive to conscious and deliberate, but
fact, it is the development of word meaning that *causes* private speech to
change! Naive social speech, private speech, inner speech, mature
social speech--these do not cause development per se, but instead are the
forms in which the development of word meaning manifests itself. This, in
prompts the question: Just what IS word meaning, and what causes it to
Sagely, Vygotsky (Chapter 7, again) points us to the answer: words develop
one direction, while meaning develops in the opposite direction. To be
about it, a child's first word does NOT correspond to its dictionary
meaning, as you might expect, but to its global, undifferentiated
meaning. In other words, the forms and functions of words and meanings are
"sprung" in an incompatible way from the start, and so development is
merely the
"unwinding" action brought about by the initial tension. This needs a
visual (I hope
this diagram comes through everyone's email system):

(Embedded image moved to file: pic14204.jpg)
The first stage (infancy) in the development of word meaning  takes place
the form of social speech with a caregiver, and is represented in the
as the relationship between "word" and "topical" meaning". A child's first
is a whole turn at talk, but all the child has available to express this
thought is a
single word. As the tension on both sides of this equation exerts pressure
on the
child, the relationship between "sentence" and "propositional" meaning
center stage. The child's focus alternates between the relationship of word
sentence, and the relationship of topical to propositional meaning. The
stage, which I believe corresponds to the abbreviation of private speech
(its final form before becoming abbreviated inner speech), moves the focus
to the relationship between conversation and lexical meaning. The process
of abbreviation, which is tied to predication, involves the child in
how individual word meanings (lexical meanings) are related to the whole
of conversation. The child must consider which words need to be uttered in
order to move the topic of conversation along, and which do not need to be
voiced. "Clots" of words are the result.

2) Dialectical materialism: What does this Marxist form of philosophical
explanation bring to the problem? Although I have been a student of Marxism
even longer than I have been a student of Vygotsky's theory, I was unable
formulate the connection between them--until recently. I believe the very
diagram above applies to Marx and Engels' explanation of the development
of human society--if you make some substitutions. For Marx and Engels, it
the forms and functions of *labor (economic) activity* and *civil
activity* that are "sprung" in an incompatible way from the start, causing
society to morph from primitive communism to slave society, to feudal
to capitalist society, and (hopefully) to socialist society and eventually
advanced communism. I don't expect everyone to agree with this
and I understand that introducing politics into the discussion is a risky
but I think it's hard to ignore (and useful to acknowledge) the parallels
Marx and Engels' stages and Vygotsky's stages of primitive social speech,
impulsive private speech, regulative private speech, abbreivated private
speech, and (hopefully) inner speech and eventually advanced (intelligent,
rational) social speech. Of course, the role of dialectical logic in
theory is much more involved than what I have outlined here, but my aim is
simply to draw attention to the "feedback" effect of Vygotsky's theory on
understanding of Marx and Engel's theory. My experience thus far is that
of these two very special theories can be helpful in understanding the

3) The relevance of Bahktin and Voloshinov's analyses to LSV's theory:
If we make an analogy between initiating and responding in conversation
to throwing and catching a ball, then we can see the design of an
"utterance" in a new light. In playing "catch" with a ball, each
must first catch the ball before he or she can throw it back. Thus, a
at catch involves first catching, then throwing. These moves are done in
a smooth transition, constituting a "turnaround". Likewise, in
a "turn" at talk involves first listening, then responding. This, too, is
(by competent communicators) in a smooth fashion. Listening flows right
into speaking, enabling the conversation to move forward. Thus, each
utterance by a speaker actually consists of *both* listening and
Just waiting patiently for one's turn at talk--without listening to the
not constitute communication with speech. One must listen (and reflect?)
before speaking, or else "responding" doesn't occur.

This leads to the notion that every utterance is linked both backwards and
forwards. Every initiation responds to some thought (even ones that start
a conversation), and every response constitutes a new initiation--a
of the ongoing conversation. Bahktin claimed that the relationship
utterances in conversation eventually develops into a relationship *within*
an utterance. Every utterance "anticipates" a response--the voice of the
"other" is contained in every utterance. (I must confess that I am only now
beginning to look at the relationship of Bahktin's work to this new
so I may be off-base). (No pun intended)

That's way more than enough for now. I don't want to dominate this

But thanks for listening and responding, XMCA-ers!  You're marvelous!

Best wishes,

             Steve Gabosch                                                 
             com>                                                       To 
             Sent by:                  "eXtended Mind, Culture, Activity"  
             xmca-bounces@webe         <xmca@weber.ucsd.edu>               
             r.ucsd.edu                                                 cc 
             10/29/2009 10:58          Re: [xmca] The Ubiquity of          
             PM                        Unicorns: conversation              
             Please respond to                                             
              "eXtended Mind,                                              

Peter, I am very happy to see how well developed the empirical and
practical side of your work is!  Have you published any of your
studies, have more writings on your theory, etc.?  I am actually quite
excited by your work.

Your argument for viewing the conversation as a unit of word-meaning
(in the Vygotskyan sense of "meaning through language use") seems
compelling.  I like this idea quite a bit, and find myself wanting to
go with it, try it out, test it, play with it, see how far it goes.
Once this simple and obvious relationship is pointed out, as you are
doing so clearly and concisely, a lot of connections seem to fall
right into place.  They are the same old connections, of course, but
more dialectical, more dynamic, more far-reaching.  For me, anyway ...

For example, this concept makes me want to ask the reverse question:
how can word-meaning be conceived **other** than in conversation?  And
how can the **parts** of conversation - parts of words, words, word-
sequences, sentences, sentence-sequences - or utterance-sequences in
general - as well as gestures, facial expressions etc. etc. - be
conceived other than as just that - as parts of **conversations**?
(This suggests that the conversation could actually be **the**
structural unit of word-meaning - with everything else being parts of
conversations - in the way we usually think of parts of sentences.  Am
I taking this too far?)  These are not new questions, of course, but
ones that your work revives in a new light.

This line of thinking makes me want to take this a step further - or
"upward" - how can a particular conversation be conceived other than
in its particular social, cultural and historical as well as
psychological "context" (its concrete reality)?  This idea, of course,
has been deeply explored by Bakhtin and others.  Have you had a chance
to relate your thinking on this to Bakhtin, Volosinov, etc.?

Thinking of word-meaning this way is one of those ideas that has "been
there" all along, but so big that I know that I for one never quite
apprehended it in this way.  I have always thought of a conversation
as some kind of combination, sequence or interchange of "word-
meanings" - not a "word-meaning" in and of itself.  This in some ways
is a mind-boggling concept.  We think of our words as being in our
conversations, but it is harder to conceptualize the opposite being
the case - that it is our conversations that are in our words, and
that our words and word-meanings actually mean nothing apart from and
outside our conversations.  Obvious, understood already in many ways,
but also new and dizzying for me.

This viewpoint - and perhaps I am taking this too far and someone is
going to have to rein me in - this view of word-meaning places, the
dialogue, or the conversation, as the transmission mechanism, the
concrete link between 1) the meanings of social relations and culture
(the interpsychological) and 2) psychological meanings (the
intrapsychological).  I like putting conversation in the middle like

On one hand, of course, this is not at all a new idea.  How else do we
communicate?  Hello?  On the other hand, by viewing conversation as
the **necessary** form that word-meaning takes - that is, that
conversation IS word-meaning - this insight opens up new possibilities
for harnessing Vygotsky's theory of word-meaning as a powerful lens
into two simultaneous realms - 1) human social relations and
activities, which are inherently language-based, and 2) individual
psychology (the higher mental functions), which, according to
Vygotsky, is also inherently language-based.

This seems to be exactly what your research is trying to do, Peter.
You are showing how to look at the totality of an activity in a given
situation - in this case, children playing while talking out loud - to
see and show how human action and human meaning-making are a
dialectical unity - using Vygotsky's classical theory of word-meaning
as a lens into **everything** that is simultaneously happening - or,
at least, that which is happening that you can figure out how to
record and analyze!

An interesting reverse question to ask here about what you have
observed in your studies so far: what are some of the **limitations**
of Vygotsky's concept of word-meaning in trying to understand what
these children are doing, thinking, saying, feeling?  Above, I speak
almost as though it can explain "everything".  Can it?  Probably not!
LOL  So what are the **limits** of this lens into human activity?
(For example, it probably does not explain roles, personalities,
membership in social classes ...)

To the extent that this approach is successful at viewing certain
multiple levels of human activity at once, one can legitimately ask:
how can this be so?  I find myself answering that question with
another:  isn't this **precisely** what Vygotsky accomplished when he
discovered word-meaning as a basic unit of analysis?  To show how to
view the **convergence** of the social and the individual?  To see
both realms at the same time at the place they meet and transform one
another and together become something new?  This is probably why I
like this idea so much, it seems to expand Vygotsky's approach to do
exactly what we want it to.

If we only apply the concept of word-meaning to "words," perhaps we
are setting our sights too low, and only on certain details, missing
the larger picture.  But if we step back and include **the
conversation** along with words and sentences, new possibilities seem
to emerge.  Maybe someone will shoot me down here and point out how I
am going off the deep end with this (or more likely, just leave me be
with my false hopes!) - but I think you are on to something, Peter.
(Maybe you think I am going too far, too!  LOL)

Another question: have you tried to apply Leontiev's activity theory
(conditions-based tasks and operations, goal-based actions and action-
sequences, lifestyle-motivated activities and activity systems, etc.)
to your work on word-meaning and conversation?  I am not aware of
anyone who has applied these concepts to the dynamics of conversation,
conversation analysis, etc.  I wonder if they could be applied ...

One last question.  Sorry about so many questions!  Not all need to be
answered all at once - there is always lots of time, no rush here, of
course.  My last question: have you given thought to how this
expansion of the Vygotskyan concept of word-meaning could be applied
to Vygotsky's work on concept formation (syncretic formations,
concrete complexes, systematic concepts, etc.) - or perhaps, the other
way around - how Vygotsky's work on concept formation might apply to
your work on word-meaning?

- Steve

On Oct 29, 2009, at 5:03 PM, Andy Blunden wrote:

> I see you use utterance as a unit of analysis. In the previous mail
> you referred to using "conversation", but utterance is surely the
> unit of which conversation is made up. Do you have Bakhtin's idea in
> mind for "utterance" at all? Otherwise the notion of "turn taking"
> in private speech is very challenging, isn't it?
> All sounds fascinating,
> Andy

Hi, Andy.

My method of eliciting the speech involves children playing a game
with a challenging cognitive component. I videotape each child
(usually 4-, 6-, and 8-year-olds) individually-- with me in the room,
at a distance, as a possible resource for interpersonal speech if the
child is feeling motivated enough to seek it out. The child is given
instructions about how the game works, what the objective is, and is
told "it's a new game, and I want to know what you think of it, and
whether other kids would like it." The game I have tended to use is a
Piagetian task in which a child is asked to connect some straight and
curved wooden tracks together in such a way that they fit between two
fixed endpoints glued to a large gameboard placed on the floor. The
children can crawl on the board and move the tracks around. The
objective is to "build as many roads as possible with the tracks" that
I provide them. To make the entire experience more fun (especially for
the younger children), I also affixed to the board a miniature
"school" area, a miniature "amusement park and zoo" area, a "lake"
area, and a "mountain" area. These areas are situated so that the
tracks can be constructed between the areas, and therefore create a
very enticing fantasy-play opportunity. This was a conscious decision,
for my aim is to elicit every type of private speech use--
particularly, "word play", "emotional expression", "descriptions of
ongoing activity", "planning", "monitoring action", and the like. I
want to have numerous examples of the full range of private speech
productions, from the most impulsive to the most regulatory, so that I
can examine the conversational structures and functions developmentally.

The analysis of the videotapes starts with a written transcript of the
speech stream. This transcript is then used in conjunction with the
videotaped speech and behavior, and serves as the place to record the
codings that are then imposed on the data. The first cut is to segment
the stream of speech into utterance units. Here we start to get quite
conceptual already; how shall we define the boundaries? In general,
the conventions for doing speech analysis are borrowed from
linguistics, but when it comes to segmenting the flow of speech,
developmental psychologists seem to have invented their own
conventions that best suit their psycholoinguistic needs. My
impression is that most analysts have adopted a set of criteria that
more or less coincides with sentence boundaries. For my purposes, this
is very helpful, for it divides the data neatly into two types: 1)
single sentences, sentence fragments, phrases, single words, and even
nonverbal behavior (such as facial expressions)--all of which are
located within an utterance's boundaries; and 2) conversation--
connections between the utterance under investigation and other
(adjacent and non-adjacent) utterances--if any. Since utterance
boundaries are a matter of convention, I would like to see researchers
make a collective, considered decision about how to slice and dice
linguistic data, so that we can best serve our research agendas. In my
case, the agenda is to test LSV's claims about the development of word

Once the data are segmented into utterance units, each unit can be
"tagged" with attributes along the following lines: 1) Is it private
speech or social speech? (direction of gaze, loudness, pitch,
intonational contour, content, relation to ongoing action--all of
these are relevant to making the decision); 2) Is it task-relevant,
supportive of fantasy play, supportive to self-regulatory action,
etc.? (evidence for these functions are marshalled as well); and, in
my own approach (which very few private speech researchers seem to
share), I ask: 3) Is the utterance in question part of a
conversational sequence? (form, function, and content relationships to
other utterances and to ongoing activity are considered); and 4) What
conversational acts (speech acts) can be inferred from the deployment,
production, and performance of the utterance? (words chosen,
grammatical structure, literal propositional meaning, inferred
meanings of individual words from sentential context, conversational
context, and context of ongoing activity; other inferred meanings of
the proposition from conversational context and context of ongoing
activity, etc.).

Naturally, all of this takes training, time, funding, and persistence--
none of which are going to happen unless researchers are convinced
that these (more extreme) measures are warranted.

So....if YOU were going to investigate LSV's concept of word meaning,
dipping into the stream of conversation (or monologue, or narration)
at various points in time to examine the key qualities that will allow
you to determine the level of development of a child's activity of
communicating (interpersonally or personally) with speech, I ask you:
doesn't conversation have to be included in such an investigation? Is
it possible to investigate LSV's claims about word meaning if the data
are limited to just words, phrases, and sentences? How about the
social, interpersonal "exchange" property? Words and phrases are just
structures, but without conversation to breathe life into them, what
have you got?

As you can tell, I have several axes to grind!

Thus, conceptual explorations of word meaning need to be tempered
against the practical and pragmatic concerns involved with
implementing and applying the concept to data. I am convinced that
both theory and practice need one another for either one of them to

Now that I've laid bare the methodology for obtaining the data that
then become the empirical "facts" about word meaning, I would
appreciate any help or suggestions you might have for improving or
deepening the conceptual structure of this methodology, so that the
"facts" we seek might be more in line with LSV's conception.

Sorry to be so verbose.


>> WHat experimental technique do you use, Peter, to observe
>> private speech?
>> Andy

>> Peter Feigenbaum wrote:
>>> Steve--
>>> Thanks for the warm welcome.
>>> I thought my response to the ongoing discussion would be useful,
>>> but I can see that it will take more work on my part to demonstrate
>>> the relevance of what I'm proposing.
>>> I should also add that my orientation to LSV, private speech, and
>>> word meaning is not just conceptual, but also practical. While I
>>> truly
>>> enjoy the erudite philosophical discussions on this listserve, and
>>> find
>>> them helpful in sharpening up crucial concepts, there is research
>>> to be
>>> done to verify these ideas, and that enterprise involves making
>>> concrete
>>> decisions about conceptual possibilities.
>>> As for your synopsis of my ISCAR presentation, you did a fine job.
>>> All I
>>> would add (and this was not spelled out sufficiently in my
>>> presentation)
>>> is a fuller description of the properties that emerge when
>>> speaking and
>>> thinking "converge". As I see it, that momentous convergence that
>>> LSV
>>> points to is none other than the momentous activity of a child
>>> uttering
>>> her first meaningful word. Several aspects of that new activity
>>> need to
>>> be made explicit.
>>> First, it's not just thinking and speaking that converge inside a
>>> child's
>>> head; the whole activity is a convergence of infant and caregiver as
>> well.
>>> That is, a child's first word is spoken "to someone", and therefore
>>> constitutes the very first instance of the infant engaging in
>> interpersonal
>>> conversation--even though the infant is unaware of the rules and
>>> conventions for conducting this new activity. Fortunately, adults
>>> step in
>>> and do the work of creating a state of mutual involvement until
>>> the child
>>> is competent enough to do so on her own.
>>> Which leads to the second point: if conversational activity is
>>> born with
>> a
>>> child's first meaningful word, then tracking the growth and
>>> development
>> of
>>> conversational understanding and skills needs to be a central aim of
>>> Vygotskian research. Since the convergence of speaking and
>>> thinking is
>>> also the same event in which word and meaning come together as one
>>> to form a new activity, tracking conversational development is
>> intertwined
>>> with the development of word meaning.
>>> To respond to your first question--is word meaning an activity of
>> "meaning
>>> through language use" or does it refer to a particular
>>> psycholinguistic
>>> structure--the answer is: both. Speech communication, word
>>> meaning, and
>>> conversation are, in my view, conceptually interchangeable. These
>>> terms
>>> all refer to the activity of making meanings with speech sounds.
>>> Thus,
>>> particular speech structures are woven into the activity. That's the
>> great
>>> thing about "activity"--it's a material process involving material
>>> structures,
>>> and from it can come "ideas" and "idealizations"!
>>> If Vygotsky's claims about word meaning and its development are to
>>> be
>>> tested using empirical data, then the concept of word meaning
>>> needs to be
>>> translated and transposed so that it can take the form of an
>>> empirical
>>> methodology--while retaining its basic features. Personally, I
>>> don't know
>>> of
>>> one private speech researcher who uses LSV's concept of word meaning
>>> when analyzing speech data--and to me, that suggests something is
>>> very
>>> wrong. So I have been focused on augmenting the concept of word
>>> meaning
>>> in order to create a corresponding method that can capture word
>>> meaning
>>> and its development from a data-analytic perspective. That involves
>> making
>>> certain conceptual choices, particularly regarding the linguistic
>>> units.
>>> Your suggestion that there could be any number of linguistic units
>>> that
>>> might
>>> be proposed is certainly valid. My justifications for choosing
>>> words,
>>> sentences,
>>> and conversation as the three major levels in the organization of
>>> speech
>>> communication are based on both conceptual and pragmatic grounds. We
>>> need a method of studying this phenomenon developmentally from a
>>> data
>>> perspective, and this scheme seems tome to be the most workable one.
>>> Does that help clarify my viewpoint?
>>> Peter
>>> -----xmca-bounces@weber.ucsd.edu wrote: -----
>>> To: "eXtended Mind, Culture, Activity" <xmca@weber.ucsd.edu>
>>> From: Steve Gabosch <stevegabosch@me.com>
>>> Sent by: xmca-bounces@weber.ucsd.edu
>>> Date: 10/28/2009 07:19AM
>>> Subject: Re: [xmca] The Ubiquity of Unicorns: conversation
>>> Welcome, Peter! I liked your presentation and paper at ISCAR 2005 a
>>> lot, and am very glad to see you getting involved here on xmca.
>>> Your post outlines a very interesting concept, extending Vygotsky's
>>> concept of word-meaning to the linguistic unit of conversation.
>>> I took a quick re-look at your 2005 paper, A Dialectical Model of
>>> Vygotsky's Theory of Speaking and Thinking.  I like the way you
>>> describe Vygotsky's three stages of thinking/speech development, and
>>> the way you propose a fourth that completes the full developmental
>>> "cycle".
>>> Here is a quick synopsis of your paper (in my clumsy terms).  As you
>>> describe them, each of these stages takes two 'forms', so to
>>> speak, a
>>> speaking form and an intellectual (thinking) form.
>>> The paper describes  Vygotsky's three stages of speaking/thinking
>>> as:
>>> 1) interpersonal speech, which combines the speaking form of
>>> external
>>> social speech dialogues, and the intellectual form of implicit
>>> practical thinking;
>>> 2) personal speech, which combines the speaking form of external
>>> private speech monologues and the intellectual form of explicit
>>> verbal
>>> thinking; and 3) personal speech, which combines the speaking form
>>> of
>>> internal inner speech monologues and the intellectual form of
>>> explicit
>>> verbal thinking.
>>> To these stages outlined by Vygotsky, you suggest a fourth stage:
>>> 4) interpersonal speech, which combines the speaking form of inner
>>> speech/social speech with internal monologues and external
>>> dialogues,
>>> and the intellectual forms of explicit and implicit verbal thinking
>>> and practical thinking.
>>> In this post, you extend Vygotsky's concept of word-meaning beyond
>>> the
>>> word and the sentence to the conversation as a unit of analysis.  A
>>> key element that emerges in the second and third stages of speaking/
>>> thinking development - that element being the monologue, or inner
>>> speech - creates the basis for the fourth stage.  And the
>>> monologue is
>>> also a key element of how you are looking at conversation.  Very
>>> interesting analysis!
>>> Please correct the rough edges of my synopsis - and where I am off-
>>> base.
>>> As for considering conversation a "third level" linguistic unit,
>>> following the word and the sentence, I have a naive question.  The
>>> term "word-meaning" that Vygotsky uses in Russian - znachenie
>>> slova -
>>> as I understand it (via Holbrook Mahn's paper at a recent AERA
>>> conference), means something more like "meaning through language
>>> use"
>>> rather than "individual-word-meaning".  In this sense, Vygotsky may
>>> not have been referring specifically to any linguistic unit when he
>>> spoke of word-meaning.  Yes?  No?
>>> As for linguistic units, I have another naive question.  Apologies
>>> for
>>> using non-technical terms here.  We have things like parts of words,
>>> words, phrases, combinations of phrases, sentences, and then perhaps
>>> sequences of directly related sentences.  And then, of course, we
>>> have
>>> conversations.  And the distinctions can get a lot more complicated
>>> than that when we look at syntax.  This leads me to ask: what
>>> permits
>>> us to say that "words" are one level, "sentences" are a second
>>> level,
>>> and that "conversations" are a third?  Are we skipping any necessary
>>> levels?  Or creating a level that we shouldn't?
>>> I don't mean to undermine the idea of that different structural
>>> levels
>>> contain, transmit and transform meaning in qualitatively different
>>> ways.  Units smaller than sentences, complete sentences, and
>>> conversations are clearly on very different levels.  My question is:
>>> how can it be demonstrated and explained that these three levels are
>>> the right ones for this job?
>>> It would be very sweet and simple if this were the case ...
>>> - Steve
>>> On Oct 27, 2009, at 8:34 AM, Peter Feigenbaum wrote:
>>>> Greetings, fellow members of XMCA.

Attachment: pic14204.jpg
Description: JPEG image

xmca mailing list