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

[xmca] Language and thought, unity and the particular



David (and anyone else who might be reading),

Sorry for the long delay in responding, two weeks seems like 
an eternity considering the pace of production of ideas on 
this listserve. Hopefully your post isn’t too far in the 
recesses of our collective memory…

I am greatly appreciative of your long and very thoughtful 
response (and no worries about sounding pedantic or 
patronizing - I'm flattered by your engagement with my 
arguments). As you’ll note by the delay in my response, it 
was a lot to chew on. I’ve read through your email a number 
of times in order to do my best to get down to 
the “concepts” that lie beneath the words (to put this in 
terms of a dualism that makes me uncomfortable), and finally 
have been able to put together a response. Yours was a 
wonderfully dense email and I particularly enjoyed the 
literary flourish at the end as you moved from Woolf to 
Austen to Hemingway to Tolstoy to Proust.

Nonetheless, I am as yet unsure that I have quite “gotten” 
the argument that you are putting forward with regard to 
Vygtosky’s thinking since it seems to me to conflict with my 
reading of Vygotsky. To this end, this email is an attempt 
to clarify what it is that you are saying and to critically 
engage with my understanding of the concepts that you (and 
Vygotsky) were proposing.

Let me also apologize in advance for the fact that this 
email surely does not do justice to your ideas and certainly 
lacks a proper tone of deference (and probably demeanor as 
well – but as Goffman notes, this reflects primarily upon 
me). My apologies as well for the fact that this response is 
far too long, but I'm not certain how else to do justice to 
the conversation.

Let me start by stating my position and how it differs from 
yours and then I’ll work through some of the ways I see this 
question/difference manifesting in your email (and I may 
well have misunderstood your email). I view language and 
thought as processes that are intimately caught up with one 
another such that it is difficult to speak of the two as 
being logically distinct. As I read your emails, you seem to 
be staking out a position that is somewhat more comfortable 
in speaking of thinking and language as logically distinct 
things.

To this point, in your email, you wrote: “The categories I 
gave earlier(“relator”, “circumstance”, “process”, “entity”, 
and “qualifier”) are semantic categories and not grammatical 
ones. Not even an “entity” is necessarily a noun; it’s 
perfectly possible to describe an apple with a clause, and 
that is in fact what we do when we define it (“the apple is 
a delicious fruit which appears in late summer on the boughs 
of Malus pumila, the name of a species of Rosaceae”)”. 


I’m still trying to understand what you mean by “semantic 
categories”. This has the feeling of a language/thought 
distinction but it is muddied by the fact that “semantic” 
has a language-y component to it. Do you mean these in the 
strict sense of “word meanings” vs. “grammatical rules”? I 
would like to suggest that grammatical categories can 
contain the kinds of things that you refer to as “semantic 
categories”. What makes a category “semantic”?

I think that this is an important question with regard to 
language and thought and the position that I am staking out 
(namely, that language and thought are not logically 
distinct and separate categories). I don’t think that it is 
necessarily impossible to speak of language and thought as 
separate things, it is obviously a useful way to carve up 
the world. And, of course, there is a pre-linguistic moment 
in which there was something called “thought” separate and 
distinct from language, but we are now well beyond this 
time. My concern is that we too easily forget that language 
and thought are caught up with one other as soon as a child 
is speaking. So, first question – does your semantic vs. 
grammatical categories map onto the categories of thought 
and language respectively? As I understand your position, 
there is some parallel between these two distinctions).

To pursue this point a bit further, I think that I differ in 
my interpretation of chapter 6 (but I’m not sure if I’ve 
quite got your position down). In my reading of Vygotsky, 
the relationship of thinking and speech is more of a 
developmentally dialectical one. Across developmental time 
(both phylogenetic and ontogenetic), thought is brought 
together with speech and both are transformed in the process 
and there are probably a few such inflections of one by the 
other (and eventually these get caught up with other 
dialectical processes like that between spontaneous concepts 
and scientific concepts which, of course, can then transform 
one’s understanding of language). Thus, the infant has 
thoughts prior to speaking, but upon learning to speak, 
those thoughts are transformed into something new. It isn’t 
just that they give voice to something that they were 
previously unable to voice but rather that in giving voice, 
the actual experience of the thing is changed. In your 
Strauss quote, I would want to push a bit further and say 
that it isn’t simply the case that language puts pain into 
words, but rather that pain actually takes on new meaning 
with language. 

Since it’s been a while since your post, I’ve copied three 
paragraphs from your email that seemed to me to be making 
the strong argument for a distinction between thought and 
language - even if the two are “linked”, as you say. (I’ve 
also left the original below in order to see the full 
context).

“In some ways, then, speech is more differentiated, more 
developed than thinking, even though, microgenetically, it 
is more recent (in the same way cultural functions are more 
developed and differentiated than natural functions, and 
science concepts and more developed and  more precisely 
differentiated than everyday ones even though they are more 
recently developed). This by itself is evidence that their 
development is one process, and not two (because if it were 
two distinct processes without any links then we would 
expect that the older process would more differentiated and 
developed). 
  
“But another way to put it (that is, another way of 
formulating the idea thinking and speaking are linked but 
distinct) is more structural and less genetic. It is to say 
that not everything we find in thinking is present in 
speaking, and not everything to be found in speaking is 
present in thinking. As soon as we put it this way, it 
becomes clear that we need to distinguish between the 
properties of language reified as a formal system for 
producing actual utterances(what Vygotsky calls its “phasal” 
properties, which is incorrectly translated 
as “phonological”) on the one hand, and the properties of 
the interface with thinking (what Vygotsky calls 
its “semantic” properties, which includes the volitional-
affective substratum of speech). 

“The semantic properties are not entirely present in 
speaking, for some of them have evaporated (“volatilization” 
is the term Vygotsky uses) in the process of speaking them. 
The former, phasal, properties are not entirely present in 
thinking, which often appears to come over us “all at once” 
rather than in a linear, left to right, progression governed 
by the rules of syntax. Thoughts occur to us, while 
utterances must be articulated.” 

Where I differ is in the sense in which these language and 
thinking are distinct. I’m pushing for a much tighter 
connection between the two such that it is very difficult to 
describe conceptual thinking without considering language as 
well. 

There were two points that you made about the “phasal” 
vs. “semantic” properties of language that did not square 
with my read of Vygotsky’s Ch. 6 of thought and language. 
First was the notion that the distinction between “phasal” 
and “semantic” properties of language is central to 
understanding the development of concepts in that the phasal 
properties of language map onto syntax and the semantic 
properties of language map onto thinking. Second was the 
notion that “phasal” and “semantic” properties are learned 
in fundamentally different ways. I think that both of these 
draw a distinction between language and thought in ways that 
doesn’t seem to me to jive with my understanding of 
Vygotsky’s project.

To this first point, as I read this distinction 
between “phasal” and “semantic” in Vygotsky, it seems that 
this distinction (discussed on pp. 196-197 of Kozulin’s 
translation, that is, assuming I have the right section) is 
used simply as part of an analogy for understanding the move 
from spontaneous to scientific concepts, and not as a 
distinction relevant to the thinking vs. speech distinction. 
Thus, I don’t read the “phasal” vs. “semantic” distinction 
as having such a central role in the theme of the chapter 
(which is about the acquisition of scientific concepts). The 
phasal vs. semantic distinction appears to me to be a 
distinction that he is making between the two sides of his 
analogy of the learning of scientific concepts with learning 
a second language (first language:second 
language::spontaneous concepts:scientific concepts). 
Vygotsky does note important differences between the two:

“However, while in the study of a foreign language attention 
centers on the exterior, phonetic, and physical aspects of 
verbal thought, in the development of scientific concepts it 
centers on semantics. And since physical and semantic aspect 
of speech develop along their own independent lines, our 
analogy cannot be a complete one. The two developmental 
processes follow separate, though similar, paths” (p. 196).

But as I read the above passage, Vygotsky is simply noting a 
difference between the two latter parts of the analogy – 
learning a second language is like learning a second 
language EXCEPT for the fact that the former centers on 
phasal aspects of verbal thought and the latter centers on 
semantics. This doesn’t suggest to me that the phasal and 
the sematic aspects are learned in different ways, just that 
they develop “along different lines”. The point for me, is 
that second language learning focuses on a different aspect 
of verbal thought. Nothing is said about the nature of 
acquisition of first language “phasal” properties as 
compared to first language “semantic” properties – more on 
this below.

But there is also another important difference. Vygotsky 
writes:
“But there is also an important difference between these two 
processes [of learning a second language and learning 
scientific concepts]. In the case of language study, the 
native language serves as an already established system of 
meanings. In the acquisition of scientific concepts, the 
*system* must be built simultaneously with their 
development. The concept of system organization thus becomes 
a crucial one” (p. 197).

It seems like the interesting and important point that he is 
making is about the development of scientific concepts is 
the importance of the relationship of the part to the whole. 

Thus, I see the phasal vs. semantic distinction not as a 
distinction between language and thought (or 
phonetic/grammatical vs. semantic) but rather as a 
distinction between part and whole. Vygotsky’s point about 
the emphasis on the phasal properties of language seems to 
me to be about how, for the second language learner, the 
part is dis-connected from the whole. This differs from the 
first language learner who learns language (even it’s phasal 
aspects) in a purely structural way. In the case of the 
second language learner, she does not learn the new word 
with respect to the whole of the language. Instead, she 
learns it as a part of and with respect to a different 
whole, namely, her first language, and this leads to an 
emphasis on the “phasal” aspects of language. E.g, for an 
English speaker learning French, the French word “boeuf“ is 
understood as the same thing as the English “beef” when 
considered only in its part-ness and only is understood as 
being different when it becomes a part of a system and the 
speaker understands that French “boeuf” refers to “beef” 
and “ox”. And one would want to add other distinctions such 
as “sheep” vs. “mutton” as opposed to the single French 
category “mouton.” And then a larger structure emerges here 
in which the eating variety of the thing is a French 
equivalent (boeuf>beef, mouton>mutton) whereas the word for 
the living beasts have Germanic roots – and voila! we soon 
discover the (cultural) concept – “French is good to eat”, 
non? (and German is beastly?).

[Note: I draw on Saussure’s examples in order to again make 
the point once again that Vygotsky and Saussure are both 
drawing on the Hegelian-cum-Marxian (in the Grundrisse) 
structuralist notion of locating the origins of meaning of 
any part with respect to the whole. This doesn’t mean that 
the two ideologies of language necessarily share identity, 
simply that there is a great affinity in this regard.]

Now, returning the point that I would push with regard to 
the distinction between the phasal/grammatical and the 
semantic is that Vygotsky seems to me to be pointing to the 
fact that both of these are similar in that they are 
structural – in both cases the unity is necessary for 
understanding the part (hence the analogy between language 
learning and learning of scientific concepts). This suggests 
that the phasal/grammatical properties of language have 
an “all at once” quality as well as the semantic ones. One 
last turn to Whorf is helpful to make this point. Whorf 
nicely articulates the importance of grammatical wholes in 
his notion of “grammatical categories” (for an example see 
my earlier post about languages with classifiers vs. 
articles and how this can encode for different processes 
of “entification” – one of the “semantic categories” you 
have proposed). Thus, grammar contains categories and 
concepts just as semantics do (and you may have been in 
agreement with me on this). And importantly, the category 
(and the concept behind it) can only be seen with respect to 
the larger system in which the grammatical category exists. 
Thus, what I’m getting at is further evidence for how it is 
that the “phasal” and the “semantic” share identity in the 
sense that both require an understanding of the relationship 
of the part to the whole in order to “get” it. The phasal 
(grammatical) aspects of language may have a linearity in 
usage, but this does not mean that it is learned linearly 
(learning a second language can have this linearity, but 
this is precisely what makes second language learning so 
difficult, we try to incorporate the second language into 
our first rather than allowing the parts to be defined 
relative to the other parts). 

Extending this to the argument about the learning of 
scientific concepts – Vygotsky’s point seems to me to be 
that the acquisition of the scientific concepts critically 
changes the nature of the spontaneous concept (which can now 
be seen “scientifically”). What I am suggesting is that this 
is the argument that Vygotsky is making with respect to 
thought and language. The acquisition of language transforms 
thought and that these new possibilities for thinking (e.g., 
concepts, and later scientific concepts) transforms the 
nature of language.

What I’m not understanding is how you are reading Vygotsky 
as splitting apart the “phasal” and the “semantic”. It seems 
that these share an identity in precisely the sense that you 
are saying that they are different. That is, they are 
similar in that they both contain the principle of part to 
whole relationship. In order to really “get” a grammatical 
or phonological aspect of language, you have to have the 
whole just like in order to really “get” a “semantic” aspect 
of language, you have to have the whole. It seems that both 
semantic and phasal learning have an “all at once” quality 
to them. This is in contrast to the argument that I took you 
to be making that the phasal involves part-like learning and 
the semantic involves more holistic learning. 

Finally, as for the reading list that you suggested, I read 
some Halliday back in the day and recall that his way of 
breaking up the world was not productive for how I’ve been 
thinking about things. But I’m willing to give him another 
shot – did you have a piece that you might recommend?
I’d also be interested in a copy of your Lim and Kellogg 
(2009) paper, I looked for it in all of the 2009 editions of 
the journal Language and Education but couldn’t find it (and 
for some reason my library doesn’t have access to these 
articles anyways), so if you’re willing to share, I’d be 
much obliged.

Again, I do sincerely appreciate your strong engagement with 
my rather lengthy emails. I find this conversation to be 
incredibly productive and helpful for me to better 
understand how to make sense of the relationship of language 
and thought, and how all of this relates to learning and 
development.

Looking forward to continued conversation,
greg


>------------------------------------------------------------
----------
>
>Message: 1
>Date: Thu, 13 Aug 2009 22:42:42 -0700 (PDT)
>From: David Kellogg <vaughndogblack@yahoo.com>
>Subject: Re: [xmca] Vygotsky, Saussure, and Wolves with 
different
>	dreams
>To: Culture ActivityeXtended Mind <xmca@weber.ucsd.edu>
>Message-ID: <72305.15901.qm@web110303.mail.gq1.yahoo.com>
>Content-Type: text/plain; charset=utf-8
>
>Greg,
> 
>Thanks for your letter, and let me first of all excuse 
myself for the pedantic quality (bordering on the downright 
patronizing) of this posting. The relentless didacticism is 
not for a moment directed towards you personally, but rather 
towards my pitiable (and as yet unknown) graduate students. 
Classes start in about a week, and I am trying to make my 
ideas as clear as possible in order to write them down in an 
in-house textbook I am finishing up. 
> 
>As I said, Greg, Chapter Six of Thinking and Speech, is all 
about non-noun concepts: “becauseâ€・ and “althoughâ€・. 
Vygotsky, and Mandelstam before him, clearly says that a 
word meaning is not a thing; it’s a process. I’m afraid 
there isn’t any other way to interpret Paula and Carol’s 
work. The concept is not in the blocks (that was Ach’s 
mistake); it’s in the process of sorting them. This is 
every bit as true of concepts as it is of heaps and 
complexes.
> 
>The categories I gave earlier (“relatorâ€・, 
“circumstanceâ€・, “processâ€・, “entityâ€・, and 
“qualifierâ€・) are semantic categories and not grammatical 
ones. Not even an “entityâ€・ is necessarily a noun; it’s 
perfectly possible to describe an apple with a clause, and 
that is in fact what we do when we define it (“the apple 
is a delicious fruit which appears in late summer on the 
boughs of Malus pumila, the name of a species of 
Rosaceaeâ€・). 
> 
>It’s just that in English the end state of most 
logogenesis about apples is nominal. So the canonical, 
default realization  of an entity like an apple is a noun. 
If English were a language spoken entirely by apple farmers 
who for some reason exported their entire crop, we would 
probably consider the root concept of “appleâ€・ a verb, 
since we would spend our days talking of how to apple or 
perhaps going appling or sitting around applizing.
> 
>I have read and I heartily approve your reading list 
(Klein, Karmiloff-Smith, etc.) Perhaps I might add to the 
list Halliday, who gave me the categories of “relatorâ€・, 
“circumstanceâ€・, “processâ€・, “entityâ€・, etc.) and 
of course Wallace Chafe (“Discourse in Timeâ€・). There is 
also my own work, especially Lim and Kellogg 2009, in 
Language and Education.
> 
>I don’t insist on the statement that articles inflect for 
time; the notions of “inflectionâ€・ and “timeâ€・ imply 
a formalist approach to language that is quite alien to the 
way I really think. I was trying to make a point about your 
taxonomy, which I find rather formalistic for my own use. In 
the classroom, where I work, articles are important for 
conveying (or, if you like, indexing) concepts. I don’t 
think that they “createâ€・ concepts any more than I think 
that tense creates time. 
> 
>I think that everything you have said about articles can be 
used to demonstrate that verbs do not inflect for time 
either (e.g. “It is high time you did your homework.â€・) 
Here too I am agreeable; we can say that verbs changes are 
called tense and not time and that tense is a purely formal 
category with only a passing acquaintance (a notional 
relationship) with the semantic category of time. Chinese 
works very well without tense, and so, for that matter, does 
English (e.g. when you have to tell a story backwards, from 
effect to cause, you do not use tense to do this). 
> 
>As I said, I think most languages are superior to English 
for the purpose of conveying concepts outside a very narrow 
range of peculiarly English ones (e.g. “Misterâ€・). But in 
general that is true of all languages; a language gets good 
at the range of activities that its speakers indulge in, and 
no language community indulges in the whole range of human 
experience. The difference is that English has apparently 
been “chosenâ€・ as our world language. As I said, I think 
it has been so because of its anti-egalitarian and 
ultimately anti-educational qualities; languages of prestige 
and power are invariable impractical in obvious ways.
> 
>Let me talk a moment about why I think it’s important to 
distinguish between the formal properties of language, which 
are expressed in linear ways, and the semantic ones, which I 
think are not so (paradoxically, time is not expressed in 
language through time; only tense is expressed through time).
> 
>The Vygotskyan position is that thinking and speech are 
linked, but distinct. One way to put this (Vygotsky’s way) 
is to say that thinking is not expressed in speech but 
realized in speech. As Richard Strauss wrote:
> 
>Flamand: The infant’s cry of pain came before language!
>Olivier: But only words can explain what pain means.
> 
>In some ways, then, speech is more differentiated, more 
developed than thinking, even though, microgenetically, it 
is more recent (in the same way cultural functions are more 
developed and differentiated than natural functions, and 
science concepts and more developed and  more precisely 
differentiated than everyday ones even though they are more 
recently developed). This by itself is evidence that their 
development is one process, and not two (because if it were 
two distinct processes without any links then we would 
expect that the older process would more differentiated and 
developed).
> 
>But another way to put it (that is, another way of 
formulating the idea thinking and speaking are linked but 
distinct) is more structural and less genetic. It is to say 
that not everything we find in thinking is present in 
speaking, and not everything to be found in speaking is 
present in thinking. As soon as we put it this way, it 
becomes clear that we need to distinguish between the 
properties of language reified as a formal system for 
producing actual utterances(what Vygotsky calls its 
“phasalâ€・ properties, which is incorrectly translated as 
“phonologicalâ€・) on the one hand, and the properties of 
the interface with thinking (what Vygotsky calls its 
“semanticâ€・ properties, which includes the volitional-
affective substratum of speech). 
> 
>The semantic properties are not entirely present in 
speaking, for some of them have evaporated 
(“volatilizationâ€・ is the term Vygotsky uses) in the 
process of speaking them. The former, phasal, properties are 
not entirely present in thinking, which often appears to 
come over us “all at onceâ€・ rather than in a linear, left 
to right, progression governed by the rules of syntax. 
Thoughts occur to us, while utterances must be articulated.
> 
>But as Goldwin-Meadow says, language is resilient stuff. 
Almost everything in the human experience (and then some) 
can make it into language as one way or another, and so it 
is possible to semanticize its phasal properties just as it 
is possible to render more phasal its semantic properties. 
When we do this, however, we often find ourselves s This was 
(I think) the great discovery of Jane Austen and other women 
English novelists (e.g. George Eliot and Elizabeth Gaskell) 
at the beginning of the nineteenth century. 
> 
>Virginia Woolf notes somewhere in “A Room of Her Ownâ€・ 
that the typical sentence at the end of the eighteenth 
century was something like:
> 
>"The grandeur of their works was an argument with them, not 
to stop short, but to proceed. They could have no higher 
excitement or satisfaction than in the exercise of their art 
and endless generation of truth and beauty. Success prompts 
to exertion; and habit facilitates success". 
> 
>You can see that the phasal structure of these sentences is 
actually quite simple: “A was B (not to C but to D). They 
had no E but F. G leads to H and H leads to more G.â€・
>You can also see that it is noun-filled (A through G) and 
static, and that it literally winds up in a circle (from G 
back to more G, which is really A, B, D, and F in disguise).
> 
>Woolf compares this with what she calls Jane Austen’s 
“shapely sentencesâ€・. She doesn’t actually tell us what 
is shapely about an Austenian sentence, but most Austen fans 
(and I am one) can tell you anyway: Austen’s sentences 
unfold a character’s feelings as thoughts, and thoughts as 
words, often in the form of a long, undulating relative 
clause or series of relative clauses. Thinking is a form of 
feeling in Austen, and both are realized in speech.
> 
>>From this alone we can see that a single text has several 
different semantic layers: mental, verbal, material. The 
organization of these layers is linked but it is also 
distinct, and when we penetrate the bottom layer the top 
layer may disappear from view. For example, if we read the 
Arabian Nights as a single utterance, we are left with a 
story about a lady who for some mysterious reason is worried 
about having her head cut off: Aladdin, Sinbad, and Ali Baba 
appear only as delaying tactics or bad jokes told on the 
scaffold. When we decompose it into its component tales (as 
it must have been composed), it is Scheherazade who 
virtually disappears. 
> 
>Mike is quite right to say that xmca threads are complexive 
and sometimes even syncretic in their structure. But when we 
decompose them into their components, we discover that the 
things that are being grouped together complexively and 
syncretically are not objects at all, but rather concepts 
(“influenceâ€・, for example, which Vygotsky reminds us 
means “in flowingâ€・). 
> 
>Hemingway, a writer whose psychology is almost entirely 
behaviorist, liked to begin stories in the way you describe, 
e.g. 
> 
>“He came back into the room. The body was still on the 
floor.â€・ 
> 
>The writer pretends to treat the reader as an insider, 
privy to direct, unmediated, visual-graphic information. 
> 
>The reader, excruciatingly conscious of his outsider 
status, has to read forward in order to catch up on what he 
has been missing. But Hemingway is a small boy playing with 
words; his technique is only a clumsy grammaticization of 
the semantic trope of Homer known as “in res mediaâ€・. He 
only succeeds in conveying the rather phoney and 
characteristically macho idea that somehow unmediated 
experience is truer, and if you gotta ask you ain’t never 
gonna know. 
> 
>Why did Hemingway, who so admired Tolstoy, learn so little 
from him? Hemingway apparently thought that he was 
rejuvenating writing by bringing it back to the simplest 
experiences, e.g. death. But of course death is NOT simple 
to write about and of course it is one of the few 
experiences about which we write that we CANNOT undergo 
first hand. One of the few writers who tried to do this was 
Marcel Proust, who apparently directed that his jottings 
from his own deathbed struggles should be incorporated into 
“A la recherche du temps perduâ€・. (Years later, another 
French writer, whose name escapes me, tried to repeat the 
experiment while he was dying of AIDS, and the only sentence 
I can remember from it is “J’ai retrouve mes muscles 
d’enfant.â€・)
> 
>Unlike Hemingway, Proust really does rejuvenate writing by 
bringing it back to the simplest experiences, but they are 
the most readily experiential experiences, like trying to 
fall asleep and eating a morsel of cake dipped in tea and 
then trying to remember something that you have forgotten 
but not forgotten about. 
> 
>And here, with the distinction between remembering and 
remembering about, we really are quite close to the 
distinction Vygotsky makes in Chapter Six between the 
“phasalâ€・ (phonological but also grammatical) properties 
of language and the semantic (volitional-affective, but also 
pragmatic). Articles are clearly of the former, but concepts 
partake of the latter.
> 
>David Kellogg
>Seoul National University of Education
>
---------------------------------------
Greg Thompson
Ph.D. Candidate
The Department of Comparative Human Development
The University of Chicago
_______________________________________________
xmca mailing list
xmca@weber.ucsd.edu
http://dss.ucsd.edu/mailman/listinfo/xmca