[Xmca-l] Re: Saussure vs Peirce

[David Kellogg]

Martin:

The point I was making was that where Bakhtin says "utterance", Sacks says
"turn" (see Sacks, H. 1992, Lectures on Conversation, Volume Two, Lecture 3
on turn-taking, and also the famous paper with Gail Jefferson, the
"simplest systemics"). In Sacks, "turn" is a physical unit of observable
behavior, discernible to dog's ears. It's still why, when you design a
listening test today, you use a conversation between a female voice and a
male voice. And following that objectivist approach is what led extreme
Conversation Analysis people (e.g. Jonathan Potter) to seriously write
about conversations without minds. Yes, later work had to revise Sacks, and
come up with a different unit. But when I read it (e.g. Heritage, or
Pomerantz, or Paul Seedhourse) I feel like they are trying to re-invent
grammar and thus re-introduce minds.

It seems to me that the distinctions that we want in the last utterance (or
turn, in early CA parlance) is much better understood as a grammatical
distinction and not a distinction between two communicative "acts". Is
Response and Evaluation two acts or one? Is "No, it's not" two acts or one?
Why the comma separating them? What is the communicative difference between
"No", "It's not" and "No, it's not"?. As far as I can see, CA explains none
of this; it even makes it hard to ask about.

A good grammatics can explain it all and much more. Systemic functional
grammar, for example, can also explain the kinds of questions I'm
interested in, e.g. why the Subject comes AFTER the verb in "Is it Chiang
Mai" but before the verb in "Who knows?". This is in interesting question
for me because I have to teach people to ask questions, and it's pretty
hard to do in English.

(You didn't explain the sudden change from South Korea to Thailand. What
happened?)

David Kellogg
Sangmyung University

New Article:
Han Hee Jeung & David Kellogg (2019): A story without SELF: Vygotsky’s
pedology, Bruner’s constructivism and Halliday’s construalism in
understanding narratives by
Korean children, Language and Education, DOI: 10.1080/09500782.2019.1582663
To link to this article: https://doi.org/10.1080/09500782.2019.1582663

Some e-prints available at:
https://www.tandfonline.com/eprint/KHRxrQ4n45t9N2ZHZhQK/full?target=10.1080/09500782.2019.1582663



On Mon, Apr 15, 2019 at 8:00 AM Martin Packer <mpacker@cantab.net> wrote:

> Hi David,
>
> In CA terms, turns are composed of ’turn construction units,’ and TCU is a
> unit of conversation that completes a communicative act. And the utterance
> “No, it is not. Who knows the capital of Thailand?” involves two
> communicative acts. The first is an evaluation of S’s response (“Is it
> Chiang Mai?”) to T’s question ("Who can tell me the capital of South
> Korea?”). The second communicative act is a (indirect) question. In other
> words, T uses their second utterance to do two things: to perform two
> communicative acts.
>
> An utterance is, as you suggest, "the stretch of speaking from one change
> of speaking subjects to another.” S is one speaking subject; T is another.
> (I am assuming that you are not equating ‘subject’ with ‘topic.’)
>
> So sentences are not involved at all. Nor is there any attempt to do away
> with linguistics (that’s to say, grammar).
>
> Or perhaps I am confused by your shifting definitions:
>
> In your first paragraph: For Bakhtin "the utterance is the stretch of
> speaking from one change of speaking subjects to another.”
>
> In your second paragraph, "a turn is pretty much what Bakhtin says it is:
> it's the stretch of speaking we find from one change of speaking subjects
> to another."
>
>
> Ack!
>
> Martin
>
>
>
> On Apr 14, 2019, at 4:14 PM, David Kellogg <dkellogg60@gmail.com> wrote:
>
> So the real source of this realist attempt to do away with linguistics is
> Bakhtin. Bakhtin it was who argued that the sentence is a fiction, and that
> the real, actual, factual unit for understanding text is the utterance. And
> the utterance is defined by the changing of speech subjects. So the
> utterance is the stretch of speaking from one change of speaking subjects
> to another. All this is real, actual, factual (i.e. perceptible with
> feline. canine, infant and foreign hears; as the Russians would say, an
> effect of rech' [speech] and not jazik [language]).The utterance is a unit
> of speech (the real, actual, factual set of things people have uttered),
> but the sentence is only a unit of language (the abstract, imaginary system
> of virtual and potential utterances). This is all in "The Problem of Speech
> Genres" ("Speech Genres and Other Late Essays", U of Texas Press 1986, p.
> 71).
>
> Martin's example (which is an invented one, as we can see from the sudden
> change of venue from South Korea to Thailand) manages to confuse things a
> very great deal. First, he uses "turn" to mean sentence, and "utterance" to
> mean "turn". In CA, which goes for radical empiricism and ontological
> realism, a turn is pretty much what Bakhtin says it is: it's the stretch of
> speaking we find from one change of speaking subjects to another. It's not
> a sentence, and in some data sets less than 11% of the data consists
> of grammatical sentences. By this standard, there are exactly three turns
> in Martin's invented dialogue, and the last turn consists of two somethings.
>
> That "something" is the problem. Bakhtin and CA try to get around this
> problem by saying that there is a virtual change of speakers (Bakhtin) or a
> potential change of speaker (CA) after the question mark. But that doesn't
> explain the question mark itself, or the peculiar order of elements we find
> in the question as opposted to the statement, or the peculiar fact that we
> find the verb before the subject in "Is it Chiang Mai" but the verb after
> the subject in "Who knows...".
>
> The real name of this mysterious "something," the real unit of speech
> communication, is not the "word", unless we assume that Vygotsky meant (as
> Russian allows) the expansive meaning of word (i.e. "In the beginning was
> the word; the word was made flesh; in a word; the last word on this
> question'"). The real unit is the wording: the clause. Clauses can be minor
> ("Hey! Martin!") or they can be elliptical ("You mean...") or they can be
> major clauses like all the examples Martin has invented. But clauses exist,
> even though you need to know the language to recognize them (hence
> languages too must exist, despite what Roy Harris and the integrationalists
> claim). Clauses are just as distinct from the utterances and turns that
> compose them as cities are from the heaps of brick and mortar that we find
> before cities are built and after they have been destroyed.
>
> David Kellogg
> Sangmyung University
>
> New Article:
> Han Hee Jeung & David Kellogg (2019): A story without SELF: Vygotsky’s
> pedology, Bruner’s constructivism and Halliday’s construalism in
> understanding narratives by
> Korean children, Language and Education, DOI: 10.1080/09500782.2019.1582663
> To link to this article: https://doi.org/10.1080/09500782.2019.1582663
>
> Some e-prints available at:
>
> https://www.tandfonline.com/eprint/KHRxrQ4n45t9N2ZHZhQK/full?target=10.1080/09500782.2019.1582663
>
>
>
> On Mon, Apr 15, 2019 at 4:15 AM Martin Packer <mpacker@cantab.net> wrote:
>
>> Hi David,
>>
>> You write: "the fact that an utterance seems to end in the middle of a
>> turn is embarrassing.”
>>
>> But (for CA) it’s not that an utterance can end in the middle of a turn.
>> It’s that a turn can end in the middle of an utterance. That’s to say, an
>> utterance can include more than one turn. Consider:
>>
>> T: Who can tell me the capital of South Korea?
>> S: Is it Chiang Mai?
>> T: No, it is not! Who knows the capital of Thailand?
>>
>> Last utterance by T is composed of two turns. And while there are plenty
>> of games - chess, tennis - where players must alternate turns, taking only
>> one turn at a time, there are plenty of other games where a player can take
>> one turn and then take another.
>>
>> I think you would agree with me that it is not the lack of “theoretical
>> preconceptions” that prevents a dog with headphones from recognizing the
>> units of speech. CA treats as unproblematic (or relatively so) the
>> understanding and transcribing of the sounds and words of a language. It
>> assumes that the researcher has a speaker’s competence in these matters: no
>> one expects someone who speaks only Spanish to conduct CA on English
>> speech. On the other hand, to a native speaker of English the sounds and
>> words are (relatively) “real, actual, and factual.”
>>
>> The (relative) absence of ‘theoretical preconceptions’ in CA refers to a
>> preference to look for, to try to identify, the methods that participants
>> are actually using to carry out their conversation. Based on concrete
>> empirical evidence. Rather than assuming from the outset a theory about how
>> speech is generated and how conversations are structured.
>>
>> So, returning to the larger issue (what the heck is language and how can
>> we study it?), CA approaches language as a concrete practical
>> accomplishment, achieved collaboratively by people whose acquired
>> competence as language speakers amounts to a set of practical skills or
>> methods. That approach will not be to everyone’s taste, but in my view it
>> has plenty to recommend it.
>>
>> Martin
>>
>>
>> On Mar 20, 2019, at 10:27 AM, David Kellogg <dkellogg60@gmail.com> wrote:
>>
>> Martin:
>>
>> From the CA point of view, there isn't anything wrong with it. Like many
>> branches of linguistics (and mathematics), CA has its ducks in a row, and
>> its system is well designed to prevent internal contradictions. For
>> example, CA insists that no theoretical "preconceptions" be brought to the
>> data, and in that sense it is "radically empiricist".
>>
>> I think the problem arises when you try to incorporate concepts from
>> abroad, including Bakhtin. If we say that turns are "real, actual, factual"
>> units ("in the air", as J.J. Gibson used to say--quite incorrectly, as it
>> turns out--of the phoneme), then the fact that an utterance seems to end in
>> the middle of a turn is embarrassing. It undoes the attempt by CA to do an
>> end run around Saussure's notion that the object of study in linguistics is
>> created only by our attitude towards it (that is, we have to understand
>> that something is language before we can study it as language and not
>> simply noise).
>>
>> From a Vygotskyan point of view, this radical empiricism will not do: a
>> dog with headphones could easily segment the "real, actual, factual" turns
>> in data, but not the utterances if we define them by potential turn
>> transition points (or TRPs, or whatever). But it's precisely units that
>> would escape a dog in headphones that make the sound signal into human
>> language, into a meaningful sign, and not simply a signal.
>>
>>
>> David Kellogg
>> Sangmyung University
>>
>> New Article;
>>
>>  David Kellogg (2019) THE STORYTELLER’S TALE: VYGOTSKY’S
>> ‘VRASHCHIVANIYA’, THE ZONE OF PROXIMAL DEVELOPMENT AND ‘INGROWING’ IN THE
>> WEEKEND STORIES OF KOREAN CHILDREN, British Journal of Educational
>> Studies, DOI: 10.1080/00071005.2019.1569200
>> <https://doi.org/10.1080/00071005.2019.1569200>
>>
>>
>> Some e-prints available at:
>>
>>
>> https://www.tandfonline.com/eprint/GSS2cTAVAz2jaRdPIkvj/full?target=10.1080/00071005.2019.1569200
>>
>>
>>
>>
>> On Mon, Mar 18, 2019 at 11:32 PM Martin Packer <mpacker@cantab.net>
>> wrote:
>>
>>>
>>> On Mar 17, 2019, at 11:06 PM, David Kellogg <dkellogg60@gmail.com>
>>> wrote:
>>>
>>>  As soon as you do this, though, you have to admit (and real, actual,
>>> practical data will support this) that there are two such transition
>>> points--not just one--WITHIN your utterance (in addition to the real,
>>> actual, practical turn transition point. .
>>>
>>>
>>> Right: CA refers to this as speaker self-selection. At a TRP (transition
>>> relevant place) the person who has been speaking continues to speak.
>>>
>>> So what’s the problem with that?
>>>
>>> Martin
>>>
>>>
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucsd.edu/pipermail/xmca-l/attachments/20190416/91f68b37/attachment.html