nl-kr-request@CS.ROCHESTER.EDU (NL-KR Moderator Brad Miller) (12/16/88)
NL-KR Digest (12/15/88 19:57:13) Volume 5 Number 40
Today's Topics:
sense of wholeness of one's surroundings
A case for the study of speech behavior (A joint experiment)
Chinese dictionary stack
Re: information, message, intention, and the like
Submissions: NL-KR@CS.ROCHESTER.EDU
Requests, policy: NL-KR-REQUEST@CS.ROCHESTER.EDU
----------------------------------------------------------------------
Date: Wed, 14 Dec 88 11:40 EST
From: VENTURA%21514@atc.bendix.com
Subject: sense of wholeness of one's surroundings
Responding to Roland Zito-wolf's message:
> I am looking for references to work dealing with the way that the
> human mind manages to maintain the impression of being aware of one's
> surroundings as a whole, in spite of the fact that the visual system
> (and other senses) can only attend to a small portion (aspect) of it
> at any one time. [....]
> A related question is: How is it that we maintain the impression of
> perceiving (or recalling, or imagining) our surroundings in great
> detail even when we've given them but the slightest glance-- as when
> we enter a room with which we are familiar. We know the texture and
> color of the walls, the contents of the room, and so forth, at some
> half-conscious level, without actually paying them much attention. We
> function quite adequately using only these "general" impressions.
One name for the idea you are describing is a "fish-eye view". The idea is:
what the mind uses as a working representation of surroundings is a skeleton -
a sketchy outline - of the full detail it actually possesses about them.
Points in the outline are fleshed out in greater detail if attention is
focussed on them. The amount of detail is proportionate to the intensity of
attention directed at that point: there is greater and greater detail the
more intensely you scrutinize a point. The idea of a working representation
containing little detail except at points of focus is where the term "fish-eye
view" came from.
I have only heard fish-eye views mentioned regarding visual "surroundings".
Also, Barbara Grosz has written on focussing in dialog and I think that also
bears on the problem - in this case regarding language and communications
"surroundings".
I am very interested in this topic. I'd appreciate it if you would post or
email me your findings.
C.Ventura
ventura@atc.bendix.com
------------------------------
Date: Thu, 1 Dec 88 07:08 EST
From: Celso Alvarez <sp299-ad@violet.berkeley.edu>
Subject: A case for the study of speech behavior (A joint experiment)
In article <178@gollum.UUCP> rolandi@gollum.UUCP (Walter G. Rolandi) writes:
>In response to Celso's:
>>In fact, I propose to the net the following joint experiment...
>
>While of limited experimental value, this sounds like it might be fun.
>I'm game.
Experimental value? Donnow. Theory-testing value, a lot.
But dis no fun. Dis serious'tuff, like the rest of our discussions.
Here it goes:
--------------------------------------------------------------------
The following is the transcription of an actual verbal exchange.
It is open for analysis or comment from any approach to the study of
speech, language or communication.
THE CASE
EXCHANGE: The initial moves of a phone conversation.
PARTICIPANTS: Albert (A, age 28) and Betty (B, age 30).
SETTING: Locale: two apartments in the same town
["environmental variables" #1 and #2? :-)]
Time: around 7:00 p.m.
SOME TRANSCRIPTION CONVENTIONS (if there's enough interest, I'll keep
reviewing the tape for more detailed transcription)
? rising tone at the end of intonation group
/ falling tone at the end of intonation group
, sustained tone at the end of intonation group
v fall-rise tone over stressed syllable
[ ] beginning and end of voice
[ ]
overlap
: sound lengthening
- false start, sound interrupt-
' sound(s) (presumably) ellided
* piano (lenis, soft) enunciation
CAPS extra PROminent syllable (extra word accent)
.. shorter pause (less than .5 sec. approximately)
... longer pause (between .5 sec. and 1 sec. approximately)
> < relatively >faster talk<
hh he laughs (vocalized aspiration)
(Lines are numbered)
THE EXCHANGE:
1 A: hello ? ... mh /
[ ]
2 B: yeah / albert ?
3 A: yeap
4 B: how come you don't answer the phone like a normal human BEing /
5 A: wh::at ' you meanhh ?
6 B: *hh *hh *hh
7 A: no / I th- I took the the the the wrong way you know ,
8 B: oh v v
9 A: I- I took it upside down mh mh ... really , really /
[ ]
10 B: he he he
11 B: u::h ... >did you git a hold of 'im< ?
12 A: yeah / he's uh he's at the departmen' so he's coming now /
13 B: oh ok /
14 A: he's yus' walking /
15 B: oh ok
16 A: *yeah /
17 B: the other thing is , .. do you have an onion ?
.. . .
QUESTIONS (Choose only one) (10 points):
a) Apply a causal model of verbal behavior to the analysis of the
previous exchange. Specifically, explain, in line 17, the utterance
"do you have an onion ?"
b) On the basis of the previous exchange, argue convincingly AGAINST the
feasibility of a causal model of verbal behavior. Specifically,
refrain from referring to line 17, "do you have an onion?" :-)
----------------------------------------------------------------------
Celso Alvarez
sp299-ad@violet.berkeley.edu
------------------------------
Date: Wed, 14 Dec 88 18:28 EST
From: Ed Lai <lai@Apple.COM>
Subject: Chinese dictionary stack
This is to announce the release of another version of the HyperCard
Chinese Dictionary stack - version 0.2H. As before the stack needs
to be run under a Chinese Macintosh operating system (Taiwan version
or any system that uses the big-5 code). No new release is available
for the simplified character version. System requirement is hard
disk and at least 2M of RAM, 4M or more RAM preferred. It works with
1.2.1 or earlier version of HyperCard. This stack and version 0.1
stacks do not work with HyperCard 1.2.2. Later I shall try to
release version 0.3 which will work with HyperCard 1.2.2 but will no
longer be compatiable with 1.2.1 or earlier.
The reason for the new release is the availability of data from the
CCAG (Chinese Character Analysis Group) in Taiwan. This enables me
to extend the dictionary from 5000+ characters to 13000+ characters,
covering all the characters in the Chinese Macintosh system. And
that is why the release is for the traditional character set only.
Information about each character now includes 5 Romanization schemes
- BoPoMoFo, PinYin, JuiYin 2nd Form, Wade-Gile and Yale; telegraph
code, 4-corner code, 3-corner code, TsangChi code, CCCII code and
Big-5 code. There is also the Cantonese spelling for a subset of the
characters. Of course most of these information are more for the
interest of those interested in Chinese computing, the language
learner will find this stack to be an overkill and should stick with
the earlier version which has less hardware requirement. The
structure of the stack is very similar to the last version. The
major improvement is in the search of character by sound, since
index table is now available for all 5 Romanization scheme mentioned
earlier. Another interesting alternative to the sound table is now
available. In stead of a list of all the sound Romanization, you can
have a matrix of the consonants and vowels, with a character with
that consonant and vowel combination being listed as a matrix
element.
As before the stack is available free to anyone who want it but you
are responsible to getting your own copy of a Chinese operating
system (not necessarily Apple's). Since the final stack is close to
3M you need to send me 3 (THREE) 800K disks and SASE to the address
below. You need also know how to join the files and decompress it
using StuffIt. If you want to get it by E-mail, I am willing to do
it once and only once. Send me mail and I shall send it to everyone
at the same time in two weeks. If for some reason you miss the
deadline or cannot receive the file you have to send me the disks as
I am not going to do it a second time
This release is possible
because the CCAG is willing to share the data they have collected.
So once again I would like to appeal for machine readable data that
you are willing to share. In particular I am looking for Chinese
phrases and the English translation, break down of the stroke count
into strokes types in writing order, break down of characters into
sub-components, big-5 to GB code conversion table, and the
equivalent of the CCAG style database for simplified characters.
The address to get the stack is
Edmund Lai
946 San Tomas Aquino Road
Campbell, CA 95008
/* Disclaimer: All statments and opinions expressed are my own */
/* Edmund K. Lai */
/* Apple Computer, MS42-C */
/* 20525 Mariani Ave, */
/* Cupertino, CA 95014 */
/* (408)974-6272 */
------------------------------
Date: Thu, 1 Dec 88 05:43 EST
From: Celso Alvarez <sp299-ad@violet.berkeley.edu>
Subject: Re: information, message, intention, and the like
In article <8876@bcsaic.UUCP> rwojcik@bcsaic.UUCP (Rick Wojcik) writes:
RW>In article <17389@agate.BERKELEY.EDU> sp299-ad@violet.berkeley.edu
RW>(Celso Alvarez) writes:
CA>I think that Rolandi is clear (and right) when he states that context
CA>("environmental variables") is directly correlated to "the probability
CA>[of occurrence] of an utterance or class of utterances" (not verbatim).
RW>Sorry, Celso, but I didn't catch your arguments for this proposition. Can
RW>you point to any literature where "an utterance or class of utterances" is
RW>assigned a probability of occurrence based on some correlation with
RW>environmental variables?
^^^^^^^^^^^^^^^^^^^^^^^
I should first say that I didn't phrase the question exactly like you do,
nor (I realize now) did my posting reflect exactly what I meant (my fault).
As I said, my approach to the study of speech behavior is not correlational,
but interpretive. Anyway, since I asked for it, I'll try to answer.
Regarding "environmental variables", no, I don't know of such literature.
I have already expressed my reservations about that concept, and I still am
not sure what is meant by it. Perhaps Rolandi could help.
As for *situational* variables (setting, participants, topic, etc.), that's
another matter. As you know, there is correlational work associating
situational factors with the production of specific turns or utterances
(e.g. the work by West on gender-based interruption patterns, the work by
Brown & Levinson on the use of utterances with T/V pronouns, work on
gender-based differences in the use of directives, etc.) which establish
observed *frequencies* of occurrence of those utterances. But, as you also
know, the issue has been raised (at least in variation theory) whether observed
frequencies correspond to actual probabilities of occurrence (I'm referring to
the debate on the pages of _Language_ between Sankoff, Labov, Kay, McDaniel,
and Cedergren on the nature of variable rules). I'll leave it there.
This is what I had in mind, and what I thought Rolandi had in mind, too.
Perhaps I was mistaken.
>I'm not too sure what you mean when you say that active and
>passive sentences are 'propositionally equivalent'. There are semantic
>differences between active and passive sentences.
In speech act theory, the propositional content of an utterance refers to
its truth value, regardless of illocutionary force and surface form.
"Joe was hit by a car" and "A car hit Joe" (and also "Was Joe hit by a car?")
is the same: agent (car)-action (hit)-object (Joe). I assume that one
semantic difference between active and passive constructions is the
foregrounding of information.
Celso Alvarez
sp299-ad@violet.berkeley.edu
------------------------------
Date: Fri, 2 Dec 88 13:50 EST
From: Rick Wojcik <rwojcik@bcsaic.UUCP>
Subject: Re: information, message, intention, and the like
In article <179@gollum.UUCP> rolandi@gollum.UUCP (Walter G. Rolandi) writes:
>I doubt many behaviorists would claim to understand active/passive
>switching as a general phenomenon. In behavior analysis, one typically
>investigates one behaving organism at a time. The covariants of a
>particular speaker's active/passive switching could most likely
>be experimentally determined.
I seem to have lost sight of what your original complaint with linguists was.
You want us to adopt your 'scientific' methodology, but you tell us that we
can't even talk about the most interesting and obvious linguistic phenomena.
We can't make generalizations about mental states because you're not sure they
exist (or can be defined in such a way that they make sense) in your
framework. For some reason, English speakers choose to use active and passive
sentences. I grant you that some modern syntacticians avoid the question of
how syntactic rules correlate with language use. But, as far as I can tell,
so do you. Some linguists can't see language use, but you can't see language
rules.
--
Rick Wojcik csnet: rwojcik@boeing.com
uucp: uw-beaver!ssc-vax!bcsaic!rwojcik
------------------------------
Date: Mon, 5 Dec 88 20:34 EST
From: w.rolandi <rolandi@gollum.UUCP>
Subject: Re: information, message, intention, and the like
In response to Rick's:
>I seem to have lost sight of what your original complaint with linguists was.
>You want us to adopt your 'scientific' methodology, but you tell us that we
>can't even talk about the most interesting and obvious linguistic phenomena.
By this, I assume you are referring to rules.
>I grant you that some modern syntacticians avoid the question of
>how syntactic rules correlate with language use. But, as far as I can tell,
>so do you. Some linguists can't see language use, but you can't see language
>rules.
If I cannot see rules, it is because my view is obscured by the patterns
of behavior that suggest them. If one can determine an external cause
for a behavior, what explanatory value does the construct of "rule" add
to that understanding?
Walter Rolandi
rolandi@ncrcae.Columbia.NCR.COM
NCR Advanced Systems Development, Columbia, SC
------------------------------
Date: Fri, 9 Dec 88 17:14 EST
From: Rick Wojcik <rwojcik@bcsaic.UUCP>
Subject: Re: information, message, intention, and the like
In article <183@gollum.UUCP> rolandi@gollum.UUCP (Walter G. Rolandi) writes:
>If I cannot see rules, it is because my view is obscured by the patterns
>of behavior that suggest them. If one can determine an external cause
>for a behavior, what explanatory value does the construct of "rule" add
>to that understanding?
If I understand your position correctly, behaviors don't correspond to
linguistic rules. At least, you haven't shown that they do. So your view
cannot be obscured by "the patterns of behavior that suggest them." If such
patterns existed for you, then I am sure you would point them out for us. It
follows that the answer to your last question is "nothing". Behaviors can't
be defined in such a way that they correspond to linguistic structures or
rules. And I'm still puzzled as to what you think your methodology adds to
the study of language. You seem to think that progress is made if you can
define a problem in such a way that it can't be addressed. In this case, the
question is why do speakers use active and passive sentence structures?
--
Rick Wojcik csnet: rwojcik@boeing.com
uucp: uw-beaver!ssc-vax!bcsaic!rwojcik
------------------------------
Date: Wed, 14 Dec 88 08:57 EST
From: wgr <rolandi@gollum.UUCP>
Subject: Re: information, message, intention, and the like
In response to Rick's:
>If I understand your position correctly, behaviors don't correspond to
>linguistic rules. At least, you haven't shown that they do. So your view
>cannot be obscured by "the patterns of behavior that suggest them." If such
>patterns existed for you, then I am sure you would point them out for us. It
>follows that the answer to your last question is "nothing". Behaviors can't
>be defined in such a way that they correspond to linguistic structures or
>rules.
I cannot tell from your response if you understood me or not. It doesn't
matter, it's my fault for speaking so figuratively. Here is what I meant:
I am less interested in linguistic rules than I am interested in actual
verbal behavior. Verbal behavior is observable and measurable and therefore
available for scientific analysis. On the other hand, we merely infer the
existence of linguistic rules, ironically, from verbal behavior. My point
is that if one can demonstrate a causal understanding of verbal behavior in
terms of external environmental variables, what does the internal construct
of linguistic rules add to that understanding? (It adds nothing.)
Obviously, language is rule-like in nature. The question is really whether
the rules of verbal behavior are causes or effects. Syntacticians seem to
think that we speak the way we do BECAUSE of rules. I have yet to hear
where rules exist and exactly how they influence our behavior. It would
seem that they exist somewhere within us and exercise agency over our
utterances.
>And I'm still puzzled as to what you think your methodology adds to
>the study of language.
A scientific basis for the analysis of verbal behavior.
>You seem to think that progress is made if you can
>define a problem in such a way that it can't be addressed.
I accept this as a sincere concern on your part and I think I can
understand why you would believe this to be true. While I would agree
that there is no progress in defining problems in ways that make them
inaccessible to science, the fact remains that many problems are
simply prescientific. How many angels can dance on a pinhead? Where
are linguistic rules physically located and what is the mechanism of
their effect on verbal behavior?
Walter Rolandi
rolandi@ncrcae.Columbia.NCR.COM
NCR Advanced Systems Development
Columbia, SC USA
------------------------------
Date: Thu, 15 Dec 88 06:11 EST
From: Clay M Bond <bondc@iuvax.cs.indiana.edu>
Subject: Re: information, message, intention, and the like
Walter Rolandi:
> Obviously, language is rule-like in nature.
How about changing this to "Obviously, language is, to a
certain point, easily explained by the mathematical mani-
pulation of symbols," with the emphasis on "to a certain
point."
> I have yet to hear where rules exist and exactly how
> they influence our behavior.
You and me both, Walter. We do, in fact, share a number
of concerns about current (and past) linguistic analysis.
Where we part, I think, is that you are assuming that
these "rules" are a function of some mental construct
("mind" for the sake of familiarity), and are therefore
not scientifically documentable. As far as symbolic,
mentalist linguistic theory is concerned, I agree with
you. However, I don't think rules exist anywhere save
as explanatory constructs, and that cognition (which you
apparently consign to the realm of mentalism) can indeed
be empirically studied and analyzed as firing patterns
among neural networks.
You can't talk about verbal behavior, Walter, without
taking cognition into account, and as long as one's
behaviorism (call me a neo-behaviorist, if you will)
is incapable of studying cognition, then it is as faulty
as mentalism, sort of a complementary distribution.
Take a gander at Rumelhart and McClelland's _Parallel
Distributed Processing_ and see what you think. No, there
lots of things we *can't* do now, but we're fledglings.
We don't assume non-empirical entities such as "mind",
and work with the brain instead.
Consider it an invitation to at least consider hopping
on in, Walter -- the water's fine.
--
<< Clay Bond Indiana University Department of Leather, uh, Linguistics >>
<< bondc@iuvax.cs.indiana.edu AKA: Le Nouveau Marquis de Sade >>
<< {pur-ee,rutgers,pyramid,ames}!iuvax!bondc
------------------------------
End of NL-KR Digest
*******************